Energy-Efficient Strategies in Wireless Body Area Networks: A Comprehensive Survey

Abstract

Wireless body area networks (WBANs) are a pivotal solution for continuous health monitoring, but their energy constraints pose a significant challenge for long-term operation. This paper provides a comprehensive review of state-of-the-art energy-efficient mechanisms, critically evaluating solutions across various network layers. We focus on three key approaches: energy-aware MAC protocols that reduce idle listening and optimize duty cycling; energy-efficient routing protocols that enhance data transmission and network longevity; and emerging energy harvesting techniques that offer a path toward energy-autonomous WBANs. Furthermore, the paper provides a detailed analysis of the inherent trade-offs between energy efficiency and other critical performance metrics, such as latency, reliability, and security. It also explores the transformative potential of emerging technologies, such as AI and blockchain, for dynamic energy management and secure data handling. By synthesizing these findings, this work contributes to the development of sustainable WBAN solutions and outlines clear directions for future research.

1. Introduction

Recent advances in electronics and communication technologies are transforming many areas of life, especially healthcare [1,2,3]. The combination of small sensor nodes and wireless networks, supported by IoT (Internet of Things) developments, has led to the emergence of smart health systems [4]. These solutions enable continuous and real-time monitoring, thereby facilitating better diagnosis, treatment, and disease prevention [5]. Remote health monitoring, in particular, is useful for managing chronic diseases, supporting rehabilitation, and providing better care for the elderly [6].

Wireless body area networks (WBANs) represent a key technological advancement in smart healthcare, as they enable seamless communication between wearable and implantable medical sensors. These networks are composed of miniature sensor nodes that can either be implanted within the human body or positioned on its surface to supervise physiological signs, such as heart rate, body temperature, blood pressure, glucose levels, electrocardiogram (ECG), and oxygen saturation [7]. These sensor nodes perform continuous health monitoring and transmit the collected data to a central node, commonly referred to as the ‘sink’ or coordinator. The sink, typically equipped with higher processing power and energy capacity, aggregates the data and forwards it to medical servers wirelessly. Cloud-based platforms further enhance these systems by enabling remote monitoring and real-time intervention by healthcare professionals [8].

As illustrated in Figure 1, a standard WBAN architecture comprises three main communication levels: intra-WBAN, inter-WBAN, and extra-WBAN communication.

• Intra-WBAN communication involves data exchange between sensors within the network, with sensor-to-sink communication largely influenced by the network topology. Given the small-scale nature of WBANs, the single-hop star topology is the most widely used due to its simplicity and energy efficiency. However, for scenarios requiring greater reliability or scalability, alternative topologies, such as multi-hop peer-to-peer, mesh, or hybrid tree-mesh, are employed to ensure robust data transmission in diverse environments. Short-range communication standards, such as Bluetooth, Zigbee, Bluetooth Low Energy (BLE) and Wi-Fi Direct, are commonly utilized to facilitate intra-WBAN interactions.
• Inter-WBAN communication enables the exchange of information between multiple WBANs, which is essential in hospital environments where multiple patients may require simultaneous monitoring.
• Extra-WBAN communication involves transmitting aggregated data to healthcare providers through long-range communication protocols, such as cellular networks, Wi-Fi, or satellite communication, thereby extending monitoring capabilities beyond hospital settings.

Designing WBANs comes with many challenges that have been widely studied [9]. These include radiation absorption by the body, protecting sensitive medical data, ensuring reliable data transmission, and dealing with interference from nearby networks. However, one of the biggest challenges is energy efficiency, especially for implanted sensors that must work for a long time without changing the battery.

A WBAN node usually uses energy during three main phases: sensing, data processing, and transmission [10], as shown in Figure 2.

• Sensing phase: refers to the power utilized by the sensing module during sampling, conditioning, and analog-to-digital signal conversion. Depending on the application, a WBAN node can perform either interval sensing or continuous sensing. Interval sensing is used in applications where data is collected at specific intervals, minimizing energy consumption. Conversely, continuous sensing is required for applications that need constant monitoring, which leads to higher energy usage.
• Information processing phase: In this phase, the sensor node’s processing module, typically containing a controller, consumes energy while performing signal processing tasks on the data received from the sensing module. Additionally, the controller manages the operation of other modules within the node, contributing further to the overall energy consumption at the processing unit level.
• Data transmission phase: In this phase, energy consumption is driven by the tasks performed by the radio transceiver, which include transmitting and receiving RF signals, switching between active and passive states, and maintaining idle mode. Since wireless communication typically requires more energy than other operations, this phase often represents the most significant source of energy consumption in WBAN nodes.

Additionally, several other factors contribute to energy wastage in WBANs, as discussed further in [11]:

• Collisions: Collisions can be a significant source of energy wastage. In WBANs, intra-BAN or inter-BAN interferences (mutual and/or cross interferences) are frequently the cause of collisions, leading to packet loss, retransmissions, and ultimately, increased energy consumption.
• Traffic retransmissions: Traffic retransmissions occur in many scenarios, especially when nodes need to resend packets due to losses caused by channel access issues or interference, for example, resulting in increased network congestion and energy consumption.
• Idle listening: Idle listening emerges as a noteworthy phenomenon in WBANs; it happens when the durations of idle modes are significant, which may lead to unnecessary energy consumption.
• Overhearing: In network communication, overhearing occurs when a node actively listens to and intercepts packets that are not intended for it. In this scenario, the node captures and processes data frames originally meant for other nodes in the network, leading to additional energy consumption.
• Additional control packets: The incorporation of control packets occurs when a node adds an additional load to the traffic payload by introducing numerous control packets. These packets are typically used for managing and regulating the flow of information within the network, such as for synchronization purposes. However, an excessive amount of control packets can lead to inefficient energy use.

To address the energy efficiency challenge in WBANs, several approaches have been proposed in the literature; some studies focus on optimizing energy consumption via Medium Access Control (MAC) layer protocols [12], while others investigate energy-efficient routing protocols [13]. Additionally, energy harvesting techniques have been proposed as a complementary approach to enhance power autonomy and sustainability [13].

1.1. Contributions

This paper provides a comprehensive review of state-of-the-art energy-efficient solutions in WBANs, with a particular focus on optimizing energy consumption across various network layers. It examines key advancements aimed at mitigating energy wastage, extending network longevity, and enhancing the overall performance of WBAN-based healthcare applications. The key contributions of this paper are as follows:

• A critical evaluation of energy-efficient MAC protocols, emphasizing mechanisms such as low power listening, scheduled contention and time division multiple access.
• An in-depth analysis of energy-aware routing strategies, highlighting approaches that enhance data transmission efficiency while minimizing power consumption.
• A brief complementary review of energy-efficient data processing techniques—namely data aggregation, data fusion, and data compression—employed to reduce the volume of transmitted data and lower communication-related energy costs.
• A systematic investigation of emerging energy harvesting techniques, assessing their potential to reduce dependence on battery power and facilitate sustainable WBAN deployments.

1.2. Paper Structure

The remainder of this paper is organized as follows:

Section 2 provides an overview of related surveys and highlights the distinctive contributions of this work. Section 3 describes the methodology adopted for selecting, categorizing, and analyzing the references included in this review. Section 4 presents a detailed analysis of energy-efficient MAC protocols, focusing on techniques to minimize idle listening, reduce collisions, and optimize duty cycling. Section 5 investigates energy-efficient routing protocols in WBANs, covering thermal-aware, cluster-based, quality of service (QoS)-aware, cross-layered, and postural movement-based approaches. Section 6 introduces complementary energy-efficient data processing techniques, such as data aggregation, fusion, and compression. Section 7 explores energy harvesting innovations, emphasizing methods that harness both ambient and body-centric energy sources. Section 8 discusses advanced strategies, including hybrid energy harvesting and the integration of artificial intelligence for intelligent energy management. Section 9 provides a critical discussion of the reviewed techniques, outlines additional challenges, and suggests future research directions. Finally, Section 10 concludes the paper by summarizing the key findings and contributions.

2. Related Works

Several surveys have been published in recent years to address energy efficiency challenges in wireless body area networks (WBANs), particularly in the context of healthcare monitoring. These studies generally focus on routing strategies, MAC protocols, or hybrid approaches to prolong network lifetime and enhance system reliability. This section reviews key existing survey papers, highlighting their contributions and delineating the unique scope and novelty of our work.

Kurian et al. [14] conducted an early survey dedicated to energy-efficient routing protocols for WBANs, analyzing classical methods, such as thermal-aware, mobility-aware, and cluster-based approaches. While this work provided foundational insights into routing, it primarily emphasized the network layer without covering other essential protocol layers or cross-layer designs.

In [15], the authors presented another broad survey of energy-efficient routing protocols tailored for WBANs. They categorized approaches into types such as cluster-based, thermal-aware, cross-layered, QoS-driven, postural movement-based, cost-effective, and secure routing strategies. Their analysis emphasized minimizing node temperature and extending network lifetime for continuous healthcare monitoring. However, their study remains focused exclusively on the network layer, offering only limited discussion of MAC-level energy optimization techniques, energy-aware data processing strategies (such as aggregation and compression), and comprehensive cross-layer approaches. Furthermore, the survey does not explore energy harvesting methods, which are increasingly vital for enhancing the long-term sustainability of WBANs.

Javaid et al. [16] provided a comprehensive survey investigating the evolution of energy-efficient approaches in WBANs, emphasizing the transformative role of artificial intelligence (AI) and machine learning (ML). Their review covered both internal and external WBAN architectures, exploring energy consumption factors from a hardware-software perspective and highlighting state-of-the-art techniques across protocol layers. They also discussed the potential of emerging technologies, such as quantum computing, nano-technology, SWIPT, and biocompatible energy harvesting. While this work offers a broad technological outlook and explores futuristic directions, it notably lacks a structured taxonomy and fine-grained classification of specific MAC and routing protocols tailored to WBANs. Additionally, it provides limited technical depth regarding protocol-specific trade-offs and detailed layer-by-layer energy optimization strategies—areas that are central to the present review.

A systematic review of energy-efficient solutions for the Internet of Wearable Things (IoWT) was presented in [17], covering applications in healthcare, activity recognition, smart environments, and general-purpose wearable systems. This review provided a detailed taxonomy of techniques by application domain, a qualitative comparison of key performance indicators, and a statistical analysis of communication technologies and evaluation tools used from 2010 to 2020. It also highlighted trade-offs between energy efficiency and usability, comfort, and processing capabilities. However, while thorough in its investigation of energy-efficient strategies in wearables, its scope extends beyond WBAN-specific systems. Consequently, it addresses WBANs primarily as a subset of the broader IoWT paradigm, offering limited technical depth on WBAN-specific architectural, routing, and protocol-level challenges.

In contrast to the existing surveys, this paper provides a distinct and comprehensive analysis with a dedicated focus on the multifaceted aspects of energy efficiency within wireless body area networks (WBANs) specifically. We offer an updated and systematic review that integrates a detailed, multi-layered examination of energy-saving strategies, encompassing energy-efficient MAC protocols, routing protocols, data processing techniques, and a systematic investigation of emerging energy harvesting methods. Crucially, this work provides a structured taxonomy and fine-grained classification of these protocols, analyzing their technical trade-offs and layer-by-layer energy optimization strategies with significant depth.

3. Methodology

3.1. Research Questions

This study is guided by the following set of specific research questions, which directly align with the paper’s key contributions and guided the entire literature search, selection, analysis, and synthesis process:

• RQ1: What are the state-of-the-art energy-efficient MAC protocols in WBANs, specifically examining mechanisms such as low power listening, scheduled contention, and time division multiple access, and how do they contribute to optimizing energy consumption?
• RQ2: What are the key strategies and advancements in energy-aware routing protocols for WBANs, and how do they enhance data transmission efficiency while minimizing power consumption?
• RQ3: How do energy-efficient data processing techniques (including data aggregation, data fusion, and data compression) contribute to reducing the volume of transmitted data and lowering communication-related energy costs in WBANs?
• RQ4: What are the emerging energy harvesting techniques for WBANs, and what is their potential to reduce dependence on battery power and facilitate sustainable WBAN deployments?
• RQ5: What are the overarching advantages, limitations, and performance trade-offs of the various energy-efficient solutions identified, and what are the remaining research challenges and promising future directions for achieving long-term energy sustainability in WBANs?

3.2. Search Strategy

To conduct this review, a comprehensive literature review was carried out based on several reputable digital libraries, including IEEE Xplore, ACM Digital Library, ScienceDirect (Elsevier), SpringerLink, and MDPI. The search targeted English-language publications that specifically address energy-efficient strategies in WBANs. The following set of keywords and Boolean combinations was used to guide the query process:

• “Energy-efficient MAC protocols in WBAN”
• “Energy-efficient routing protocols for WBANs”
• “Energy harvesting techniques in WBANs”
• “Energy-efficient data processing techniques in WBANs”
• “Challenges and current trends in WBANs”.

The initial dataset was restricted to works published between 2002 and 2025 to ensure that both foundational research and the most recent developments were covered. Duplicate articles, non-peer-reviewed documents, and works not explicitly focused on WBANs or energy-saving strategies were excluded. Non-English documents were also omitted. In addition, previous surveys or review articles were examined to identify further relevant studies not retrieved in the initial search.

A summary of the inclusion and exclusion criteria is provided in

Table 1. Inclusion and exclusion criteria.

Inclusion Criteria	Exclusion Criteria
English-written papers published between 2002 and 2025	Non-English language papers
Peer-reviewed journals and conferences	Duplicate sources
Focus on WBANs and energy efficiency	General papers not targeting WBANs
Discuss MAC, routing, and energy harvesting	No discussion of energy saving techniques

.

3.3. Selection and Classification Criteria

After applying the search filters, a total of 131 articles were retained for detailed analysis. All selected articles addressed either MAC and network layer strategies, data processing techniques with the goal of optimizing energy consumption in WBANs, or energy harvesting techniques used in WBANs to enhance the energy efficiency.

The retained studies were categorized into the following main groups:

• Energy-efficient MAC protocols (based on low power listening, scheduled contention, and time division multiple access)
• Energy-efficient routing protocols (thermal-aware, cluster-based, cross-layered, QoS-based, and postural movement-based protocols)
• Energy-efficient data processing techniques
• Energy harvesting techniques in WBANs
• Current trends in WBANs

This classification allowed a layered analysis of energy-saving strategies at different protocol levels.

3.4. Dataset Statistics

Among the 178 retained articles, 146 were journal papers (82%) and 32 were conference publications (18%). All papers were published in English. The temporal distribution shows that early research on energy-efficient WBANs was limited, with only three papers published between 2002 and 2005, and 20 papers between 2006 and 2010. Interest grew significantly in the subsequent years, with 44 papers published between 2011 and 2016, and 50 papers between 2017 and 2020. The most recent period (2021–2025) accounts for 61 papers (34.3% of the total), confirming a sustained and strong research focus on addressing energy efficiency challenges in WBANs.

3.5. Quality Assessment of Sources

An evaluation of journal quality based on Scopus quartile ranking showed that approximately 76% of the selected journal articles were published in Q1 or Q2-ranked journals, including high-impact venues, such as IEEE Access, Sensors, Scientific Reports, Nature Communications, Advanced Materials, Energies, and various IEEE Transactions. This highlights the inclusion of high-impact, peer-reviewed research from leading publishers. The remainder came from Q3 or Q4 journals and reputable conferences with established scientific relevance in IoT, healthcare monitoring, wireless networks, and biomedical engineering, ensuring comprehensive coverage while maintaining quality standards.

4. Enhancing Energy Efficiency in WBANs via Optimized MAC Layer Protocols

The investigation and development of effective MAC protocols for WBAN systems are currently prominent topics in the healthcare sector [18]. These protocols are essential for optimizing energy consumption while ensuring robust health data communication. In wireless communications, the MAC layer, located directly above the physical layer, handles the allocation of channel resources among the different connected devices. In the context of WBANs, this layer manages the access times for sensor nodes sharing a common communication channel. However, to meet the strict WBAN QoS requirements [19], such as energy efficiency, throughput, reliability, and latency, the MAC layer must optimize resource allocation. Therefore, a well-designed MAC protocol should take into account various types of health data while maintaining QoS and reducing energy consumption caused by collisions and idle listening [20].

In general, WBAN MAC protocols are classified into three types: contention-free, contention-based, and hybrid protocols. Contention-free protocols use scheduling to allocate channels among biosensor devices, preventing collisions and reducing energy usage. Time division multiple access (TDMA) is one of the main protocols used in this category. On the other hand, contention-based protocols allow devices to compete for channel access without a fixed schedule, offering flexibility and adaptability but potentially leading to collisions, which can also cause high power consumption [21]. Examples include carrier sense multiple access (CSMA) and ALOHA. ALOHA permits instantaneous transmission and uses acknowledgements to detect collisions, facilitating retransmissions when necessary. In contrast, CSMA requires devices to listen to the channel before transmitting to avoid collisions. Hybrid protocols combine both contention-free and contention-based methods to optimize energy efficiency and channel utilization [20].

In the same context of WBAN MAC protocols, IEEE 802.15.4 and IEEE 802.15.6 serve as foundational communication standards for these systems. IEEE 802.15.4 is widely utilized in various IoT applications, such as industrial monitoring and smart grid systems [22], due to its effective MAC layer functionalities, such as beacon management and channel access. The IEEE 802.15.4 MAC sublayer interfaces with the MAC sublayer management entity (MLME) service access point (SAP) to provide two primary services: the MAC data service and the MAC management service. MAC protocol data units (MPDUs) can be sent and received over the PHY data service thanks to the MAC data service. Beacon control, channel access management, GTS handling, acknowledged frame transmission, association/disassociation processes, and frame validation are some of the major functions of the MAC sublayer. However, despite its channel access mechanisms, IEEE 802.15.4 is inherently susceptible to the hidden terminal problem. This occurs when two nodes transmit without detecting each other’s presence, yet both target the same receiver, resulting in data collisions. Furthermore, IEEE 802.15.4 falls short in supporting high data rate WBAN applications [23]. To fill this need, IEEE 802.15.6 was released specifically for WBANs [24], providing improvements in short-range communication and increased data transfer rates. This standard is tailored to meet various requirements of in, on, or near-body applications, thereby enhancing patient mobility and lowering healthcare expenses.

To make WBANs more energy efficient, it is important to improve how devices communicate at each layer by using specialized protocols. This section focuses on MAC protocols that are specifically designed to save energy when transmitting health data. In the literature, the main techniques used to reduce energy consumption in MAC protocols are low power listening (LPL), scheduled contention, and TDMA [25], as illustrated in Figure 3. LPL reduces idle listening by letting nodes wake up at regular intervals to check for activity, which works well for periodic data. Scheduled contention mixes scheduling and competition to prevent collisions and support scalability. TDMA gives each node a fixed time slot to send data, allowing long sleep periods, but it needs synchronization, which can consume extra energy. These methods help reduce wasted energy from unnecessary listening and data collisions. Table 2 gives a comparison of various energy-efficient MAC protocols used in WBANs.

Building on TDMA, battery-aware TDMA protocols [26] have been suggested to further enhance energy efficiency in healthcare-related WBANs. These protocols take into account the packet queuing characteristics, wireless channel unpredictability, and battery electrochemical properties in order to prolong the battery life of sensor nodes while guaranteeing fast and reliable data delivery, which is essential for patient monitoring. The coordinator sends periodic beacons in the same manner as the IEEE 802.15.4 standard [27]. Three time slots make up the time axis: (1) the active time slot, (2) the inactive period, and (3) the beacon slot. In order to accommodate various WBAN applications, the frame structure is flexible and adaptable. A periodic wakeup technique is implemented to prevent nodes from listening inactively. Every node has a specific time slot (Ts) during which, upon receiving a signal, the end node transmits data. Each node has a dedicated guaranteed time slot (GTS) assigned to it, which increases packet delivery time and reliability. During the idle phase, end nodes remain in sleep mode to prevent using further energy. Despite these advantages, the protocol exhibits limitations—it does not manage emergency data and relies on prolonged packet buffering, which can lead to increased drop rates and delays. Importantly, the authors provided a quantitative analysis of the trade-offs between energy efficiency and quality of service (QoS). For instance, Scheme-IV, which prioritizes battery recovery time, transmits up to 26 times more packets than IEEE 802.15.4 under specific conditions, substantially extending battery life. However, this gain results in higher average packet delays. In contrast, latency-optimized schemes, such as conventional TDMA, offer faster delivery at the expense of battery longevity. The study also highlights variations in packet drop rates and idle probabilities across different schemes and traffic conditions, offering concrete metrics to guide protocol selection based on specific WBAN application requirements.

An additional energy-efficient MAC algorithm for WBANs has been introduced in [28]. In the architecture of study, body-worn sensor nodes transmit vital sign data, such as body temperature, heart rate, and activity levels, to a central master node through a single-hop master–slave architecture. To conserve energy, sensor nodes stay in standby or sleep mode until their assigned time slot. Furthermore, in order to avoid collisions, the protocol uses a clear channel assessment algorithm that applies a listen-before-transmit (LBT) strategy. It also adds a wakeup backup time to handle possible time slot overlaps. While the proposed energy-efficient MAC protocol offers significant benefits for WBANs in pervasive healthcare, it still has some limitations. In fact, the dependence on a master-slave architecture may hinder scalability, i.e., the chosen architecture may not scale well with an increasing number of nodes, as the central master node could become a bottleneck. Additionally, fixed time slots and centrally managed communication can limit flexibility, potentially leading to latency and issues in handling varying data transmission needs.

Timmons et al. [29] introduced a TDMA-based MAC protocol for WBANs known as MedMAC, incorporating two key energy-saving mechanisms: the adaptive guard band algorithm (AGBA) and the drift adjustment factor (DAF). In order to rectify clock drift and allow nodes to enter sleep mode for extended periods of time, AGBA uses a guard band (GB) between time slots in conjunction with timestamping to ensure synchronization between the coordinator and nodes. On the other hand, DAF improves bandwidth utilization by adjusting GB based on real-world conditions, preventing slot overlaps. According to the obtained results, MedMAC demonstrates superior performance compared to IEEE 802.15.4 in low- and medium-data-rate applications. In fact, the assignment of GTS for each node reduces synchronization overhead and energy waste from collisions. However, MedMAC’s effectiveness diminishes in high data rate applications, typically in both in-body and on-body WBAN applications, which could limit its applicability in scenarios requiring the transmission of large amounts of data.

The S-MAC protocol aims to save energy by addressing common sources of waste, such as collisions and idle listening [30]. It achieves this by having nodes periodically go to sleep and wake up in a coordinated manner. While this approach is very effective for saving power, especially in networks with low traffic, it comes with a trade-off of increased data transmission delay. For example, S-MAC can use up to six times less energy than standard protocols, such as IEEE 802.11, in low-traffic situations. Although it introduces some overhead in high-traffic conditions, it still provides significant energy savings at the source node. The protocol’s ability to adjust its sleep duration based on traffic load makes it highly flexible, prioritizing energy efficiency and longer battery life over low latency.

Moreover, an additional MAC protocol for WBANs that is energy efficient and operates with a low duty cycle, used for remotely monitoring physiological parameters, such as EEG and ECG, has been introduced in [31]. By implementing a TDMA strategy, it minimizes power consumption through reduced idle listening and long sleep times for sensors, while ensuring collision-free data transfer. The protocol enhances resilience to communication errors with redundancy, supports real-time patient monitoring, and leverages the static nature of WBANs for effective TDMA timing control with minimal overhead. However, the use of TDMA and the need for synchronization might introduce latency, which could be problematic for applications requiring ultra-low latency communication. The proposed protocol also lacks forward error correction strategies to enhance data transmission reliability. To assess energy efficiency, the authors developed a detailed power consumption model using empirical measurements from a real RF transceiver. Results show that, for a sampling rate of 1250 bit/s and a communication rate of 34.56 kbit/s, the average transmission power is approximately 2.04 mW with a duty cycle of 4.51%. The protocol also performs well under higher data rates; for instance, when transmitting EEG data at 4096 bit/s using a 200 kbit/s link, the duty cycle drops to 1.35%, yielding an estimated power consumption of only 0.61 mW. Furthermore, in scenarios employing onboard seizure detection, the transmission load can be reduced by a factor of up to 70, decreasing the average power required for data transfer to around 25 µW.

The BodyMac protocol [32] is designed to save energy and reduce delays by using a structured approach for communication. It divides the communication time into different slots for sending and receiving data, allowing nodes to sleep when they are not active. BodyMac is highly flexible, with three different strategies for allocating bandwidth depending on the type of data being sent. This helps it use energy more efficiently by avoiding idle listening and being more responsive. Compared to other common protocols, BodyMac is much faster and more energy efficient. For instance, in a scenario with 54 nodes, the average end-to-end delay is reduced from 62 ms with standard CSMA/CA to less than 22 ms. Energy-wise, BodyMac significantly reduces node power consumption; its sleep mode cuts average power consumption from 0.006 W (GTS-based access) to 0.003 W. This makes it an excellent choice for applications that need both timely and long-lasting data transmission.

The heartbeat-driven MAC (H-MAC) [33] protocol is also a TDMA-based approach tailored for star topology WBANs, enhancing energy efficiency by using heartbeat rhythm information to synchronize nodes without the need for periodic signals. Each node is allocated a dedicated time slot to avoid collisions, supported by a synchronization recovery scheme. While this method reduces energy costs, it struggles with sporadic events due to its fixed TDMA slots, resulting in low bandwidth efficiency during low traffic. The heartbeat rhythm, which varies with patient condition, can sometimes fail to provide accurate synchronization. A potential solution is to dynamically allocate time slots based on node traffic and receive synchronization packets only when necessary.

IsMAC [34] is a multi-channel TDMA-based protocol designed to improve energy efficiency and network lifetime in medical WBANs. It introduces a rotating wireless coordinator node (WCN) mechanism, selecting nodes based on residual energy, previous WCN roles, and data priority. This rotation balances energy use across nodes. The protocol also uses adaptive transmission power control and dedicated time slots to reduce idle listening and collisions. A channel hopping strategy further prevents interference between coexisting WBANs. Simulations show that IsMAC achieves an EED below 4 ms with 40 WBANs (compared to 18 ms in ZigBee), consumes ~2 J per WCN (vs. 3.78–6.75 J in ZigBee), and maintains throughput close to 1 packet/slot, outperforming ZigBee’s < 0.5 packets/slot.

The quasi-sleep-preempt-supported (QS-PS) protocol introduced in [35] offers another approach to energy-efficient MAC for e-health WBANs. The protocol also operates on a TDMA basis, allowing nodes to send data packets within allocated time slots and enter Q-Sleep mode during idle slots to conserve energy. A notable feature of QS-PS is its ability to handle emergency packets by broadcasting awakening messages (AMs), which awaken the entire network and preempt the current slot for urgent transmissions, thereby reducing delay. However, the proposed protocol may introduce complexities in synchronization and coordination among nodes when waking up the entire network for emergency packets, which could potentially impact overall network performance.

Rezvani et al. [36] proposed a MAC protocol for WBANs that aims to improve the transmission of medical data in a star topology. The protocol uses structured superframes divided into periods for synchronization, contention, and scheduled data transmission. To handle emergencies, it includes dedicated slots for alarm messages and ensures collision-free emergency communication. It dynamically adjusts transmission cycles based on network conditions and prioritizes critical health data to ensure it is sent quickly and reliably. Simulation results showed that in emergency scenarios, the protocol reduced the power consumption of impermanent nodes to below 0.002 W, compared to over 0.012 W with IEEE 802.15.4. It also improved the number of successful transmissions, delivering over 400 packets in 10⁴ slots, outperforming both IEEE 802.15.4 and IEEE 802.15.6. Permanent nodes achieved data rates up to 35 kbps, whereas IEEE 802.15.6 only reached 20 kbps. Emergency transmission delays stayed below 20 ms, even with more nodes, while IEEE 802.15.4 delays exceeded 35 ms. However, the protocol may face synchronization issues in larger or denser networks, which could reduce its overall efficiency.

FT-MAC (few-transmit MAC) [18] is a specialized MAC protocol designed for WBANs, focusing on nodes operating in the few-transmit mode. It optimizes energy efficiency by structuring superframes with long inactive periods interspersed with short active insertion slots for data transmission. Quantitative evaluations using OMNeT++ simulations reveal that for low traffic scenarios (average data interval ≥ 1000 s), FT-MAC reduces node power consumption by over 50% compared to IEEE 802.15.4. In a WBAN of 20 nodes, the use of dual DATA sections further decreases collision rates and energy consumption. However, as traffic increases, FT-MAC exhibits a noticeable rise in average frame delay and energy usage, eventually surpassing IEEE 802.15.4 under dense traffic. This indicates a trade-off between energy efficiency and latency, making FT-MAC more suitable for health-monitoring applications with sparse event generation.

In line with these developments, the authors of [37] proposed a MAC protocol for WBANs that is both energy-efficient and delay-sensitive, focusing on the remote monitoring of physiological signals, such as EEG and ECG. They proposed to enhance the IEEE 802.15.4 standard by allowing sensors to interrupt normal operations for emergency data transmission, prioritizing immediate action to prevent casualties. Medical data traffic has been classified into normal and emergency categories, ensuring low-latency delivery of emergency data without waiting for regular transmissions. Simulation results confirm the protocol’s superiority over IEEE 802.15.4; when 5% of nodes generate emergency data, approximately 95% of packets are successfully delivered, versus a lower success rate in the standard. Even with 10% of emergency nodes, the protocol maintains a delivery rate of 83%. Furthermore, it achieves reduced average transmission time and lower energy consumption per bit, owing to dedicated time slots and emergency preemption. For instance, energy usage remains significantly below that of IEEE 802.15.4 under equivalent network loads. However, as the number of nodes increases, average transmission delay and energy consumption rise slightly due to concurrent access attempts. However, real-world validation in diverse operational environments, including dynamic medical settings, is essential to confirm the protocol’s reliability and performance under practical conditions.

The IEEE 802.15.6 CSMA/CA protocol has several shortcomings, including high control packet overhead, lack of effective prioritization for emergency traffic, and inefficient collision handling, which leads to increased delays and reduced throughput. Additionally, it does not offer a robust mechanism for adaptive sink node selection in dynamic environments, resulting in inefficient data transmission and higher energy consumption. For all these reasons, Shah et al. [38] proposed several enhancements to the IEEE 802.15.6 CSMS/CA by implementing a block acknowledgment policy to minimize control packet overhead and enhance energy efficiency. They prioritized emergency and normal traffic by assigning different contention window sizes and transmission power levels, allowing high-priority nodes to access the channel first, thereby reducing delays and collisions. Additionally, they proposed a new superframe structure to further support this prioritization. To optimize sink node selection and improve data transmission efficiency, they introduced a mobility link table (MLT) that dynamically selects the sink node based on the number of connections, addressing issues related to body mobility and posture variations. The proposed scheme reduced mean network delay and achieved up to 15% higher throughput and bandwidth efficiency compared to baseline IEEE 802.15.6 CSMA/CA. Energy efficiency was also significantly improved through reduced transmission power for normal traffic and minimized control overhead via block acknowledgment—sent once every five packets. In adaptive sink scenarios, power consumption dropped further thanks to intelligent sink assignment through MLT, where nodes with the highest link availability were selected, avoiding congestion and improving transmission accessibility. However, while emergency traffic is effectively prioritized, the strategy may lead to unequal bandwidth distribution, potentially delaying normal data. Thus, while the protocol enhances responsiveness and energy conservation, future research should explore fairness guarantees in multi-priority environments under dynamic WBAN conditions.

SDC-HYMAC (coordinated superframe duty cycle hybrid MAC) is another MAC protocol that has been proposed in [39] with the aim of optimizing energy efficiency and extending the operational lifespan of biomedical devices in IoT-enabled WBANs. It combines a coordinated superframe duty cycle mechanism with priority-based slot allocation to optimize time-slot utilization and minimize energy waste. Simulation results demonstrate that this approach leads to substantial gains in performance. For instance, with a network of nine devices, the protocol achieves a reduction in energy consumption down to 265 mJ, representing an improvement of up to 14% compared to standard configurations under identical conditions. Additionally, it maintains a low packet drop rate of approximately 15% and supports extended device lifetime—with improvements reaching up to 60% depending on the transmission probability. The use of a continuous-time Markov model for state transition estimation contributes to more accurate energy consumption predictions. Despite increased protocol coordination, the overall results highlight the efficiency and robustness of the proposed mechanism for reliable and energy-conscious WBAN communications.

The adaptive scheduling protocol for energy efficiency (ASPEE) suggested in [40] improves WBAN performance by dynamically adjusting node sleep schedules in response to variable traffic patterns, thereby reducing unnecessary energy expenditure. The protocol leverages a randomized threshold to mitigate frequent transitions between sleep and active states, which are a significant source of energy waste due to switching inefficiencies. Simulation results demonstrate that ASPEE achieves a reduction in the energy consumption ratio as the number of sensor nodes increases. Additionally, the protocol lowers the average packet delay and hop count while improving throughput and network lifetime. For instance, the proposed approach reduces the energy consumption by a significant margin, enhances throughput, and extends network lifetime, achieving performance improvements of up to 96.3% across multiple metrics, including latency, transmission rate, and packet delivery efficiency. These results confirm the protocol’s capacity to enhance energy efficiency without compromising data transmission quality, although synchronization challenges due to clock drift and topological variations still require robust mechanisms.

While energy-saving strategies, such as LPL, scheduled contention, and TDMA, significantly enhance the energy efficiency of WBAN MAC protocols, they are not without limitations. A primary challenge across many of these protocols is the inherent trade-off between energy conservation and other key QoS metrics. For instance, protocols designed for low power consumption, such as S-MAC [30], often introduce higher delay, which can be problematic for real-time applications. Similarly, TDMA-based protocols are highly efficient but can struggle with synchronization issues and may not be well-suited for high data rates or sporadic event-driven traffic. Other protocols may face scalability limitations due to their reliance on a single master node or complex coordination, as seen with FT-MAC [18] in high-traffic scenarios. These challenges show the need for MAC protocols that can adapt to changing WBAN conditions, save energy, and still deliver fast, high-rate, and reliable communication.

Table 2. Comparative analysis of the studied energy-efficient MAC protocols for WBANs.

MAC Protocol	Energy-Saving Approach	Advantages	Limitations
Battery-Aware TDMA [26]	TDMA	Prolongs battery life, fast and reliable data delivery, minimizes idle listening	No emergency data handling, packet buffering increases drop rates and delays.
[28]	Scheduled contention	Energy conservation, collision avoidance	Scalability issues, fixed time slots may cause latency.
MedMAC (Timmons et al.) [29]	TDMA	Superior performance at low/medium data rates, reduced sync overhead	Limited applicability at high data rates.
S-MAC [30]	LPL Scheduled contention	Reduced collisions, energy-efficient	Increased delay for energy efficiency.
[31]	TDMA	Minimizes idle listening, collision-free data transfer	Potential latency for ultra-low latency applications, no forward error correction.
BodyMAC [32]	TDMA	Energy-efficient sleep mode, precise synchronization	Overhead with synchronization, issues with frequent short bursts of data.
H-MAC [33]	TDMA	Reduced energy costs, avoids collisions	Low bandwidth efficiency, synchronization issues with varying heartbeat rhythms.
isMAC [34]	TDMA	Balances energy consumption, reduces retransmission and idle listening	Managing interference in dense environments.
QS-PS [35]	TDMA	Energy conservation, reduced delay for emergencies	Complexity in synchronization and coordination for emergency packets.
[36]	TDMA	Prioritizes critical data, reliable transmission	Synchronization challenges in star topology, potential errors.
FT-MAC [18]	TDMA	Energy efficiency, manages sporadic traffic	Handles sporadic traffic, reliable delivery.
[37]	TDMA	Low latency for emergencies, enhanced QoS.	Requires real-world validation for reliability.
Enhanced IEEE 802.15.6 CSMA/CA) [38]	Scheduled contention	Improved energy efficiency, adaptive sink node selection.	Potential unfair bandwidth allocation.
SDC-HYMAC [39]	Scheduled contention	Reduces energy consumption, minimizes contention overhead.	Challenges with dynamic adjustment and synchronization
ASPEE [40]	Scheduled contention	Optimizes energy consumption during traffic spikes.	Synchronization challenges due to clock drift and variable traffic loads.

Table 2. Comparative analysis of the studied energy-efficient MAC protocols for WBANs.

MAC Protocol	Energy-Saving Approach	Advantages	Limitations
Battery-Aware TDMA [26]	TDMA	Prolongs battery life, fast and reliable data delivery, minimizes idle listening	No emergency data handling, packet buffering increases drop rates and delays.
[28]	Scheduled contention	Energy conservation, collision avoidance	Scalability issues, fixed time slots may cause latency.
MedMAC (Timmons et al.) [29]	TDMA	Superior performance at low/medium data rates, reduced sync overhead	Limited applicability at high data rates.
S-MAC [30]	LPL Scheduled contention	Reduced collisions, energy-efficient	Increased delay for energy efficiency.
[31]	TDMA	Minimizes idle listening, collision-free data transfer	Potential latency for ultra-low latency applications, no forward error correction.
BodyMAC [32]	TDMA	Energy-efficient sleep mode, precise synchronization	Overhead with synchronization, issues with frequent short bursts of data.
H-MAC [33]	TDMA	Reduced energy costs, avoids collisions	Low bandwidth efficiency, synchronization issues with varying heartbeat rhythms.
isMAC [34]	TDMA	Balances energy consumption, reduces retransmission and idle listening	Managing interference in dense environments.
QS-PS [35]	TDMA	Energy conservation, reduced delay for emergencies	Complexity in synchronization and coordination for emergency packets.
[36]	TDMA	Prioritizes critical data, reliable transmission	Synchronization challenges in star topology, potential errors.
FT-MAC [18]	TDMA	Energy efficiency, manages sporadic traffic	Handles sporadic traffic, reliable delivery.
[37]	TDMA	Low latency for emergencies, enhanced QoS.	Requires real-world validation for reliability.
Enhanced IEEE 802.15.6 CSMA/CA) [38]	Scheduled contention	Improved energy efficiency, adaptive sink node selection.	Potential unfair bandwidth allocation.
SDC-HYMAC [39]	Scheduled contention	Reduces energy consumption, minimizes contention overhead.	Challenges with dynamic adjustment and synchronization
ASPEE [40]	Scheduled contention	Optimizes energy consumption during traffic spikes.	Synchronization challenges due to clock drift and variable traffic loads.

5. Enhancing Energy Efficiency via Optimized Routing Protocols

Routing protocols enable the selection of the best path for packets to reach a destination node in a network. In WBANs, this path selection must be highly efficient, as the sensed data can be critical. Therefore, routing protocols in WBANs must also take into account the different challenges mentioned above [41], with limited energy resources being the most critical among them. Mobility poses another problem, as the patient is intended to move, and changing network topologies requires dynamic path selection for data transmission [42]. High QoS is also required in many WBAN applications, with factors such as timing, latency, bandwidth, and robustness being critical [43]. Security is also a major concern, with the necessity for ensuring data authenticity, integrity, confidentiality, and availability [44]. Real-time synchronization is crucial to avoid delays that could affect patient health, particularly in chronic disease monitoring. Additionally, data management is important due to the significant amount of information collected, necessitating efficient methods to handle and transmit this data. Therefore, developing robust and adaptive routing protocols is vital for the effective functioning of WBANs.

WBAN routing protocols can be classified into five categories based on the specific challenges discussed above [41,45]: thermal-aware, cluster-aware, cross-layered, QoS-aware and postural movement-based routing protocols. Thermal-aware routing protocols [46] use node temperature as a criterion for selecting the best route. They aim to minimize node overheating by avoiding nodes that are considered hotspots, which prevents body tissues from being damaged. However, managing heat generation does not always mean reducing energy consumption, as avoiding hotspots can sometimes lead to longer routing paths that impact overall energy efficiency. Therefore, to effectively reduce heat generation while achieving energy efficiency, a routing protocol should integrate both thermal awareness and energy-efficient strategies [47]. Cluster-aware WBAN protocols explicitly focus on enhancing energy efficiency by organizing nodes into clusters, with each cluster managed by a leader (cluster head) that communicates with the sink, based on the energy-efficient low energy adaptive clustering hierarchy (LEACH) protocol [48]. Protocols that consider QoS focus on both energy efficiency and key metrics, such as reliability and delay. Cross-layered protocols aim to improve overall network performance by sharing information between multiple communication layers (mainly between MAC and network layers), with energy efficiency often being a secondary advantage [45]. Postural movement-based routing protocols adapt to topology changes from body movements by updating a cost function to select the best path, focusing on reducing packet delivery delay and lowering energy consumption.

Given the range of challenges addressed by WBAN routing protocols, this paper will specifically focus on energy-efficient routing protocols within each of the identified categories, as presented in Figure 4. By narrowing the scope to those protocols that prioritize energy conservation, this paper aims to provide a more in-depth analysis of the methods used at the network layer level to enhance energy efficiency, whether through clustering, thermal management, or dynamic adaptation to body movements.

5.1. Thermal-Aware Energy-Efficient Routing Protocols

Thermal-aware routing protocols are important for the effective operation of WBANs as they help manage the temperature of sensing devices. Classic key protocols include TARA (thermal-aware routing algorithm for implanted sensor networks) [49], which, to the best of our knowledge, was the first thermal-aware routing protocol designed for WBANs. It routes data away from overheated nodes by defining hotspots and monitoring the temperature increase potential (TIP) based on the specific absorption rate (SAR) [50]. This approach balances the temperature among nodes but faces challenges in network lifetime and packet reliability. Enhancements such as LTR [51] (least temperature routing) and ALTR [51] (adaptive least temperature routing) build on TARA by selecting the next hop based on the lowest temperature and adapting hop limits to avoid packet loss. The least total route temperature (LTRT) [52] combines LTR and shortest path routing, using Dijkstra’s algorithm to select the coolest path and adjusting sensor node temperatures based on activity, although it requires comprehensive temperature knowledge and monitoring energy consumption. Hotspot preventing routing (HPR) [53] includes setup and routing phases to prevent delays and hotspots by using average node temperatures for routing decisions.

Most classic temperature-aware protocols aim only to minimize temperature generation [54], neglecting energy conservation. Therefore, the following discussion focuses exclusively on thermal-aware energy-efficient routing protocols suggested in the literature for WBANs, and summarized in Table 3. These protocols aim to optimize both temperature management and energy efficiency, ensuring effective and sustainable network performance. Furthermore, a presentation of the advantages and limitations of each of these protocols is presented in Table 3.

Bag et al. [55] proposed the RAIN protocol (routing algorithm for a network of homogeneous and ID-less biomedical sensor nodes) to reduce both power consumption and temperature rise in WBAN sensor nodes. RAIN works in three phases: setup, routing, and status updates. In the setup phase, hello packets are used to assign temporary IDs to nodes (the sink is assigned an ID of zero). During routing, each packet carries a unique combination of node ID (N), creation time (T), and a random number (R), along with a hop count. This helps avoid routing loops. Nodes track previously seen packet IDs to prevent retransmissions and also exchange temperature data with neighbors. Depending on the nearby temperature, a node decides whether to forward the packet or send it directly to a neighbor. In the status update phase, the sink notifies its neighbors when it receives a packet, helping them conserve energy. However, sending frequent status updates from the sink may increase communication overhead.

The M-ATTEMPT (mobility-supporting adaptive threshold-based thermal-aware energy-efficient multi-hop protocol) enhances WBAN performance by addressing key challenges, such as heat generation, mobility-related disconnections, and energy consumption [56]. It uses direct communication for critical data to reduce delay, and multi-hop for regular data to conserve energy. The protocol integrates thermal-aware routing to avoid hotspots and employs a linear programming model to optimize energy use and data extraction. M-ATTEMPT operates through four phases—initialization, routing, scheduling, and data transmission—using TDMA to assign time slots for nodes. Simulation results show that it significantly prolongs network lifetime, with the first node death occurring around the 2700th round in a 5000-round simulation. It also achieves an approximately 29% improvement in packet delivery and a 29.5% improvement in overall network longevity, thanks to its adaptive routing and hybrid communication model [57]. However, it may lead to uneven energy consumption. To overcome this, RE-ATTEMPT was proposed in [58], enhancing energy efficiency and reliability by introducing backup routes, improving energy balancing, and equipping relay nodes with higher energy. It demonstrated a 7.46% reduction in average energy consumption, extended the stability period (first node death at round 1479), and achieved reliable data delivery, with 6930 packets received out of 9889 sent.

TEAR (thermal and energy aware routing) [47] is a routing protocol designed to address the critical challenges of energy preservation, thermal dissipation, and reliable communication in WBANs. It selects routing paths based on a weighted combination of three metrics: energy consumption, heat dissipation, and link quality. The protocol dynamically builds a gradient cost field where each node updates its cost based on neighbors’ information, enabling the selection of routes that minimize power usage and thermal impact. Transmission power is adjusted in real time according to received signal strength to ensure communication reliability while reducing energy usage and heat generation. Simulation results demonstrated that TEAR significantly reduces thermal stress on tissues by maintaining only 70% of nodes in a heated state, compared to full network heating under static transmission strategies. Moreover, transmission power was maintained at its minimum level (–20 dBm) for most of the protocol’s lifetime, effectively minimizing the Heating Ratio while preserving packet delivery quality. However, the protocol’s reliance on periodic updates from neighboring nodes for cost calculations could also lead to delays, particularly in dense networks, impacting real-time data transmission.

The M2E2 multi-hop routing protocol is also a thermal-aware and energy-efficient protocol that has been introduced in [59], it operates in four phases: initialization (nodes determine distance from the sink), routing (data sent to the medical server via direct link if the home signal is present, otherwise multi-hop), scheduling (assigns communication time slots), and data transmission (data sent within allocated slots). The protocol also aims to minimize energy consumption and extend network life by selecting high data rate parent nodes close to the sink. It supports two modes: single-hop communication when at home, and multi-hop when the home signal is unavailable.

A novel thermal-aware clustering and routing protocol for multi-WBANs was proposed in [60], using fuzzy logic and a hybrid optimization algorithm to manage temperature and energy challenges. The protocol includes two fuzzy logic controllers (FLCs): the first organizes sensors into clusters based on factors such as cluster head temperature, number of neighbors, remaining energy, and path loss; the second helps CHs send data to coordinators by considering sink load, distance, and packet delivery ratio. To optimize the FLCs’ rules, the authors introduced HAOA—a hybrid algorithm combining Aquila and arithmetic optimization—to improve performance and avoid local optima. The HAOA-optimized protocol consistently reduced average sensor temperature and hotspot occurrence in various settings (fixed or mobile patients with 20–30 sensors). For example, in a 20-sensor fixed scenario, it achieved 38.77 °C average temperature, 0.20 hotspot ratio, and extended network lifetime with first, half, and last node death times of 802, 1543, and 2531 s, respectively. However, its performance may be affected by interference when nearby WBANs share the same frequency, highlighting the need for further evaluation under mutual interference conditions [61].

TAEO (thermal-aware and energy-optimized routing) [62] is another temperature-aware energy-efficient routing protocol that focuses on controlling thermal effects and optimizing energy consumption in WBANs. The primary goal of TAEO is to manage node temperature and energy use to prevent tissue damage and extend network lifetime. The protocol achieves this by detecting and avoiding hot-spot nodes using SAR values, and by selecting data forwarders (DF) based on temperature, residual energy, and proximity to the sink. During the initialization phase, nodes broadcast their location, energy level, heat value, and identifier. In the routing phase, nodes exceeding the temperature threshold are temporarily suspended from transmitting data to allow their temperature to return to normal. Utilizing a multi-hop scheme, TAEO adjusts transmission power based on the distance between nodes to further optimize energy use. Analytical simulations comparing TAEO with ATTEMPT and SIMPLE protocols demonstrated significant enhancements. TAEO showed a remarkable improvement in the stability period, with the first node dying in the 6000th round, representing a 50% improvement compared to ATTEMPT and 28% compared to SIMPLE [63]. This optimized energy consumption was further evidenced by residual energy analysis, where TAEO nodes consumed significantly less energy, with energy utilization in 6000 rounds comparable to ATTEMPT’s consumption in 2000 rounds and SIMPLE’s in 4000 rounds. In terms of thermal management, TAEO achieved a 70% saving in heated nodes, meaning only 30% of total nodes became heated after 16,000 rounds, a stark contrast to existing protocols, where 100% of nodes typically become heated.

Table 3. A comparative overview of thermal-aware energy-efficient routing protocols for WBANs. ✓ indicates that the feature is present; — indicates that the feature is not applicable.

Protocol	Mobility	PLR Reduction	Latency Reduction	Energy Efficiency	Thermal Management	Fault Tolerance	Main Advantages	Main Limitations
RAIN [55]	✓	✓	—	✓	✓	—	Avoids loops & retransmissions.	Frequent sink status updates may increase communication overhead.
M-ATTEMPT [56]	✓	✓	—	✓	✓	—	Addresses heat, mobility-induced disconnections, & energy consumption.	May lead to non-uniform load distribution under certain conditions.
RE-ATTEMPT [58]	✓	✓	—	✓	✓	✓	Improved reliability & energy balancing.	Potential for increased computational/communication overhead.
TEAR [47]	—	✓	✓	✓	✓	—	Energy, thermal, & reliability-aware routing; Dynamic power adjustment.	Reliance on periodic updates from neighbors can cause delays in dense networks.
M2E2 [59]	✓	—	✓	✓	—	—	Supports Adaptive single/multi-hop modes.	Potential for non-uniform energy consumption/hotspots if “high data rate parents” are consistently preferred.
[60]	✓	✓	—	✓	✓	—	low avg. temp & hotspot ratio.	Performance can be impacted by interference.
TAEO [62]	—	✓	✓	✓	✓	—	Hotspot detection/avoidance (using SAR); Improved stability period.	Reactive hotspot avoidance might cause temporary data delays.
THE [64]	✓	✓	✓	✓	✓	—	Utility function for parent selection (multi-metric).	Possible complexity in managing data priority & mode switching.
WETRP [54]	—	✓	✓	✓	✓	✓	Hybrid routing metric (temp, energy, latency); Minimizes temp rise, optimizes energy.	Reliance on accurate/timely node status info may not be feasible in dynamic environments.

Table 3. A comparative overview of thermal-aware energy-efficient routing protocols for WBANs. ✓ indicates that the feature is present; — indicates that the feature is not applicable.

Protocol	Mobility	PLR Reduction	Latency Reduction	Energy Efficiency	Thermal Management	Fault Tolerance	Main Advantages	Main Limitations
RAIN [55]	✓	✓	—	✓	✓	—	Avoids loops & retransmissions.	Frequent sink status updates may increase communication overhead.
M-ATTEMPT [56]	✓	✓	—	✓	✓	—	Addresses heat, mobility-induced disconnections, & energy consumption.	May lead to non-uniform load distribution under certain conditions.
RE-ATTEMPT [58]	✓	✓	—	✓	✓	✓	Improved reliability & energy balancing.	Potential for increased computational/communication overhead.
TEAR [47]	—	✓	✓	✓	✓	—	Energy, thermal, & reliability-aware routing; Dynamic power adjustment.	Reliance on periodic updates from neighbors can cause delays in dense networks.
M2E2 [59]	✓	—	✓	✓	—	—	Supports Adaptive single/multi-hop modes.	Potential for non-uniform energy consumption/hotspots if “high data rate parents” are consistently preferred.
[60]	✓	✓	—	✓	✓	—	low avg. temp & hotspot ratio.	Performance can be impacted by interference.
TAEO [62]	—	✓	✓	✓	✓	—	Hotspot detection/avoidance (using SAR); Improved stability period.	Reactive hotspot avoidance might cause temporary data delays.
THE [64]	✓	✓	✓	✓	✓	—	Utility function for parent selection (multi-metric).	Possible complexity in managing data priority & mode switching.
WETRP [54]	—	✓	✓	✓	✓	✓	Hybrid routing metric (temp, energy, latency); Minimizes temp rise, optimizes energy.	Reliance on accurate/timely node status info may not be feasible in dynamic environments.

E. Selem et al. [64] proposed the temperature heterogeneity energy (THE)-aware routing protocol for WBANs in IoT health applications, addressing the network layer gap in IEEE 802.15.6. These deficiencies mainly refer to the lack of standardized protocols and mechanisms to manage data transmission, energy efficiency, and temperature control in on-body sensors. THE is at the same time a one-hop, multi-hop and clustering-based protocol that aims to control on-body node temperatures to prevent skin harm while optimizing network performance. It employs a utility function to select parent nodes based on remaining energy, data rate, proximity, and temperature, adapting between one-hop and multi-hop transmission based on data priority. Monte Carlo simulations demonstrated that THE protocol increases network longevity by 11% over SIMPLE [63] (Stable increased-throughput multi-hop protocol for link efficiency) and 6% over iM-SIMPLE [65] (iMproved stable increased-throughput multi-hop link efficient routing protocol). These are both classic routing protocols in WBANs. SIMPLE focuses on power efficiency and high throughput by using a multi-hop topology and a cost function to choose nodes with high residual energy and minimal distance to the sink, maximizing network stability and packet delivery. iM-SIMPLE, as an improved version of SIMPLE, enhances these features by incorporating mobility considerations and optimizing energy consumption and throughput for physiological data transmission from sensor nodes on the human body.

Bhangwar et al. [54] proposed the WETRP protocol (weighted, QoS-based, energy and temperature-aware routing protocol) to extend the lifetime of WBANs through efficient multi-hop communication using relay nodes. WETRP selects routes based on a hybrid metric that equally considers node temperature, residual energy, and link latency. This approach ensures balanced traffic distribution, reduces temperature rise, and improves the delivery of critical data. The protocol includes a route discovery phase to find optimal paths and a maintenance phase to avoid hotspots and low-energy nodes. Simulations showed that WETRP achieves lower average delay across different data rates and better thermal management, maintaining a lower temperature rise even at 100 Kbps compared to TARA and HPR. It also demonstrated a longer network lifetime due to efficient use of network resources. However, its performance depends on accurate, real-time knowledge of node status, which can be challenging in dynamic environments.

Figure 5 illustrates the operational steps of four representative thermal-aware routing protocols for WBANs: RAIN, M-ATTEMPT, TEAR, and TAEO.

Thermal-aware energy-efficient routing protocols help control node temperature and prolong network lifetime, but they face certain drawbacks. One major issue is the trade-off between temperature control and energy savings, as avoiding hotspots can sometimes necessitate longer, less energy-efficient routing paths. For specific protocols, limitations include the communication overhead associated with frequent sink updates in RAIN [55], the potential for non-uniform load distribution in M-ATTEMPT [56], and the reliance on periodic neighbor updates in protocols such as TEAR [47], which can cause delays in dense networks. Furthermore, a heavy dependence on accurate and timely node status information, as seen in WETRP [54], may not be feasible in highly dynamic environments, and the need for comprehensive temperature monitoring can be computationally expensive. These challenges highlight the ongoing need for research into protocols that can dynamically balance the conflicting requirements of heat dissipation, energy efficiency, and low latency for reliable WBAN operation.

5.2. Cluster-Based Energy-Efficient Routing Protocols

Cluster-based routing is a significant approach in WSNs (wireless sensor networks) generally and in WBANs in particular that aims to enhance network lifespan by lowering energy consumption. In this strategy, the network is divided into clusters, each electing a CH responsible for communicating with the base station, thereby conserving the energy of other sensor nodes. Building on this concept, two main classical protocols have been proposed in the literature, which are hybrid indirect transmission [66] and AnyBody [67].

Hybrid indirect transmission (HIT) [66] is a protocol that uses a cluster-based hybrid architecture with multi-hop indirect transmissions to enhance energy efficiency and minimize network delay. It operates using a structured, round-based approach. Each round includes a brief cluster setup phase, where clusters and multi-hop routes are established, followed by an extended data transmission phase. Data is relayed to cluster-heads via indirect paths and fused along the way, minimizing the overall data size sent to the base station. HIT supports parallel transmissions both within and between clusters through an independently calculated TDMA schedule for each sensor. This method is resource-efficient and eliminates the need for CDMA-capable nodes or a remote base station, making it a more streamlined and effective solution. Simulations demonstrated HIT’s significant improvements in energy efficiency and network longevity compared to LEACH, PEGASIS, and direct transmission. HIT lasted approximately 1.05 times as long as PEGASIS and 1.44 times as long as LEACH, directly translating to an extended network lifetime. However, while HIT supports parallel transmissions, scalability could be challenging in large networks, as managing multiple TDMA schedules might lead to synchronization issues and latency. For instance, maintaining precise clock synchronization across numerous distributed nodes in a dynamic WBAN is inherently difficult, as even small clock drifts can lead to slot misalignment, collisions, and increased retransmissions, thereby accumulating latency and consuming more energy [67]. This challenge is exacerbated in large or mobile networks where topology changes frequently, requiring complex scheduling updates that can introduce additional overhead and delays.

AnyBody [67] is a self-organization routing protocol designed for WBANs, particularly in hospital environments. The protocol focuses on creating a functional network structure to facilitate multi-hop data transmission from medical and environmental sensors to a central sink node, serving as a gateway to the hospital’s wired network. It operates in five steps: First, nodes discover their neighbors by exchanging hello messages. Second, nodes calculate their density and elect CHs based on this information. Third, nodes join clusters led by the highest-density nodes. Fourth, CHs identify and connect gateway nodes to form a virtual backbone that links clusters. Finally, routing paths are established from each cluster to the central sink node using a gradient-based method. Nevertheless, the protocol’s gradient-based routing paths might not adapt to sudden modifications in network topology, such as node mobility or failure, which may cause increased packet loss.

In [68], a cluster-based routing protocol for WBANs was introduced to improve energy efficiency using a hierarchical clustering model. The protocol begins by setting up the network and classifying nodes based on their energy levels. It operates in rounds, where cluster heads (CHs) are dynamically selected according to residual energy to balance the load and extend network lifetime. The system includes phases for network initialization and data transmission, with energy updates at each round. To further optimize performance, the protocol uses multi-level clustering and mixed integer linear programming (MILP) to calculate the optimal probability for CH selection. Simulations in MATLAB (https://www.mathworks.com/products/matlab.html) with 100 sensor nodes over a 10,000 m² area showed that the protocol significantly outperformed LEACH in terms of energy use and network lifespan—maintaining network activity for over 6000 rounds compared to less than 1500 rounds with LEACH. However, the frequent re-election of CHs based on energy levels may introduce communication delays and reduce stability in dynamic environments.

On the other hand, Srinivas et al. [69] proposed an energy-efficient routing algorithm for WBANs focused on real-time monitoring of patients’ vital signs in hospital settings. The algorithm combines ant colony optimization (ACO) with clustering to find optimal paths for data transmission, adapting to changing traffic patterns and ensuring timely delivery. It improves energy efficiency, load balancing, and network lifetime by rotating cluster heads and using a breadth-first search to avoid routing loops. To reduce overhead, it limits node participation while maintaining reliable transmission of critical data. Simulations in OMNeT++, under different traffic and power settings (e.g., −25 dBm), showed that the proposed CH-ACO algorithm achieved higher successful transmission rates, lower energy use, and better load distribution. However, because the algorithm is iterative, it may introduce delays during heavy traffic or rapid changes in a patient’s condition.

The secure optimal path routing (SOPR) is another cluster-based routing protocol introduced in [70], designed for energy efficiency and security in WBANs. It uses a multi-criteria decision-making approach for path selection, considering factors such as energy consumption, path reliability, and security levels. By integrating the balanced energy-efficient and reliable (BEER) algorithm, SOPR effectively balances energy usage across sensor nodes, preventing premature battery depletion and prolonging network lifespan. Additionally, SOPR includes security mechanisms that protect data integrity and confidentiality, ensuring that sensitive medical information remains secure during transmission. This dual focus makes SOPR particularly well suited for critical WBAN applications. SOPR-BEER demonstrated a higher packet-delivery ratio and extended network lifetime, notably retaining 3% to 6% more active nodes on average compared to ATTEMPT and M-ATTEMPT. For security, it detected black-hole nodes faster and with less communication overhead due to local discovery. Its overall energy consumption was also considerably lower, achieved by optimizing intra- and inter-cluster communication. However, SOPR’s intricate decision-making process might limit its effectiveness in large or dynamic high-density WBANs due to potential scalability challenges.

Another cluster-based routing approach for WBANs has been proposed in [71] to enhance energy efficiency in real-time health monitoring systems. In this protocol, sensor nodes are divided into two groups, each with a dedicated sink, and elect a CH within each group based on a cost function. The CH is responsible for aggregating and transmitting data to the sink. To minimize energy dissipation, nodes either send data directly to the CH or through a forwarder node, selected based on distance and residual energy. Additionally, the protocol incorporates solar energy harvesting to extend battery life, enabling long-term monitoring without frequent battery replacements.

Zaman et al. [72] proposed an energy-efficient distance and link-aware body area (EEDLABA) protocol. It operates by dividing the network into clusters, with each cluster managed by a CH. It uses two path loss (PL) models to adaptively select energy-efficient routing paths. These models help determine the most efficient routes based on nodes’ residual energy and proximity to the CHs. This dual PL approach minimizes energy consumption and signal degradation, enhancing network stability and extending the overall lifespan of WBANs. EEDLABA achieved the lowest average path loss, minimizing signal power reduction during transmission, and it also consumed the least energy, leading to significantly higher average residual energy compared to other approaches. This directly translated to an extended network lifetime, with the first node dying at 4205 s, a notable improvement over existing protocols (which saw first node death around 2000–2100 s). However, because node movement may often require re-evaluating routing paths, clustering and PL-based routing may bring latency and instability in dense or high-mobility systems.

In the same context, Arafat et al. [73] proposed a distributed energy-efficient two-hop-based clustering and routing protocol (DECR) integrated with the Wearable Internet of Things (WIoT). The protocol addresses the challenges of high interference, mobility, and limited energy by employing a distributed approach to clustering and routing. It utilizes two-hop neighbor information to enhance cluster formation, with key metrics including the two-hop connectivity ratio (TCR), energy factor (EF), and node stability factor (NSF) to optimize CH selection. The protocol employs a modified grey-wolf optimization (MGWO) algorithm for efficient CH selection and routing, ensuring energy-efficient data transmission from CHs to the sink. DECR’s design includes an analytical model for determining the optimal number of clusters, aiming to minimize overall transmission distances and energy consumption. The simulation results demonstrate that DECR significantly improves performance metrics, such as the packet delivery ratio (PDR), the average end-to-end delay, control overhead, cluster building time, and overall network energy consumption, compared to existing protocols.

Within a similar framework, another cluster-based routing protocol has been proposed in [48]. It selects CHs and innovative commander nodes (ICNs) based on parameters, such as innovative bandwidth, distance, and residual energy. The sensor nodes deployed on the human body calculate these values, and those with the highest rank are chosen as CHs and ICNs. The CHs then aggregate data from their respective clusters and forward it to the selected ICNs, ensuring efficient routing and communication within the network. However, the frequent recalculations and reassignments needed for CH and ICN roles could cause delays and overhead, affecting real-time data transmission in high-mobility contexts, as with active patients.

In addition to the aforementioned clustering-based protocols, recent advances in multi-view clustering can offer new perspectives for enhancing WBAN routing performance. Recent work by Sinaga and Yang [74] on Globally Collaborative Multi-View k-Means Clustering (GCMVKM) offers useful insights for improving clustering-based routing in WBANs. This method combines information from multiple data “views” to produce more accurate and robust clusters. In a WBAN context, such an approach could use different sensor parameters—such as energy level, temperature, and link quality—as separate views. This would allow clusters to be formed more adaptively, helping to balance energy use, manage heat, and maintain stable network performance in changing conditions.

Table 4. Comparative analysis of cluster-based energy-efficient routing protocols in WBANs. ✓ indicates that the feature is present; — indicates that the feature is not applicable.

Protocol	Mobility	PLR Reduction	Latency Reduction	Energy Efficiency	Security	Fault Tolerance
HIT [66]	✓	✓	✓	✓	—	—
AnyBody [67]	✓	✓	—	✓	—	—
68]	—	✓	✓	✓	—	✓
[69]	✓	✓	—	✓	—	—
SOPR [70]	—	✓	✓	✓	✓	✓
[71]	✓	✓	✓	✓	—	—
EEDLABA [72]	—	✓	—	✓	—	—
DECR [73]	✓	✓	✓	✓	—	—
[48]	✓	✓	—	✓	—	—

presents a comparative analysis of cluster-based energy-efficient routing protocols based on key criteria, including mobility support, packet loss rate (PLR) and latency reduction, energy efficiency, security, and fault tolerance. This comparison highlights the distinguishing features and performance aspects of the discussed protocols.

While cluster-based routing protocols are highly effective for enhancing energy efficiency in WBANs, they come with a distinct set of limitations. A primary challenge is their adaptability to dynamic environments, as the frequent re-evaluation of routing paths and re-election of cluster heads (in high-mobility scenarios can lead to increased communication delays and reduced network stability. For instance, protocols such as HIT [66] may face scalability issues due to the complexity of managing multiple TDMA schedules, which can lead to synchronization problems and latency. Similarly, the gradient-based routing in AnyBody [67] might not adapt quickly enough to sudden topology changes, potentially causing packet loss. The iterative nature of protocols such as CH-ACO [69] can also introduce delays during periods of heavy traffic or rapid changes in patient condition. Furthermore, the intricate decision-making process in protocols such as SOPR [70] may limit its effectiveness in large or dynamic high-density WBANs due to potential scalability challenges. These limitations highlight the need for future research to develop cluster-based solutions that can dynamically balance energy efficiency with the strict latency and stability requirements of mobile WBAN applications.

5.3. Cross-Layered Energy Efficient Routing Protocols

Cross-layer design improves protocol efficiency by merging layers in the protocol stack, unlike the traditional OSI model, where each layer operates independently. This approach addresses the limitations of layered designs in complex and performance-sensitive environments, such as WBANs, which demand high QoS, energy efficiency, and adaptability to variable channel conditions. A cross-layer design can enhance performance through new interfaces that link non-adjacent layers, merging adjacent layers into a unified super-layer, vertical calibration of parameters across layers, and even replacing the entire stack with an integrated, optimization-based architecture.

While cross-layer designs offer significant performance gains, especially in resource-constrained environments, they must be balanced against potential issues with scalability and compatibility, requiring a careful evaluation of the benefits and trade-offs [75]. Furthermore, there are two main categories of cross-layer protocols: loosely coupled and tightly coupled cross-layer designs. Loosely coupled designs optimize higher layers using parameters from lower layers while maintaining the layered structure, whereas tightly coupled designs integrate layers into a unified solution, potentially offering greater performance gains but at the cost of protocol transparency and maintenance complexity [75].

The present section will focus on the key cross-layered routing protocols proposed in the literature to enhance energy efficiency in WBANs. These protocols are summarized and compared in Table 5.

TICOSS (TImezone COordinated Sleep Scheduling) is a cross-layer protocol [76] designed to improve energy efficiency in healthcare-related WBANs by managing the sleep and wake cycles of sensor nodes. It introduces the concept of “time zones,” grouping nodes based on their distance (in hops) from the network coordinator and synchronizing their sleep schedules accordingly. This helps overcome some key limitations of the IEEE 802.15.4 standard [77], such as its support for only single-hop communication, lack of hidden terminal mitigation, and uncoordinated sleep scheduling. TICOSS coordinates both the MAC and routing layers to enable efficient multi-hop communication. It uses a V-table at the MAC layer to assign time slots for sending and receiving data, ensuring that nodes in adjacent time zones wake up at the same time for data forwarding. This alignment reduces packet collisions, improves connectivity, and saves energy. The protocol also helps mitigate the hidden terminal problem by structuring transmissions in a TDMA-like manner. Simulation results showed that adding TICOSS to IEEE 802.15.4 networks significantly extends network lifetime—nearly doubling it under high traffic conditions—while maintaining better performance across different traffic levels. However, in larger networks, coordinating time zones becomes more complex, and synchronization delays may affect communication efficiency.

In the same context, Braem et al. introduced the WASP (wireless autonomous spanning tree protocol) [78], a cross-layer protocol that combines MAC and routing functions to improve coordination and energy efficiency in multi-hop WBANs. WASP builds a dynamic spanning tree rooted at the base station, where each node selects a parent based on signal strength and energy levels. A key innovation of WASP is its use of special control messages called “WASP-scheme” messages. These messages serve two purposes at once: they assign time slots for MAC-level communication and carry routing information about data requirements. This dual function reduces control overhead and energy consumption by avoiding the need for separate MAC and routing messages. WASP also schedules sleep times for each node, so they only turn on their radios when needed, saving more energy. Simulations showed strong performance benefits; for instance, WASP achieved a consistent delay of about 0.32 s with no packet loss, while a CSMA-based protocol showed more delay variation (over 0.35 s) and dropped up to 30% of packets. Thanks to its scheduled and collision-free design, WASP enables high sleep durations for nodes (e.g., one node slept through 16 out of its time slots), significantly reducing energy use while maintaining reliable data transmission.

An additional cross-layer routing protocol has been proposed in [79]. It focuses on establishing efficient data paths between sensors and the personal coordinator, optimizing for energy efficiency, reliability, and QoS by managing different data priorities and mitigating network congestion. This protocol achieves its cross-layer functionality by tightly integrating network layer traffic prioritization with MAC layer channel access and scheduling. The network layer establishes a reverse tree routing mechanism, initiated by the personal coordinator broadcasting beacon messages to form paths with minimal delay. Crucially, it categorizes data into priorities (emergency, medical, general monitoring). This criticality information is directly passed to the MAC layer, which then leverages IEEE 802.15.6 features. Specifically, high-priority (emergency and delay-sensitive) packets are preferentially scheduled within exclusive access phases (EAP1/EAP2) of the MAC superframe, while lower-priority packets utilize random access phases (RAP1/RAP2) and a contention access phase (CAP). Furthermore, for contention-based access methods, the contention window (CWmin, CWmax) for CSMA/CA and contention probabilities (CPmin, CPmax) for Slotted Aloha are adaptively adjusted based on the packet’s priority (e.g., smaller for Emergency packets to ensure faster channel access). This direct mapping of network-level criticality to MAC-level channel access mechanisms resolves the conflict between QoS requirements and uncoordinated resource access, ensuring critical data’s timely and reliable delivery while saving energy by reducing retransmissions. To prevent packet starvation of lower-priority data during congestion, a remaining time (RT) scheduling mechanism is implemented at the MAC layer, ensuring eventual transmission by prioritizing packets that have waited longest in the buffer. Additionally, the coordinator dynamically manages traffic load based on these priorities (e.g., reducing general monitoring traffic by 75% under congestion but not affecting emergency traffic), preventing network congestion while preserving critical data integrity. Simulation results confirmed the protocol’s advantages. In the two-hop topology setup phase, it consumed less energy than existing protocols (e.g., OLSR), attributed to fewer beacon exchanges. Furthermore, the packet delivery ratio (PDR) evaluation across different packet categories demonstrated its effectiveness. While PDR generally decreased with increased traffic load, the proposed scheme ensured that emergency and delay-sensitive packets maintained significantly higher PDRs than general monitoring packets. The implementation of the remaining time (RT) technique notably improved the PDR for general monitoring packets, preventing their starvation and ensuring better overall network reliability for all data types.

Another cross-layer-based routing protocol has been introduced in [80]. It proposed several key enhancements to improve energy efficiency, network lifetime, and communication reliability, while also minimizing health impacts from electromagnetic exposure. This protocol achieves its multifaceted optimization through a novel cross-layer interaction primarily driven by a comprehensive cost function at the network layer and adaptive adjustments at the MAC layer. At the network layer, route selection is governed by a cost function that linearly combines three crucial metrics: energy ratio, assessing the residual energy of nodes to extend network lifespan; link reliability, evaluating signal strength and stability (derived from node distribution and distances) to ensure robust paths; and a specific absorption rate (SAR) optimization approach, which minimizes electromagnetic radiation exposure to body tissues by considering transmission power and node placement during route selection. This holistic routing decision, informed by energy levels, channel quality, and health constraints from lower layers, directly addresses the conflicts between extending network lifetime, ensuring reliable connectivity, and safeguarding user health. Subsequently, the MAC layer (based on IEEE 802.11) is adaptively adjusted; it introduces modified contention window (CW) sizing and backoff intervals based on the characteristics of the selected route (e.g., energy availability and link stability). These adaptive mechanisms dynamically manage channel contention to enhance throughput and reduce packet collisions, complementing the network layer’s routing choices. Additionally, a cooperative relay strategy is employed where nodes decide to forward route requests and replies based on residual energy, distance, and signal-to-noise ratio, further improving the efficiency of data relaying. Simulation results demonstrated significant improvements across various performance indicators. Compared to conventional protocols, the proposed method achieved a network lifetime of 93.1 s, a throughput of 0.655 Kbps, a packet delivery success rate of 0.866, and a residual network energy of 63.31 Joules after a 120-s simulation period with eight source nodes. The parametric variations of the cost function variables confirmed the protocol’s ability to balance trade-offs, showing that the contention window adjustment at the MAC layer, combined with the energy and reliability-aware routing, effectively enhanced packet delivery success rates and throughput by reducing packet queues and retransmission attempts. Nevertheless, the proposed protocol overlooks some important factors, such as body mobility and fluctuating traffic scenarios, which can impact optimal routing decisions in highly dynamic WBAN environments.

A cross-layer opportunistic MAC/routing (COMR) protocol has been proposed in [81] to enhance the reliability of WBANs, which is crucial for medical applications. The COMR protocol addresses key challenges, such as environmental interference, body shadowing, and energy efficiency, largely by using a sophisticated timer-based relay selection mechanism that integrates information from both the MAC and network layers. COMR’s cross-layer intelligence is centered on how the MAC layer’s channel access is dynamically controlled by network-level awareness. When a source node needs to transmit, it initiates a standard RTS/CTS handshake. However, instead of a random selection, potential relay nodes contend to send their CTS replies based on a calculated delay, τd. This τd is a composite metric directly incorporating network layer-relevant parameters: the node’s received signal strength indicator (RSSI), which implicitly reflects its proximity to the sink and the current link quality; and its residual energy, indicating its energy reserves. Nodes with higher residual energy and those closer to the sink (with better RSSI to the sink) will compute a lower τd value, enabling them to reply faster and win the MAC contention. This unique integration ensures that the “best” relay node—one with sufficient energy, a strong link, and positioned to move data efficiently towards the sink—is selected for the opportunistic forwarding, resolving the conflict between random MAC-level channel access and optimal network-level routing. The protocol’s use of a four-way handshake also allows non-participating nodes to enter a sleep state via network allocation vector (NAV), further contributing to energy savings. Simulation results demonstrated that COMR significantly outperforms the simple opportunistic routing (SOR) protocol [82], which lacks such cross-layer intelligence and relies on random relay selection. When varying payload size, COMR consistently showed higher network lifetime (e.g., improving as payload size increased due to better energy management) and lower end-to-end (ETE) delays (due to minimizing hops). It also exhibited improved energy efficiency (lower energy used per bit) because of its energy-aware relay selection, leading to higher packet reception at the sink. Critically, COMR achieved a higher packet delivery ratio (PDR), showcasing its enhanced reliability, as it avoids selecting relays that might forward packets away from the sink or are prone to premature death. These performance gains were also observed when varying the number of nodes, with COMR maintaining higher network lifetime and lower ETE delay compared to SOR due to more intelligent power consumption load division and minimized hop counts. Nevertheless, the proposed protocol overlooks some important factors, such as body mobility and fluctuating traffic scenarios, and no real-world results have been produced to validate its efficiency under actual interference conditions (mutual and cross interference), which makes determining its practical usefulness difficult. The same authors extended their research on the COMR protocol in another work [83] by adapting it for the Internet of Health Things (IoHT). They tried to address mutual interference by focusing on multi-WBAN setups where networks can collaborate for data transmission. The protocol was tested in intra- and inter-WBAN environments, showing again significant advantages over the SOR protocol in terms of network lifetime, reliability, and energy use. The inter-WBAN configuration, in particular, demonstrated enhanced performance, making the COMR protocol well suited for complex IoHT scenarios, such as medical wards, with multiple WBANs as well. Nevertheless, there is still a need to verify the efficiency of the proposed protocol in cross-interference scenarios as well.

A cross-layer design optimization (CLDO) scheme was proposed in [84] to improve energy efficiency, transmission reliability, and network lifetime in WBANs by coordinating decisions across the network, MAC, and physical (PHY) layers. At the PHY layer, CLDO uses power control algorithms to set the optimal transmission power for each link, ensuring low energy use while meeting delay and jitter requirements. At the network layer, it selects relay nodes based on their remaining energy, balancing the load and extending network lifetime. It also adjusts the power of each node depending on its energy level—nodes with more energy take on heavier tasks—improving overall performance and avoiding network bottlenecks. Additionally, CLDO optimizes packet size by minimizing header overhead from all protocol layers, which helps reduce energy consumption and improve responsiveness. Both theoretical and experimental results showed that CLDO increased transmission reliability by over 44.6% and extended network lifetime by up to 33.2%.

An optimized, cross-layer and thermal-aware converge cast protocol for IoHT was proposed in [85] to overcome key limitations of existing protocols, such as long delays, single points of failure, and high overhead. This protocol improves data transmission from sensor nodes to the central node by combining optimizations across different layers. At the MAC layer, it reduces redundant broadcasts and acknowledgments during setup, while at the network layer, it avoids single-root failures by using multiple parent-child relationships, lowering communication overhead. It also introduces a smart routing decision mechanism using a maximum benefit-cost function (MBCF), which considers multiple parameters, such as residual energy, temperature, path loss, link quality, hop count, and bandwidth. This ensures better load balancing, energy efficiency, and thermal management. The protocol uses a hybrid data collection strategy that switches between gossip-based and minimum attenuation methods, adapting dynamically to network and thermal conditions. For example, when a hotspot is detected, data is rerouted through cooler paths to maintain performance. Simulation results showed improved outcomes: a 19.4% delay reduction, a throughput increase from 8% to 13.75%, and only a 0.3% packet loss while controlling temperature rise. However, in dynamic environments with frequent node changes, the hybrid strategy may lead to unpredictable behavior, and managing multiple parent-child links could limit scalability as the network grows.

Table 5. Comparison of cross-layered energy-efficient routing protocols for WBANs. ✓ indicates that the feature is present; — indicates that the feature is not applicable.

Protocol	PLR Reduction	Latency Reduction	Mobility	Fault Tolerance	Interference Management	Data Prioritization
TICOSS [76]	✓	✓	✓	—	—	—
WASP [78]	✓	✓	✓	✓	—	—
[79]	✓	✓	—	✓	—	✓
[80]	✓	✓	—	✓	✓	—
COMR [81]	✓	✓	—	✓	✓	✓
COMR (IoHT) [83]	✓	✓	—	✓	✓	—
CLDO [84]	✓	✓	✓	✓	✓	—
[85]	✓	✓	—	✓	✓	✓

Table 5. Comparison of cross-layered energy-efficient routing protocols for WBANs. ✓ indicates that the feature is present; — indicates that the feature is not applicable.

Protocol	PLR Reduction	Latency Reduction	Mobility	Fault Tolerance	Interference Management	Data Prioritization
TICOSS [76]	✓	✓	✓	—	—	—
WASP [78]	✓	✓	✓	✓	—	—
[79]	✓	✓	—	✓	—	✓
[80]	✓	✓	—	✓	✓	—
COMR [81]	✓	✓	—	✓	✓	✓
COMR (IoHT) [83]	✓	✓	—	✓	✓	—
CLDO [84]	✓	✓	✓	✓	✓	—
[85]	✓	✓	—	✓	✓	✓

While cross-layered protocols are highly effective at enhancing energy efficiency by integrating decisions across multiple layers, they introduce several limitations and trade-offs. The primary challenge lies in their increased complexity and maintenance, as tightly coupled designs can compromise protocol transparency and make it difficult to manage and debug the system. Protocols such as TICOSS [76] may face scalability challenges in coordinating time zones and managing synchronization delays as networks grow larger or denser. Furthermore, some protocols, such as that in [80], may overlook crucial factors such as body mobility and fluctuating traffic, which can impact routing decisions in dynamic WBANs. Finally, a lack of real-world validation, as with the COMR [81] protocol, makes it difficult to assess their practical usefulness under actual interference conditions, which is a critical requirement for medical applications. These limitations underscore the need for future research to develop cross-layer protocols that can efficiently manage these trade-offs and provide robust, scalable performance in real-world WBAN environments.

5.4. QoS-Based Energy Efficient Routing Protocols

QoS is a very important factor for WBANs due to the unique challenges posed by these networks, such as severe resource constraints, critical environmental conditions, and the need for real-time data transmission. As opposed to traditional WSNs, WBANs must handle diverse traffic types from various sensors with different QoS requirements, making priority management essential. The main QoS considerations include minimizing transmission delays for time-sensitive healthcare data, ensuring energy efficiency to prolong the network’s lifetime—especially for implantable sensors with irreplaceable power sources—and addressing significant path loss in body-based communication. Effectively managing these aspects is crucial for reliable and efficient WBAN performance, particularly in healthcare applications [86]. In their paper, Yessad et al. [86] divided QoS-aware WBAN routing protocols into two main categories: multi-sink and single-sink approaches. Multi-sink protocols aim to enhance reliability and transmission speed by routing data to multiple sinks. The single-sink approach is further split into single transmission mode, using multi-hop routing, and hybrid transmission mode, combining single-hop and multi-hop routing based on data priority.

This section reviews the main QoS-aware WBAN routing protocols that have been proposed in the literature to enhance energy efficiency in these systems. These protocols are summarized and compared in

Table 6. Comparative analysis of QoS-aware energy-efficient routing protocols in WBANs. ✓ indicates that the feature is present; — indicates that the feature is not applicable.

Protocol	Energy Efficiency	Fault Tolerance	Latency Reduction	PLR Reduction	Mobility	Load Balancing	Data Prioritization
DMQoS [87]	✓	✓	✓	✓	—	✓	✓
EBRAR [88]	✓	✓	✓	—	—	✓	✓
[89]	✓	—	✓	✓	—	—	—

.

The DMQoS protocol proposed in [87] is a data-centric, multi-objective QoS-based routing protocol for WBANs that focuses on optimizing energy efficiency, alongside improving delay and reliability. DMQoS aims to differentiate route selection based on the type of data packets, addressing QoS requirements in delay, reliability, and energy cost to enhance overall network performance. The protocol employs a modular architecture with five key components, including a dynamic packet classifier that categorizes data into ordinary, critical, delay-driven, and reliability-driven packets, forwarding each to its respective module. For energy efficiency, the energy-aware geographic forwarding module selects the nearest next-hop node while considering residual energy, thus balancing energy consumption across the network. The modular approach allows DMQoS to use localized, hop-by-hop routing decisions without the need for global path discovery and maintenance, which conserves energy and enhances scalability to large-scale sensor networks. However, while DMQoS excels in reducing operational energy overhead and managing diverse traffic flows, its reliance on localized routing decisions can lead to increased traffic loads, potentially causing congestion and affecting end-to-end latency or reliability, and the assumption of having location detection capabilities can be challenging in WBAN environments.

Ababneh et al. [88] introduced the energy-balanced rate assignment and routing protocol (EBRAR) designed to enhance WBAN performance by balancing energy consumption and maintaining QoS for real-time data streaming. Unlike static routing approaches, EBRAR dynamically selects routes based on node depth and residual energy to evenly distribute the data forwarding load and extend network lifetime. The protocol employs adaptive resource allocation to prioritize high-utility data streams, adjusting bandwidth as conditions change to ensure reliable data collection. EBRAR operates in four phases—topology discovery, routing tree construction, rate assignment, and state transition—with a focus on energy balance and load distribution. Simulation results demonstrated that EBRAR’s path-energy (PE) strategy significantly outperformed its path-degree (PD) and shortest-path (SP) counterparts. Specifically, EBRAR-PE achieved the best performance in maintaining higher residual energy across nodes, leading to a notably extended network lifetime. For instance, it showed a lifespan of over 4000 s, which was considerably longer than the other strategies, depending on the sink location. This approach helps maintain the network’s operational efficiency and extends its lifetime, though EBRAR-PE may experience a drop in utility performance when the number of high-priority nodes exceeds 10, and it faces challenges related to optimal rate assignment and maintaining sensor node availability.

The authors of [89] have introduced a relay-based routing protocol designed to enhance energy efficiency in in vivo WBANs. The proposed protocol minimizes energy consumption by reducing communication distances through the use of relays and a coordinator placed on the patient’s clothing. This setup ensures that in-body sensors communicate directly with nearby relays rather than each other, thus minimizing the energy required for data transmission and processing. The protocol employs linear programming models to optimize network lifetime and ETE delays. MATLAB simulations comparing the proposed BAN against single-hop BANs, multi-hop BANs, and CH-rotate BANs demonstrated significant quantitative improvements across key metrics. Notably, the proposed BAN exhibited the highest remaining energy and achieved the maximum stability period and network lifetime, extending operational duration to around 8000–10,000 rounds, considerably outperforming other protocols, which ceased operation much earlier (e.g., CH-Rotate BAN around 2000–3000 rounds). Furthermore, it showed the least packet drop rate, maintaining a low number of dropped packets (~1–2 at 10,000 rounds), and proved to be the most stable system with the lowest average number of packets in the system. However, the protocol does not explicitly address issues related to packet retransmission in case of transmission errors, and the deployment of sensor nodes on the body may not always be practical.

QoS-based protocols aim to balance energy efficiency with performance but face challenges such as high computational complexity, scalability issues, and added overhead from traffic prioritization methods [87]. Some rely on specific hardware [89], and many fail to address key reliability concerns, such as packet retransmission, which is critical in medical applications. These gaps highlight the need for more efficient QoS-aware protocols that ensure both energy savings and reliable data delivery in complex WBAN environments.

5.5. Postural Movement-Aware Energy-Efficient Routing Protocols

Postural movement-aware routing protocols are mainly designed to address the challenges posed by body movements and limited radio frequency (RF) transmission ranges in WBANs, which can cause network partitioning and link disconnections [90]. To mitigate these issues, various protocols have been developed that dynamically adjust the communication cost factor, which is defined as the ratio of the total energy consumed in the network to the average packets successfully delivered to the sink. These protocols aim to minimize the energy cost of transferring data from the patient’s body to the network coordinator.

This section focuses on postural movement-aware energy-efficient routing protocols for WBANs. A review of the main QoS-aware routing protocols is presented, highlighting their contributions to enhancing energy efficiency in these systems. These protocols are summarized and compared in Table 7.

In their paper, Quwaider et al. [91] addressed the challenge of routing in WBANs that are frequently disrupted by short RF transmission ranges and human postural movements. WBANs, commonly used for health monitoring, rely on low-power RF transceivers with limited communication ranges, which makes them vulnerable to network partitioning. Network partitioning occurs when wireless links between body-worn sensors are interrupted due to movement or RF limitations. To mitigate this issue, the paper proposes store-and-forward routing algorithms, probabilistic routing protocol with link cost (PRPLC) and distance vector routing protocol with link cost (DVRPLC), that aim to minimize end-to-end packet delays and reduce energy consumption by dynamically selecting routes with lower storage delays and fewer hops. The proposed protocols are based on a stochastic link cost model, which accounts for spatial and temporal disconnection patterns caused by body movement. A prototype WBAN is developed to characterize these disconnections and validate the proposed algorithms. The routing protocols, which include both probabilistic and distance vector frameworks, are evaluated experimentally and through simulations. In a seven-sensor prototype network with varying human postures, the adaptive versions of the proposed protocols significantly reduced end-to-end packet delays. Experimentally, DVRPLC achieved an average packet delay of 3.11 s, and PRPLC achieved 3.6 s, demonstrating superior performance compared to existing Delay Tolerant Network (DTN) routing methods, such as PROPHET (simulated 5.79 s), UTILITY (simulated 4.05 s), and especially OPPT (simulated 34.8 s). Simulations also showed that including a transitive link likelihood factor update further improved packet delivery delay by approximately 6%. While the packet hop count (PHC), a direct measure of energy expenditure, was sometimes slightly higher for the proposed protocols to achieve these lower delays, their overall approach demonstrated improved energy efficiency compared to other single-copy DTN methods. However, the proposed routing approach may be less effective in larger WBANs with a large number of sensor nodes due to increased complexity in managing multiple routes and potential bottlenecks, and does not explicitly address packet retransmission errors.

In the same context, Maskooki et al. [92] addressed the issue of energy consumption in WBANs, where dynamic link quality and frequent disconnections due to body movements pose significant challenges. They proposed an opportunistic routing scheme that exploits natural body movements, such as hand motion during walking, to enhance network lifetime. The protocol dynamically determines whether to send data directly to the sink node or use a relay node, depending on the positioning of the nodes and line-of-sight conditions. For instance, when the sensor on the chest cannot directly communicate with the sink on the wrist due to body movement, the relay node on the waist is activated to facilitate the communication. Simulation results demonstrated that the proposed opportunistic routing significantly reduces energy consumption at the individual node level. Specifically, it showed the least energy consumption in the sensor node compared to multi-hop and single-hop schemes. More remarkably, the energy usage in the relay node for the opportunistic scheme was half the amount of the multi-hop scheme. While the overall network average energy consumption per bit for the opportunistic scheme was 5.78 nJ (compared to 3.85 nJ for single-hop and 7.70 nJ for multi-hop), indicating an overhead from the relay, this was effectively managed by using the relay only when necessary. Crucially, by efficiently managing energy resources through strategic relay node placement and dynamic route selection, the protocol achieved the best network lifetime among all compared schemes by reducing energy consumption in both sensor and relay nodes simultaneously, all while maintaining similar bit error rates (BER). This approach effectively reduces the energy overhead, especially in the relay node, which is typically the most power-hungry component in the network.

On the other hand, Goyal et al. [93] presented a hybrid data delivery mechanism for Wireless WBANs, designed to improve energy efficiency and address the challenges posed by postural mobility and network partitioning. The protocol combines RF and body-coupled communication (BCC) links to optimize communication, prioritizing BCC for emergency data due to its lower energy consumption and reliability. When network partitioning occurs, a secondary path using the BCC link ensures continuous communication, particularly for critical data. The protocol uses a TDMA approach to prevent collisions, further reducing energy consumption by eliminating retransmissions. The reconfiguration process is triggered by the personal digital assistant (PDA), which updates the network routing information using BCC to minimize energy use. However, while BCC gives priority to emergency data, it may still be sensitive to interference, which may affect data transmission quality.

In her thesis, Haq [94] introduced a postural movement-based routing mechanism designed specifically for WBANs. It addresses the unique challenges posed by the mobility of sensor nodes attached to the human body. As the user changes posture, the protocol allows sensor nodes to dynamically select optimal forwarder nodes based on their positions and distances from each other, categorized into best, average, and worst-case scenarios. This adaptive routing strategy ensures that data is transmitted efficiently with minimal delay and reduced energy consumption. By prioritizing closer nodes for data forwarding, the protocol minimizes the energy expenditure associated with longer-distance transmissions, effectively reducing the overall temperature rise of sensor nodes. Additionally, the implementation of a cost function that considers residual energy, node temperature, and hop count further enhances energy efficiency, allowing for prolonged network lifetime and stability, all while maintaining reliable data transmission.

Table 7. Comparative analysis of postural movement-based energy-efficient routing protocols. ✓ indicates that the feature is present; — indicates that the feature is not applicable.

Protocol	Energy Efficiency	Fault Tolerance	Latency Reduction	PLR Reduction	Data Prioritization	Route Stability	Network Stability
[91]	✓	—	✓	—	—	—	—
Maskooki et al. [92]	✓	—	—	—	—	—	—
Goyal et al. [93]	✓	—	✓	—	—	—	—
Pm-EEMRP [95]	✓	—	—	—	✓	—	✓
Newell and Vejarano [96]	✓	—	—	✓	—	—	—
EPRS [97]	✓	—	✓	✓	—	✓	—
Yang et al. [98]	✓	✓	✓	—	—	—	—

Table 7. Comparative analysis of postural movement-based energy-efficient routing protocols. ✓ indicates that the feature is present; — indicates that the feature is not applicable.

Protocol	Energy Efficiency	Fault Tolerance	Latency Reduction	PLR Reduction	Data Prioritization	Route Stability	Network Stability
[91]	✓	—	✓	—	—	—	—
Maskooki et al. [92]	✓	—	—	—	—	—	—
Goyal et al. [93]	✓	—	✓	—	—	—	—
Pm-EEMRP [95]	✓	—	—	—	✓	—	✓
Newell and Vejarano [96]	✓	—	—	✓	—	—	—
EPRS [97]	✓	—	✓	✓	—	✓	—
Yang et al. [98]	✓	✓	✓	—	—	—	—

In [95], a postural movement-based energy-efficient routing protocol called Pm-EEMRP is proposed for intra-WBSNs, focusing on challenges caused by body movement. The protocol uses relay nodes placed in the patient’s clothing to collect and forward data from biosensors to the body network controller (BNC). It separates critical data (sent directly to the BNC) from normal data (sent via relay nodes), improving both energy use and reliability. MATLAB simulations showed strong results; Pm-EEMRP improved network stability by 184% and 19.41% compared to multi-hop and forward-based protocols, and achieved 156.49% and 45.79% higher throughput, respectively. It also increased network lifetime by 11.26% over multi-hop, though it was slightly lower (−8.19%) than forward-based protocols. However, Pm-EEMRP maintained better energy levels across nodes, showing effective energy balancing through relay use. These results make it a promising option for health monitoring in dynamic conditions. Still, interference issues were not evaluated, leaving its performance in noisy environments uncertain.

Newell et al. [96] developed a WBAN protocol that improves energy efficiency by adjusting routing paths and transmission power based on body movement. It uses a multi-hop strategy and a token management system to avoid loops and reduce delays. Sensors increase their power just enough to ensure reliable packet delivery to the base station (BS), reducing the chance of token loss. Tests on a prototype WBAN showed notable energy savings; the dynamic routing and transmission power control (TPC) algorithm cut average power use by 23.4% to 50.4% compared to always transmitting at maximum power. It also maintained high reliability, with packet delivery ratios (PDRs) between 91% and 95%, exceeding the 90% target. The algorithm further reduced RSSI fluctuations—for example, lowering RSSI standard deviation by 60.2% on key links. Compared to other body-motion-aware methods, this approach achieved similar or better energy savings with lower overhead, needing only about 11% of the power required by competing protocols. However, it depends on commercial transceivers with limited power control precision and may struggle with posture estimation due to noisy motion data. It also assumes all nearby nodes receive transmissions, which might not always be the case.

In the same context, authors in [97] introduced the enhanced path reliable stable (EPRS) routing protocol for WBANs, focusing on maintaining route stability and reliable data transmission in healthcare applications despite disruptions caused by body movements. EPRS addresses the issue of network disconnections by dynamically selecting the most stable and energy-efficient links using two key factors: the route stability metric and the expected probability of a link remaining connected. By minimizing unnecessary re-routing and avoiding unreliable links, the protocol conserves energy while ensuring the timely dissemination of critical health data. Evaluated using network simulator-2 (NS-2) across varying data rates (50–250 Kbps), EPRS demonstrated superior quantitative performance compared to existing state-of-the-art protocols, such as SRE, MCDR, and AODV. Specifically, EPRS exhibited significantly lower end-to-end delays due to minimized packet re-transmissions and reduced routing packet flow. It also achieved a substantially lower normalized routing load, indicating a more cost-effective management of routing overhead by maintaining stable routes for longer durations. Furthermore, EPRS consistently showed significantly higher route stability, which directly contributed to its improved throughput performance, especially under high network loads where other protocols experienced decline. By minimizing unnecessary re-routing and avoiding unreliable links, the protocol conserves energy while ensuring the timely dissemination of critical health data.

On the other hand, Yang et al. [98] proposed a probabilistic, behavior-aware routing mechanism (BAPR) for WBANs that leverages inertial sensor data and historical link quality to address the challenges of dynamic network topology caused by human movement. By employing multi-hop routing, the protocol enhances communication reliability and efficiency by selecting reliable relay nodes in real time, which reduces channel conflicts and improves data delivery ratios. This approach is particularly effective during rapid body movements, allowing for adaptive routing based on user posture. Additionally, it significantly lowers energy consumption compared to single-hop communication, as it optimizes the use of lower transmission power across multiple hops, ultimately extending the operational life of the network. Experimental evaluations in walking and running scenarios on a five-node WBAN platform demonstrated BAPR’s quantitative advantages. In walking scenarios, BAPR’s average delivery ratio was comparable to optimal forwarding routing (OFR), a lower bound benchmark, and was 30% higher than the PRPLC protocol. While the delivery ratio decreased in running scenarios, BAPR still outperformed PRPLC. In terms of average end-to-end delay, BAPR, OFR, and PRPLC generated similar satisfactory delays across both walking and running scenarios. Furthermore, regarding resource utilization and energy consumption, BAPR achieved an average number of hops slightly better than PRPLC and several times lower than OFR, indicating improved energy efficiency.

While postural movement-aware protocols are highly effective for maintaining network connectivity and energy efficiency in dynamic WBANs, they are not without limitations. A primary challenge is the complexity of managing multiple routes and the potential for increased overhead, particularly in larger networks with a significant number of sensor nodes. For instance, protocols such as DVRPLC and PRPLC [91] may face scalability issues and do not explicitly address packet retransmission errors, which can affect data reliability. The protocol in [95] also raises a concern about un-evaluated interference issues, making its performance in noisy environments uncertain.

6. Complementary Energy-Efficient Data Processing Techniques

In addition to MAC and routing strategies, improving energy efficiency in WBANs can be effectively achieved through data reduction techniques, including data aggregation, data fusion, and data compression. These approaches reduce the volume of transmitted data—one of the most power-intensive operations in sensor nodes—thereby conserving energy and extending network lifetime. Figure 6 provides a visual summary of these three concepts.

6.1. Data Aggregation

Data aggregation involves combining multiple data packets into a compact form before transmission, minimizing redundancy and reducing communication overhead. Mehmood et al. [99] proposed a mobile agent-based data aggregation scheme tailored to WBANs, where the network is divided into clusters and a mobile agent, dispatched from the base station, visits each cluster-head to collect aggregated data. This approach significantly reduces transmission frequency, lowers energy consumption, and improves reliability.

Similarly, Samanta et al. [100] developed DEBA (delay- and energy-efficient big data aggregation), a quality-driven framework for cloud-assisted WBANs. DEBA optimizes both intra-BAN (sensor to local managing unit) and inter-BAN (local managing unit to base station/cloud) communications. Simulations involving 400 WBANs, each comprising eight sensors, deployed over a 3.5 km × 3.5 km area under the IEEE 802.15.6 standard, demonstrated that DEBA reduces aggregation delay and cost while improving served traffic by 5–7% and decreasing energy consumption by 7–8%.

6.2. Data Fusion

Unlike simple aggregation, data fusion combines information from multiple sensors to produce more accurate and informative results while reducing the number of transmissions. One notable work [101] presented an energy-aware decision-making framework for WBANs, fusing physiological signals (e.g., heart rate, SpO₂, temperature) at the coordinator level. Using fuzzy set theory and a decision matrix based on the national early warning score (NEWS), the system interprets uncertain measurements to support medical decisions. It adjusts monitoring frequency based on patient condition and triggers immediate responses only when necessary. Simulations using real-world medical data confirmed reduced transmissions, improved energy efficiency, and enhanced responsiveness.

In a similar context, FASR-LED [102] is an energy-efficient WBAN strategy that combines fuzzy logic and high-level data fusion. At the sensor node, it filters unnecessary data locally; at the coordinator, it uses a fuzzy inference engine to analyze and merge vital signs. This dynamic adjustment of sampling rates reduces redundant transmissions and avoids complex synchronization across sensors. The method reduces energy consumption by 50% and the volume of transmitted data by 30%.

6.3. Data Compression

Compression techniques further reduce the size of data before transmission and are categorized into lossless, lossy, or hybrid methods. The EDC-ER approach [103] combined long short-term memory (LSTM) neural networks for data prediction (lossy) with run-length encoding (RLE) (lossless) to compress repetitive physiological information. Operating directly at the sensor level, EDC-ER significantly lowers data transmission and battery usage. Tests on real signals (e.g., heart rate, blood pressure, respiration, SpO₂) show energy savings of up to 98%, with low prediction error and high compression ratios.

GROWN [104], a hybrid compression method, adapts the technique based on the type of physiological signal by combining lossy and lossless compression. It uses a modified Exponential-Golomb algorithm and includes a compression module within wearable devices that encodes variations in sensor readings. The decompression is handled at the sink device. Experiments using Arduino-based wearables and smartphones demonstrate energy savings up to 53.73%, while maintaining real-time constraints for medical applications.

The work in [105] introduced a secure WBAN framework that integrates compressive sensing (CS) to sample and compress physiological signals, significantly reducing data volume and transmission overhead. It incorporates a lightweight blockchain-based consensus protocol tailored for constrained environments. This design achieves compression ratios up to ~88%, reduces signal distortion (PRD) by up to ~34%, and reaches consensus ~24% faster with CPU usage under 14%.

In the same context, SHDC (smart hybrid data compression) [106] addressed high WBAN energy consumption by combining LSTM-based data prediction with RLE compression of the prediction error. This hybrid approach is particularly suitable for physiological data with high temporal redundancy. Executed directly at the sensor layer, SHDC achieves an average of 98% energy savings while maintaining high prediction accuracy and outperforming standard methods, such as Huffman, LZW, and arithmetic coding.

7. Toward Green Communication in WBANs: Innovations in Energy Harvesting

Energy harvesting (EH) technology is one of the most important approaches to offset the scarcity of energy resources in WBANs; it makes use of ambient and human body sources capable of generating significant energy to power wearable and implantable devices effectively [107]. This approach enhances the operational lifespan of energy-limited devices, promoting self-sufficiency in communication networks. EH in WBANs follows two main architectures: the “harvest-store-use” and “harvest-use” models. In the “harvest-store-use” model, harvested energy is stored in batteries or supercapacitors before powering devices, while the “harvest-use” model directly utilizes harvested energy, offering a simpler and more cost-effective alternative [108].

Based on energy sources, EH techniques are classified into ambient and human body categories [107], as illustrated in Figure 7. Ambient energy sources, such as solar or RF energy, are ideal for charging wearable devices in on-body applications. In contrast, human body energy sources, including biochemical, biomechanical and thermal energy, are better suited for powering implanted sensors in in-body applications.

7.1. EH Based on Ambient Sources

7.1.1. Photovoltaic (PV) Energy

PV technology harnesses solar energy through solar cells, converting it into electrical power for WBAN nodes. This power is used during active data transmission and for battery recharging during rest periods [109]. Furthermore, given the intermittent nature of solar energy, it becomes crucial to store excess energy in the battery during periods of availability. This stored energy serves as a reserve, ensuring an uninterrupted power supply for nodes with higher energy demands, such as relay nodes in multi-hop WBAN topologies. The conceptual architecture of a photovoltaic (PV) energy harvesting system for WBANs is illustrated in Figure 8.

The potential of solar EH to power nodes within WBANs has been extensively explored in various studies. Wu et al. [110] presented a prototype of on-body sensor nodes powered by solar energy and operating through Bluetooth low energy (BLE) technology. This innovative design is tailored for critical healthcare applications, including heartbeat monitoring, body temperature measurement, and fall detection. The proposed EH module incorporates a flexible solar panel and employs an output-based maximum power point tracking (MPPT) technique to enable continuous operation of sensor nodes, even under variable light conditions. The findings demonstrated that by leveraging efficient active-standby mode procedures, the sensor node could achieve continuous, autonomous operation for 24 h, showcasing the feasibility of integrating solar EH into WBAN applications. The work presented in [110] builds upon an earlier on-body sensor prototype introduced by the same authors. This earlier design also leveraged solar EH, focusing on applications such as heartbeat measurement and patient activity monitoring [111]. However, the aforementioned prototype [111] consumed significantly more energy (28.98 mW) and had a lifespan of only 12 h when the solar panel was exposed to sunlight for 3 h.

Similarly, Tran et al. [112] introduced an EH module employing a flexible solar panel for an on-body sensor node specialized in monitoring heartbeats and ECG. The solar EH technique also depends on an MPPT controller, which adjusts to both indoor and outdoor conditions, ensuring optimal energy extraction. Moreover, this module significantly enhances the lifespan of the proposed sensor. In an indoor environment with a minimal radiation of 0.22 mW/cm², the sensor’s lifespan extends to 41 h. However, in shaded areas characterized by a radiation of 2.9 mW/cm², this duration extends up to 384 h.

In outdoor environments, battery charging is feasible. Furthermore, the charging duration depends on the radiation intensity. With a radiation level of approximately 78.7 mW/cm², battery charging can be completed in 4 h, while this duration increases to 6 h when the radiation is at 24 mW/cm². Despite these enhancements in terms of lifespan, the proposed energy recovery module consumes approximately 41.25 mW of energy when transmission is continuous.

More recently, and with the ongoing objective of increasing sensor lifespan through solar energy utilization, Mohsen et al. [113] have also introduced a prototype of an on-body WBAN node based on photovoltaic EH using a flexible solar panel designed to operate on the human body surface. This innovation ensures a lifespan of 246 h while consuming a power of 4.97 mW. The proposed sensor node employs BLE technology for data transmission and is dedicated to monitoring body temperature and heart rate.

7.1.2. Mechanical Energy

The generation of electrical energy from diverse mechanical sources, including pressure, vibrations, and tension/distortion of materials, represents a pivotal frontier in the realm of EH. This phenomenon is facilitated through various transduction mechanisms, such as electromagnetic, electrostatic, piezoelectric, and triboelectric processes.

Electromagnetic energy harvesting specifically leverages Faraday’s Law of Induction to convert mechanical motion into electrical energy. Its working principle involves the relative movement between a coil (conductor) and a magnetic field. As the magnet moves through or near the coil, it induces a current in the wire. This mechanism is robust and can potentially generate significant power, especially from larger, more consistent motions [114]. A key advantage is that it does not require an initial charge or external power source to initiate the energy conversion. However, for WBAN applications, significant challenges include achieving sufficient miniaturization and low-profile designs for wearable integration, especially for the antenna/rectenna systems. Furthermore, maintaining stable performance (e.g., consistent signal matching and radiation patterns) is difficult due to the significant impact of human body proximity, curvature, and varying tissue layers. Optimizing energy conversion efficiency from ambient RF sources, which can fluctuate in strength and frequency, and ensuring robust performance despite dynamic body movements and changing orientations, also remain active areas of research [115].

Electrostatic harvesting employs a variable capacitor that converts mechanical movement into energy, but needs an initial charge. Piezoelectric harvesting converts mechanical strain into electricity using piezoelectric materials, requiring no external power source, making it efficient and suitable for biomedical and WBAN applications [116]. Triboelectric EH operates on a similar principle by converting mechanical energy from body movements, such as friction or vibration, into electrical energy through the triboelectric effect, using materials with different electron affinities. This method is particularly advantageous for wearable devices due to its simple design and ability to efficiently harvest energy from human motion [107].

In the context of WBANs, the human body can serve as a source of mechanical energy during voluntary movements, such as walking, running, and jumping, as well as inherent physiological processes, such as breathing and heartbeats (biomechanical energy sources). These biomechanical energy sources will be further discussed. Bodily motions provide mechanical tensions and deformations that serve as input stimuli for EH devices, such as piezoelectric transducers and triboelectric generators [117]. These transducers convert mechanical vibrations into electrical energy, thereby supplementing the power requirements of WBAN nodes.

One of the primary hurdles of mechanical EH in WBANs lies in the stochastic nature of body movements, characterized by the prevalence of low-frequency vibrations typically below 1 kHz. Hence, linear resonance-based vibrational EH systems demonstrate optimal performance solely when stimulated near their precise resonance frequencies, constraining their efficacy to narrow frequency bands [118].

In the literature, various studies have suggested WBAN nodes utilizing mechanical EH techniques. Hamid et al. [119] introduced an initial prototype incorporating piezoelectric and electromagnetic EH. The “WE harvest” prototype is designed to extract energy from low-frequency vibrations, primarily originating from body movements. The authors employed a transducer utilizing the motion of magnets to induce current flow in copper wires and exert kinetic force on piezoelectric diaphragms, thereby generating electrical energy in the form of an alternating current (AC). However, for meeting the power demands of sensor node electronic modules and battery charging, a consistent direct current (DC) is essential. Consequently, the authors integrated a power conditioning circuit to regulate the output current from the transducer, ensuring stability. In a subsequent study, the same authors [120] enhanced the initial prototype by integrating three conditioning circuit topologies to optimize energy extraction from human body movements. The refined prototype underwent testing across a range of physical activities, including walking and exercising. Remarkably, during a bicycle ride at speeds of 35–38 km/h, the developed prototype yielded 75.6 µJ of energy.

The hybrid approach to harnessing mechanical energy from the human body has once again been adopted in [121] where the authors proposed a self-powered sensor node. This time, they employed a hybrid model of mechanical energy recovery, which combines a Peltier module with a piezoelectric module. The analysis delved into the amount of energy harvested by these two modules across various scenarios: (1) rest state, (2) walking state, (3) running state, and (4) fast running state.

The incorporation of rotational piezoelectric modules for capturing energy from human body movements was initially introduced in [122]. The efficacy of these rotational modules stems from their ability to harvest energy from the random movements of the human body. Figure 9 presents the conceptual framework of mechanical energy harvesting in WBANs.

7.1.3. Thermal Energy

Thermal EH (TEH) involves the generation of thermal energy, subsequently converting it into electric current through various methods. In WBANs, it consists of making use of the natural heat that comes from the human body or other nearby sources to power sensor nodes, and thus enhancing energy efficiency in these systems. Among TEH devices, thermo-electric generators (TEG) stand out as a widely recognized example. An illustrative commercial instance is the thermo-electric watch introduced by Seiko, designed to harness the heat emitted by the human body. TEG devices exhibit a maximum power generation capacity ranging from 1 to 60 micro watts per square centimeter at a 5 K temperature [123].

Researchers have noted variations in TEH systems employing thermo-electric conversion under different environmental conditions, such as air and ground. Notably, TEH systems have been strategically positioned in greenhouses utilizing solar thermal EH systems. In this context, TEG devices play a crucial role in recharging batteries. A specific example involves the generation of approximately a 25 K delta T, yielding energy capable of rapidly recharging an 80 mAh battery [124].

In contrast, commonly available thermo generators necessitate temperatures ranging from 10 to 200 degrees Celsius to generate thermal energy. This requirement poses challenges in enclosed environments due to diverse weather conditions, making the consistent production of thermal energy a complex task [125].

7.1.4. Radio Frequency Energy (RF)

RF energy harvesting is a promising technology for extending the lifespan of sensors, particularly in applications such as long-term, low-power medical monitoring. By converting ambient RF signals from sources such as Wi-Fi and cellular networks into electrical energy, it enables sensor recharging without relying on external environmental factors, such as sunlight. Despite the limited energy harvested, the use of efficient components, such as optimized antennas, rectifiers, and power management units can enhance energy conversion and storage [107]. Compact and flexible antenna designs are particularly valuable for wearable device integration. Furthermore, dedicated relay nodes are often deployed to use the harvested energy for forwarding sensor data, ensuring consistent operation even when RF signals are intermittent. With these advancements, RF EH offers a sustainable solution for powering WBAN devices while addressing energy efficiency challenges [108].

EH techniques for monitoring human health have been widely discussed in the literature. However, when it comes to animals, there has been limited research. In [126], the use of RF EH-based sensor networks for monitoring animal health is explored. The paper addresses energy management and routing challenges in radio frequency EH sensor networks (RF-EHSNs) for animal healthcare. It introduces a path-oriented method to optimize energy requests, balancing energy efficiency and transmission costs via an energy tunnel. The paper also tackles dynamic network topologies resulting from animal movement and the frequent state switching of EH Nodes (EHNs). It proposes species-dependent routing strategies and gesture recognition to predict topology changes, ensuring efficient and reliable data transmission despite intermittent node connectivity. The main contributions include energy-efficient request strategies and dynamic routing approaches for RF-EHSNs.

Figure 10 illustrates a conceptual overview of radio frequency (RF) energy harvesting in WBANs. The diagram presents how ambient RF signals from sources such as Wi-Fi access points and cellular base stations are captured by compact, flexible antennas, converted into DC power via rectifiers and power management circuits, and stored for later use by WBAN sensor nodes. It also highlights the role of dedicated relay nodes in maintaining data transmission when RF signals are intermittent.

7.2. EH Based on Human Body Sources

The human body can serve as a valuable source of energy to improve the efficiency of sensors used in in-body and on-body applications. Energy generated by biological and mechanical processes, such as muscle movements or other physical activities, can be captured and converted into electrical power. By reducing dependence on external power sources, human body-based EH contributes to more reliable and autonomous operation of WBANs [107].

Human body-based EH takes advantage of biochemical and biomechanical processes that occur within the human body to produce energy. Hence, numerous research efforts have focused on developing EH modules from the human body to power sensors within WBAN systems. These studies explore various mechanisms, including biofuel cells [127], thermal [128], piezoelectric [129], triboelectric [130], electromagnetic [131], and electrostatic generators [132], as well as hybrid solutions [133], to efficiently capture biochemical and biomechanical energy. Such approaches aim to address the dual challenges of energy autonomy and device biocompatibility, offering promising avenues for powering wearable and implantable medical devices.

7.2.1. Biochemical Energy Sources

Biochemical energy harvesting (EH) uses the body’s natural chemical processes—such as glucose oxidation, lactate metabolism, or pH changes—to produce electricity through devices such as biofuel cells. This approach takes advantage of chemical reactions in the body to power wearable or implantable medical devices. Biofuel cells (BFCs), a type of fuel cell, generate energy by using biocatalysts to move electrons between an anode and a cathode [134]. There are two main types: microbial BFCs, which use microorganisms in a bioreactor and can work for up to five years, and enzymatic BFCs, which rely on enzymes to break down substances, such as glucose and lactate, in body fluids, such as sweat, blood, or tears. Enzymatic BFCs have a shorter lifespan (around 7–10 days) [135], but their use of readily available biofluids makes them very suitable for wearable and implantable uses [136]. Moreover, using flexible, biocompatible materials—such as carbon-based electrodes—improves their performance and integration in WBAN systems.

7.2.2. Biomechanical Energy Sources

Biomechanical energy harvesting (EH) converts energy from the body’s movements—both voluntary (such as walking) and involuntary (such as heartbeat or breathing)—into electrical power. This is achieved using technologies such as piezoelectric materials, triboelectric nanogenerators (TENGs), electromagnetic generators (EMGs), electrostatic generators (ESGs), and reverse electrowetting on dielectric (REWOD). Piezoelectric materials produce electricity when stressed, making them suitable for wearables such as shoes or wristbands that capture energy from motion. TENGs generate power through the triboelectric effect from movements such as breathing or arm swings, offering high efficiency at low frequencies and low cost [137]. EMGs use the relative motion between magnets and coils (based on Faraday’s law) to generate energy, ideal for repetitive actions, such as walking or joint movements [138]. ESGs harvest energy from small vibrations or motion by changing the gap between electrodes, working best at high frequencies [139]. REWOD creates electricity by pressing conductive liquid droplets between electrodes, increasing capacitance and producing power without external voltage—making it effective for low-frequency applications, such as motion sensors [140]. These technologies help build self-powered WBANs by reducing reliance on batteries and enabling continuous, long-term health monitoring.

Finally, energy harvesting offers a promising way to power WBANs, but it faces several key challenges. Ambient and body-based sources, such as sunlight for photovoltaic systems or body movements for mechanical harvesters, are often irregular and unpredictable, making a stable power supply difficult to maintain. Thermal harvesting requires significant temperature differences, which are not always available. Miniaturization and ensuring consistent performance remain difficult [107], as human body proximity and movement can reduce the efficiency of RF antennas and mechanical devices. Moreover, some biochemical and photovoltaic prototypes have short lifespans or produce insufficient power, underscoring the need for more efficient, durable, and reliable harvesting technologies.

8. Potential Improvements in EH for WBANs

8.1. Hybrid EH

The utilization of energy in WBANs can be further enhanced by adopting hybrid EH solutions. Hybrid EH involves integrating multiple EH techniques or sources to improve overall energy efficiency and reliability. By combining diverse methods, such as biochemical and biomechanical processes, or ambient sources, such as photovoltaic and RF energy, hybrid systems leverage the strengths of different mechanisms to overcome the limitations of individual approaches. This strategy is particularly critical in WBANs, where relying on a single energy source can lead to constraints in functionality and reliability, especially for devices connected to the human body. A hybrid approach not only addresses the energy insufficiency problem but also provides extra power and enhances the autonomy and operational stability of wireless nodes. Furthermore, hybrid solutions incorporate various transduction mechanisms, mixed materials, and innovative structures to maximize EH efficiency [107].

8.2. Advanced Machine Learning Techniques

Reinforcement learning (RL) and deep learning (DL) are two powerful machine learning techniques that can significantly improve EH in WBANs [107]. RL helps systems make smart decisions by interacting with the environment and using feedback. This allows WBANs to optimize energy use, adjust to changing conditions, and distribute energy efficiently between devices. RL also helps with battery management, communication strategies, and real-time decision making, which extends the life of devices and networks. In multi-WBAN settings, RL can also enable energy sharing between systems to improve performance [141].

On the other hand, DL uses advanced neural networks to predict energy availability from sources such as solar, thermal, and kinetic energy. By learning from past data, DL can forecast energy levels, improve charging strategies, and better allocate resources. DL also supports modeling user behavior, detecting anomalies, and ensuring the system’s reliability through multi-modal data fusion and fault tolerance. Together, RL and DL provide smart, adaptive solutions to energy challenges in WBANs, leading to more efficient and sustainable systems [142].

9. Discussion, Further Challenges and Future Directions

This section provides a comprehensive synthesis of the findings from the reviewed literature, directly addressing the research questions posed in Section 3.1. We first highlight the key mechanisms for energy efficiency, followed by a quantitative analysis of current research trends and gaps. Finally, we explore overarching challenges, promising future directions incorporating emerging technologies, and other critical considerations crucial for the sustainable and reliable deployment of WBANs.

9.1. Synthesis of Findings: Addressing Energy Efficiency in WBANs

This subsection summarizes the primary findings across the energy-efficient mechanisms reviewed, directly addressing research questions RQ1, RQ2, RQ3, and RQ4.

• Energy-Efficient MAC Protocols: At the MAC layer, multiple mechanisms have been developed to minimize power consumption, notably time division multiple access (TDMA), low power listening (LPL), and scheduled contention. These techniques aim to reduce idle listening, collisions, and retransmissions. TDMA-based approaches offer predictable scheduling and energy savings but are limited by synchronization overhead and reduced adaptability in dynamic environments. On the other hand, contention-based and hybrid methods provide greater flexibility, yet often suffer from increased latency or reduced reliability in high-density scenarios.
• Energy-Efficient Routing Protocols: Beyond medium access, the design of energy-aware routing protocols plays a pivotal role in achieving energy efficiency. The literature shows a broad categorization of protocols, including thermal-aware, cluster-based, QoS-driven, cross-layered, and postural movement-based strategies. These routing families are tailored to address WBAN-specific constraints, such as mobility due to body movement, node heterogeneity, real-time data delivery, and thermal safety. However, most studies treat routing in isolation, without adequately integrating it with other network layers or with application-level requirements, such as reliability and security.
• Energy-Efficient Data Processing Techniques: Complementing protocol optimization, this review also examined energy-efficient data processing techniques, such as data aggregation, data fusion, and data compression (addressing RQ1). These methods are crucial for reducing the volume of transmitted data, thereby directly lowering communication-related energy costs. By minimizing redundant transmissions and consolidating information at the source or intermediate nodes, these techniques effectively extend network lifetime, particularly in multi-sensor WBAN deployments, representing a vital strategy for optimizing energy consumption.
• Energy Harvesting (EH) Techniques: Complementing protocol optimization, energy harvesting (EH) has emerged as a promising solution to extend WBAN lifespan and reduce battery dependence. A wide variety of ambient and body-centric energy sources—such as photovoltaic, thermoelectric, and biomechanical—have been explored. Nevertheless, the power output of these sources often remains intermittent and insufficient for continuous operation. Hybrid harvesting systems, combining multiple energy sources, together with intelligent scheduling and predictive models based on machine learning, represent a promising yet underdeveloped research direction.

9.2. Quantitative Analysis of Research Trends and Gaps

This subsection provides a quantitative overview of the reviewed literature, identifying prominent research trends, highlighting areas of focus, and pinpointing existing research gaps, particularly concerning practical validation.

Figure 11 illustrates the growing research interest in energy-efficient WBANs over time. Only 1.7% of the reviewed articles were published between 2002 and 2005, followed by 11.2% between 2006 and 2010. Interest expanded considerably in the subsequent periods, with 24.7% of the papers published between 2011 and 2015 and 28.1% between 2016 and 2020. The most recent period (2021–2025) accounts for 34.3% of the total publications, confirming the sustained and increasing research focus on addressing energy efficiency challenges in WBANs.

Figure 12 presents the distribution of research focus across various categories of energy-efficient WBAN studies. Routing strategies (22.91%) and energy harvesting techniques (22.91%) jointly account for the largest share of the literature, closely followed by challenges and future trends (20.11%). MAC protocols represent 12.85% of the works, while specific surveys contribute 11.17% and general surveys 5.59%. Data processing techniques receive comparatively less attention, with only 4.47% of the references. This distribution highlights a balanced emphasis on routing and harvesting, alongside significant attention to overarching challenges, but also suggests opportunities to expand research in data processing and targeted protocol design to achieve more comprehensive and integrated energy-efficient solutions.

Figure 13 highlights a notable trend in validation methodologies employed in the reviewed literature. Our analysis indicates predominant reliance on simulation-based validation, with 64.2% of studies evaluating their proposals through simulations, while 23.5% employ theoretical analysis. Physical testbed implementations or prototype developments represent only 10.1%, and even fewer real-world deployments or clinical validations account for merely 2.2%. This observation points to a crucial research gap: there is a clear need for greater emphasis on experimental validation to bridge the gap between theoretical and simulated performance and the practical feasibility of energy-efficient WBAN solutions in dynamic, real-life environments.

9.3. Overarching Challenges and Future Directions

This subsection synthesizes the overall advantages and limitations of the discussed energy-efficient solutions (addressing RQ5) and proposes key future research directions, particularly by integrating the potential of emerging technologies.

9.3.1. Analyzing Key Trade-Offs in Energy-Efficient WBAN Design

While optimization at individual layers offers significant energy gains, a major future direction for WBANs lies in holistic, cross-layer optimization. The complex interplay between different protocol layers (MAC, network, application) and their impact on energy consumption necessitates solutions that coordinate decisions across the entire WBAN stack.

In this context, the design of a practical and robust WBAN is governed by multiple inherent trade-offs. Improving one performance metric, such as energy consumption, often comes at the expense of another, such as latency or reliability. Research must therefore continue to develop adaptive mechanisms that dynamically manage these trade-offs based on application requirements and environmental conditions. This section analyzes the most significant trade-offs and outlines strategies proposed in the literature to balance them.

• Energy Efficiency vs. Latency: The conflict between energy efficiency and latency is one of the most fundamental challenges. Mechanisms designed to minimize power consumption at the MAC layer, such as low power listening (LPL) and long duty cycles, directly increase the transmission delay [143]. To balance this, researchers have proposed priority-based scheduling schemes for critical medical data and adaptive duty-cycling protocols that adjust wake-up periods based on the type of data being transmitted [144].
• Energy Efficiency vs. Network Reliability: High reliability is important for life-critical medical data, yet achieving it requires energy-intensive mechanisms, such as retransmissions, forward error correction (FEC) codes, or maintaining redundant data paths [145]. A common approach to balancing this is the use of adaptive modulation and coding, which dynamically selects the most efficient scheme based on channel quality [146]. Additionally, reliability levels can be defined at the application level to tolerate lower reliability for non-critical data [147].
• Energy Harvesting vs. System Performance: The main challenge with energy harvesting is the intermittent and variable nature of the source, which directly impacts a WBAN’s ability to provide consistent performance [148]. To address this, research focuses on hybrid energy harvesting systems, using a supercapacitor or battery as a buffer [149]. Predictive power management schemes, often based on machine learning, can also proactively adapt the transmission schedule to match predicted energy budgets [150].
• Energy Efficiency vs. Security: Security protocols, essential for protecting sensitive data, are often computationally and energy-intensive [151]. To balance this trade-off, the literature explores lightweight cryptographic algorithms for resource-constrained devices [152] and offloading security-intensive tasks to the WBAN gateway [153]. Hierarchical security models can also be used to differentiate security complexity between nodes [154].

9.3.2. Need for Extensive Real-World Validation

As highlighted in the quantitative analysis (Section 9.2), a major overarching challenge and crucial future direction for energy-efficient WBANs is the lack of extensive real-world validation and deployment studies. While simulation provides valuable initial insights, the complex and dynamic nature of the human body and diverse operational environments necessitate more rigorous experimental validation. Future research must prioritize developing and testing energy-efficient protocols on physical prototypes and testbeds that mimic realistic physiological and environmental conditions.

9.3.3. Other Critical Challenges for WBANs

Despite the advancements reviewed, the literature reveals several technical aspects that remain insufficiently addressed. In particular, the following cross-layer challenges are crucial to the sustainable deployment of WBANs in real-world healthcare environments.

Minimizing Latency for Critical Data

At the MAC layer level, one of the most significant challenges is minimizing transmission delays, particularly for critical medical data, where latency can have serious consequences. In medical applications, latency should not exceed 125 ms, whereas in non-medical applications, it should remain below 250 ms. To address this issue, several mechanisms have been proposed in the literature, including priority-based data management and the direct transmission of urgent packets, which help ensure that essential information reaches medical personnel with minimal delay [155].

Enhancing Reliability

The reliability of data transmission in WBANs is a critical challenge, particularly when handling life-critical medical information. Packet loss, errors, or transmission delays can significantly compromise patient safety. To enhance reliability, several approaches can be employed, including the adoption of fault-tolerant transport mechanisms, the reduction of redundant data through periodic reporting systems, and the maintenance of optimal bit error rates (BER) using techniques such as adaptive modulation and channel coding at the physical layer [155].

Improving Security

Medical applications, particularly those based on WBANs, handle highly sensitive and personal data, which necessitates the implementation of core security principles, such as data integrity, confidentiality, encryption, and privacy. These measures are essential to ensure the protection of patient data throughout its lifecycle—from initial collection by sensors, through processing and transmission, to final storage on dedicated medical servers. Therefore, it is crucial to design secure communication protocols that can prevent data tampering, unauthorized access and eavesdropping [156].

Optimizing Throughput

To enhance the overall performance of a WBAN, it is essential to optimize throughput. This refers to the system’s data transmission capacity—that is, the amount of data a WBAN can reliably transfer over a communication link within a specific time frame [157]. Since WBANs frequently handle critical health data, maintaining optimal throughput is crucial for ensuring efficient and timely data transmission.

Handling Mobility

In WBANs, data transmission occurs in a dynamic environment—at the level of the human body—where the user is not expected to remain static. Physical and physiological movements can lead to signal fluctuations, multipath fading, or even transmission interruptions [158]. Therefore, designing communication protocols that are aware of these environmental constraints is essential to ensure greater stability and overall performance of WBAN systems.

Managing Interference

WBANs are particularly vulnerable to two major forms of interference: mutual interference, which occurs when neighboring WBANs operate on the same frequency band, and cross-technology interference, which arises when WBANs share spectrum with ubiquitous wireless systems, such as Wi-Fi, Bluetooth, or Zigbee, especially in the crowded 2.4 GHz ISM band [159]. To address the interference problem and its impact on system performance and energy consumption, various mitigation schemes have been proposed in the literature, broadly categorized into interference reduction and interference avoidance. Key power control approaches, such as game theory-based algorithms, such as proactive power update (PAPU) [160] and various power control games (PCGs) [161], dynamically adjust transmission power to optimize network utility and minimize energy drain, often incorporating adaptive pricing or reinforcement learning [162]. Complementing these, cross-layer interference management (CLIM) [163] adapts transmission rates and power based on the interference-limited communication range, allowing concurrent transmissions while mitigating their impact. For interference avoidance, MAC layer protocols are crucial; prominent examples include dynamic coexistence management (DCM) [164] and interference-aware channel switching (InterACS) [165], which utilize channel hopping and beacon management to resolve collisions. Dynamic resource Allocation (DRA) [166] and graph coloring schemes, such as random incomplete coloring (RIC) [167] assign orthogonal channels or time slots to avoid concurrent transmissions among coexisting WBASNs. Furthermore, asynchronous inter-network interference avoidance (AIIA) [168] enables flexible, asynchronous scheduling through information exchange to relocate conflicting transmission periods. Finally, the cognitive radio approach, exemplified by fast dynamic cognitive radio (FDCR) [169], empowers WBASNs to dynamically sense and utilize idle spectrum, thereby opportunistically avoiding occupied frequency bands and preventing interference. These diverse strategies collectively aim to enhance the reliability, throughput, and energy efficiency of WBASNs in dense and dynamic interference environments.

9.3.4. Emerging Technologies for Enhanced WBAN Sustainability

Beyond conventional approaches, the integration of cutting-edge technologies offers transformative potential for future energy-efficient WBANs, addressing the reviewer’s comment on expanding this focus.

• AI/Machine Learning (ML)-Driven Optimization: Artificial intelligence (AI) and machine learning (ML) are evolving into powerful tools for proactive and dynamic energy management in WBANs [170]. While their role in predictive energy management for EH (e.g., using deep learning to forecast energy availability from intermittent sources) is crucial, their potential extends significantly [150]. Reinforcement learning (RL), for instance, can enable WBANs to learn optimal policies for adaptive duty cycling [171], intelligent power control [172], and dynamic routing decisions in real-time, adjusting to network load, interference, and channel conditions to minimize energy waste [173]. The true strength of AI is seen in its potential to enable holistic, cross-layer energy optimization, where algorithms could analyze data from multiple layers to make coordinated decisions that optimize overall system energy consumption without compromising performance. This potential extends to intelligent battery management systems that could learn discharge/charge cycles to prolong battery life and facilitate energy sharing in multi-WBAN settings.
• Blockchain for Enhanced Security and Energy Implications: The security and confidentiality of sensitive health data are paramount in WBANs. Blockchain technology is emerging as a promising solution to enhance data integrity and privacy [174]. It offers a decentralized, immutable ledger for health data, ensuring authenticity, non-repudiation, and resistance to tampering from sensor to cloud. While traditional proof-of-work blockchains are highly energy-intensive, research is focusing on lighter-weight alternatives suitable for resource-constrained WBANs [175]. These include proof-of-stake (PoS) or consortium blockchains, or off-chain storage with on-chain verification, which significantly reduce computational and communication overhead. Paradoxically, by providing inherent trust and reducing the need for complex, energy-consuming encryption/decryption at every hop in certain scenarios, a well-designed lightweight blockchain could contribute to overall system efficiency by streamlining security measures.
• Integration with 5G/6G Networks and Edge Computing: The increasing demand for ubiquitous connectivity, ultra-low latency, and massive device support for WBAN applications necessitates their seamless integration with next-generation cellular networks, such as 5G and the upcoming 6G [176]. This integration brings significant benefits for energy management and data handling. Features such as ultra-reliable low-latency communication (URLLC) are vital for critical medical data, ensuring timely transmission with minimal retransmissions (thus saving energy). Edge computing (MEC) drastically reduces the amount of data transmitted to distant cloud servers by processing data closer to the WBAN devices, thereby significantly lowering communication energy consumption, reducing latency, and enabling real-time processing [177]. The intrinsic energy efficiency of 5G/6G infrastructure itself also indirectly benefits connected WBAN devices through more efficient communication channels.

9.3.5. Toward a Unified Energy-Efficient WBAN Framework

While this review has extensively analyzed individual energy-efficient mechanisms across MAC protocols, routing strategies, data processing techniques, and energy harvesting, the true potential for sustainable WBAN operation lies in their cohesive integration. The current literature often treats these solutions separately, leading to localized optimizations that may not translate to optimal system-wide performance. To address this, a conceptual framework that integrates these diverse mechanisms into a unified WBAN system, offers a holistic view for future design and implementation, as presented in Figure 14.

Physical Layer (PHY) and Energy Harvesting (EH) Module

This foundational layer is responsible for raw data transmission and, crucially, for the acquisition of energy. The energy harvesting module continuously captures ambient (e.g., solar, RF) and body-centric (e.g., biomechanical, biochemical) energy. It interfaces directly with the battery management system (BMS). This module provides the power supply, reducing reliance on the battery. Information on harvested power levels and battery state is critical for upper layers.

Data Link (MAC) Layer

Manages access to the wireless medium, organizing transmissions to avoid collisions and minimize idle listening. In this layer reside energy-efficient MAC protocols (e.g., TDMA, LPL, Scheduled Contention). This module optimizes node duty cycles and transmission schedules based on data queue status, channel conditions, and critically, feedback from the BMS and EH Module. An energy-aware MAC can, for instance, extend sleep periods if battery levels are low or increase activity if harvested energy is abundant.

Network Layer (NET)

This layer is responsible for routing data packets from source sensors to the destination (e.g., a hub or gateway). Energy-aware routing protocols (e.g., thermal-aware, cluster-based, QoS-driven) are implemented here. This module selects the most energy-efficient paths based on factors such as hop count, link quality, and crucially, the remaining energy of intermediate nodes. It can also adapt routes based on network congestion or user mobility.

Data Processing Layer

This layer is responsible for processing raw sensor data before transmission, including data aggregation, fusion, and compression. It can reside partially on the sensor node, a local WBAN hub, or an edge device. It mainly aims at reducing the volume of data that needs to be transmitted, thereby significantly lowering communication energy consumption. Intelligent algorithms can determine optimal compression ratios or aggregation windows based on data criticality and energy budget.

Intelligent Management and Optimization Layer (Cross-Layer AI/ML)

This crucial layer represents the true strength of a unified framework. It leverages artificial intelligence (AI) and machine learning (ML), particularly reinforcement learning, to orchestrate energy management decisions across all other layers. It analyzes real-time data from the EH module, BMS, MAC, network, and data processing layers. For example, an AI agent might observe low harvested energy, instruct the data processing layer to increase compression, and simultaneously signal the MAC layer to extend duty cycles, while the network layer selects longer but more energy-efficient routes. This layer also encompasses intelligent battery management systems (BMS) that learn optimal charge/discharge cycles to prolong battery life and facilitates energy-sharing policies in multi-WBAN settings.

Developing and validating such unified frameworks is a critical area for future research. This necessitates comprehensive experimental setups that can test the interplay of these integrated solutions in real-world scenarios, addressing the current gap in practical validation highlighted previously. This holistic approach promises to unlock the full potential of energy-efficient WBANs, paving the way for truly sustainable and reliable healthcare monitoring systems.

Finally, although this review has mainly focused on healthcare applications, it is important to note that the energy efficiency challenges of WBANs extend to other promising application domains, such as sports performance monitoring and military and security operations. These domains introduce unique requirements: sports and fitness emphasize the harvesting of kinetic energy from movement and long battery life, while the military field demands high reliability, security, and resilience in hostile environments [178]. The holistic design principles, trade-off management, and AI-based adaptation discussed here are directly applicable to these contexts. However, the balance between energy efficiency, latency, and security must be rethought to meet the specific needs of each application, thereby highlighting the need for ongoing research into more adaptable and generalizable energy management solutions.

10. Conclusions

Wireless body area networks (WBANs) are poised to revolutionize continuous health monitoring by leveraging wearable and implantable sensors. However, the inherent energy constraints of these nodes remain a primary obstacle, necessitating robust and efficient strategies to optimize power consumption while maintaining reliable communication.

This comprehensive review has synthesized the current landscape of energy-efficient mechanisms in WBANs, highlighting the key contributions of this work:

• A critical evaluation of energy-aware MAC protocols, which are crucial for minimizing idle listening and optimizing channel access.
• An in-depth analysis of energy-aware routing strategies, which contribute to energy conservation and improve data transmission reliability.
• A complementary review of energy-efficient data processing techniques (e.g., data aggregation and compression) to reduce communication-related energy costs.
• A systematic investigation of emerging energy harvesting techniques, which offer a promising solution to extend network lifespan.

Despite these advancements, the review identifies several critical challenges and future directions. The most significant challenge lies in balancing the inherent trade-offs between energy efficiency and other critical performance metrics, such as latency, reliability, and security. Furthermore, the integration of emerging technologies, such as AI and blockchain, presents a new frontier for dynamic energy management and secure data handling.

In conclusion, a holistic, cross-layer approach is essential to address these challenges. Future research should focus on developing adaptive, context-aware algorithms that intelligently balance energy efficiency with quality of service (QoS), ultimately paving the way for sustainable, autonomous, and highly reliable WBAN solutions.