Consensus on the Internet of Vehicles: A Systematic Literature Review

Abstract

The Internet of Vehicles (IoV) revolutionizes transportation by enabling real-time communication and data exchange among vehicles (V2V), infrastructure (V2I), and other entities (V2X). These capabilities are crucial for improving road safety and traffic efficiency. However, achieving reliable and secure consensus across network nodes remains a significant challenge. Consensus mechanisms are essential in IoV for ensuring agreement on the network’s state, enabling applications such as autonomous driving, traffic management, and emergency response. This paper presents a systematic review of IoV consensus mechanisms, examining 78 peer-reviewed publications from 2010 to June 2025 using the PRISMA framework. Our analysis highlights challenges, including scalability, latency, and energy efficiency and identifies trends such as the adoption of lightweight algorithms, edge computing, and AI-assisted techniques. Unlike previous reviews, this work introduces a structured comparative framework specifically designed for IoV environments, enabling a detailed evaluation of consensus mechanisms across key features such as latency, fault tolerance, communication overhead and scalability to identify their relative strengths and limitations.

1. Introduction

The field of smart transportation has recently seen the emergence of the Internet of vehicles (IoV) as a revolutionary innovation that is reshaping how automobiles communicate with each other, infrastructure, and their surroundings [1]. According to 3GPP, the Internet of Vehicles (IoV) is an advanced form of vehicular communication built upon Cellular Vehicle-to-Everything (C-V2X) technology, enabling vehicles, infrastructure, and network entities to exchange information over 4G and 5G networks to support cooperative driving, safety, and intelligent traffic management [1]. By enabling real-time communication between vehicles, infrastructure, and pedestrians, predictive analytics, accident reduction, IoV improves not only individual driving experiences but also the overall efficiency of transportation networks. This collaboration promises to address significant challenges in smart urban mobility such as road safety, traffic congestion, and environmental sustainability [2].

One fundamental technology of IoV is the concept of consensus. Consensus mechanisms ensure secure and trustworthy data sharing among vehicles (V2V), roadside units (V2I), and edge servers (V2N) [2]. They also enable critical applications such as real-time traffic coordination, autonomous vehicle decision-making, and secure data exchange in dynamic networks. By maintaining agreement across distributed nodes, consensus protocols enhance safety, reliability, and integrity in intelligent transportation systems [3]. Consensus mechanisms ensure a consistent and a synchronized understanding of the network’s state among vehicles, traffic signals, and other network participants in the dynamic IoV environment [4]. For example, in cooperative driving scenarios, vehicles must agree on actions like lane changes or merging, impacting efficiency and safety [5]. Similarly, in applications such as coordinated fleet operations, traffic management, and collision avoidance, the ability to reach real-time consensus is critical to improve traffic flow and prevent accidents [6].

However, the IoV ecosystem presents unique constraints that complicate the consensus process. The high mobility of vehicle nodes leads to rapidly changing network topologies making it challenging to maintain reliable communication links [7]. Intermittent connectivity, usually caused by physical impediments or changing network circumstances can disrupt the critical flow of information [8]. Furthermore, many IoV applications require low-latency consensus protocols that can adapt to the unpredictable and dynamic nature of vehicle networks [9]. The review was rooted in the following two objectives; First, to systematically evaluate and compare the performance of various consensus mechanisms in the dynamic and decentralized environments of the Internet of Vehicles (IoV). This includes assessing their efficiency, reliability, and adaptability to changing network conditions.

Secondly to identify and analyze specific adaptations required to enhance the scalability of consensus mechanisms in IoV systems, ensuring they can handle increasing numbers of vehicles and data transactions without compromising performance. This review also aimed to address the following key questions: How do various consensus mechanisms perform in the dynamic and decentralized environments of the Internet of Vehicles (IoV)? What specific adaptations are necessary to improve scalability and minimize latency in consensus mechanisms used within Internet of Vehicles (IoV) systems? By answering these questions, we seek to provide a framework for understanding the challenges and requirements of consensus mechanisms within the IoV context. Many existing surveys predominantly focus on the Internet of Things (IoT) and general blockchain applications, often neglecting the unique challenges posed by the Internet of Vehicles (IoV). Specifically, these reviews tend to neglect the dynamic constraints associated with high mobility, which significantly impact communication protocols because of the rapid network changes and consensus mechanisms in vehicular networks. Again, prior literature also fails to address interoperability challenges that arise when integrating diverse systems and technologies within the IoV ecosystem. By explicitly citing these gaps, this paper aims to emphasize the need for IoV-specific adaptations in consensus mechanisms and to highlight the critical areas that require further exploration to enhance the robustness and efficiency of IoV applications.

To address these challenges, this review aims to provide a comprehensive analysis of consensus mechanisms specifically developed for IoV. By reviewing peer reviewed literature published between 2010 and June 2025, the study reports that the current state of the knowledge identifies trends and highlights critical research gaps in the field [10]. The review investigates a wide range of consensus protocols from conventional algorithms to emerging approaches that utilize advances in artificial intelligence and machine learning. Furthermore, the paper explores the impacts of these consensus mechanisms on the broader transportation ecosystem, including their potential to improve energy efficiency, enhance vehicle-to-everything (V2X) communication, and contribute to the development of autonomous vehicles [11].

The main contribution of this paper is a comprehensive and systematic review of state-of-the-art consensus mechanisms specifically tailored for the Internet of Vehicles (IoV). The study systematically identifies and evaluates the key challenges, requirements, and performance characteristics that these mechanisms must address in dynamic vehicular environments. Building on this analysis, the paper introduces a structured comparative framework that enables the objective assessment of consensus mechanisms across critical IoV features such as scalability, latency, energy efficiency, and fault tolerance. Furthermore, the proposed comparative framework serves as a design guideline for developing novel consensus models by aligning algorithmic features with IoV-specific use cases and by revealing research gaps in existing approaches. This work provides essential insights to advance the design of secure, efficient, and adaptive IoV systems and acts as a valuable reference for researchers and practitioners in the evolving domain of intelligent transportation.

The remainder of the paper is organized as follows: Section 2 reviews related work. Section 3 explains the methodology. Section 4 presents the results and findings, including a comparative framework and its application to IoV scenarios. Section 5 discusses limitations and recommendations, and Section 6 concludes the paper.

2. Related Literature

This section reviews key related literature that forms the foundation for this systematic review, examining the works of other academics and identifying gaps in understanding consensus mechanisms in various contexts. This study emphasizes multiple aspects of consensus, including theoretical frameworks, techniques, and practical applications, as explored in existing research. Authors in [3] provide a comprehensive analysis of distributed ledger technology (DLT) and consensus mechanisms, examining 130 different consensus algorithms and 185 articles. In this study consensus mechanisms are categorized into three classes: Byzantine Fault Tolerance (BFT)-Compliant algorithms (which guarantee reliability even in the presence of malfunctioning nodes), Proof-Compliant mechanisms (such as Proof of Work and Proof of Stake), and Cross-Compliant hybrids that combine elements from several classes. This taxonomy facilitates understanding how decentralized systems to reach agreement [12].

However, the study leaves significant gaps, particularly regarding the adaptation of consensus mechanisms for the context of IoV. Although various consensus methods are explored, there is little discussion on how they might be tailored for IoV applications, which necessitates real-time decision-making and high security standards because of the vital feature of vehicle communication. Ref. [13] comprehensively reviews the use of blockchain technology to support a decentralized and private in Internet of Things (IoT). Of the eighteen use cases of blockchain technology identified only four are specifically designed for Internet of Things applications. Nevertheless, many of them focus on private and decentralized data management, aligning with the objectives of creating a secure privacy-preserving IoT ecosystem.

The study examines the factors affecting the integrity, anonymity, and scalability of blockchain in IoT. It discovers that whereas large blockchain systems, such as Bitcoin, provide high security, they suffer severe scaling issues that make them less suitable for IoT networks with vast numbers of devices. Furthermore, the study shows that most blockchain systems merely provide pseudonymity rather than full anonymity, highlighting the necessity for additional privacy-enhancing techniques to properly safeguard user identities.

Thereafter, the authors advise creating IoT applications with a layered architecture on top of existing secure and scalable blockchains to overcome these difficulties. This approach could mitigate the trade-offs between security, privacy, and performance in IoT deployments. The study also highlights the significance of further research to improve anonymity, achieve consensus in IoT, and adapt blockchain technologies to satisfy the unique needs of IoT networks. While the study provides valuable insights into consensus mechanisms in IoT, and potentially IoV as a subset of IoT, it does not explore the unique features and requirements of IoV, leaving a gap that this work seeks to fill. Ref. [14] presents an in-depth of blockchain technology in intelligent transportation systems (ITS) and IoV. It traces the development of blockchain from the pre-Bitcoin era to the Blockchain 2.0 phase, including the use of Hyper Ledger and Ethereum. The study groups blockchain-based IoV solutions into six categories: energy, transportation applications, security, communication and networking, data management, and payments and optimization. Moreover, it identifies open challenges and future research directions in BIoV, including reaching consensus in the Internet of Vehicles IoV and IoT, improving blockchain performance, integrating machine learning, leveraging big data, and incorporating 5G networks.

While the authors of [15] provide an analysis of blockchain technology in intelligent transportation systems (ITS) and the Internet of Vehicles (IoV), several gaps remain that highlight the necessity of our work. Firstly, their categorization of blockchain-based IoV (BIoV) solutions lacks a detailed exploration of consensus mechanisms, interoperability challenges between different blockchain platforms, which is crucial for seamless integration. Additionally, the authors do not address the real-world implementation barriers, such as regulatory hurdles and the scalability of proposed solutions. While this study focuses on consensus mechanisms in IoV, it is important to recognize that such mechanisms often depend on data collected through perception systems such as vehicle detection and tracking [16].

Recent advances in object detection using domain adaptation and deep learning [17,18] contribute to improving the accuracy of input data, which may enhance the reliability of consensus processes in collaborative environments. In parallel, privacy-preserving intelligence has also evolved through Federated Learning (FL), which enables decentralized model training without exposing sensitive vehicular data. FL-based Intrusion Detection Systems (IDS) have demonstrated strong potential in safeguarding vehicular networks by detecting anomalies and preventing cyberattacks while maintaining data confidentiality [7]. Such approaches integrate Generative Adversarial Networks (GANs) and hybrid deep learning models to improve detection accuracy and resilience against intrusions. Incorporating FL-inspired mechanisms within IoV consensus frameworks could enhance trust management, data integrity, and adaptive fault detection in dynamic vehicular environments [8]. This is particularly relevant in relation to data trust, validation, and cross-vehicle agreement, where the quality of perception directly impacts the decisions reached through consensus. Unlike these previous works, our review explicitly compares consensus mechanisms using IoV-specific features such as mobility, intermittent connectivity, and real-time decision-making.

Table 1. Comparison with Existing Surveys on IoV, IoT and Blockchain Consensus.

Authors	Focus Area	IoV Coverage	Limitations	Novelty and Distinction of This Work
[1]	Blockchain for IoV	Partial	Lacks consensus level comparison feature bench marking	Establishes a PRISMA-based IoV-specific framework that systematically evaluates consensus mechanisms beyond general blockchain discussion.
[2]	Secure IoV with blockchain	High	No structured performance evaluation or scalability analysis	Introduces a comparative PRISMA framework incorporating quantitative and feature-based performance metrics.
[3]	Blockchain applications in IoV	Medium	Focuses on application layer; lacks analytical treatment of consensus protocols	Integrates consensus-level benchmarking and analytical synthesis connecting design features with IoV requirements.
[17]	Blockchain-based IoV security survey	High	Descriptive narrative Without methodological rigor or comparative framework	Adds a systematic PRISMA-based selection and performance-oriented comparative analysis of IoV consensus algorithms.
[18]	Blockchain consensus for intelligent transportation systems	High	Limited coverage of quantitative benchmarking and scalability–latency trade-offs	Extends comparative analysis with numeric performance indicators and an adaptive analytical framework for IoV scenarios.
Our Work	Consensus in IoV	Full Coverage		Presents the PRISMA-based analytical comparison, IoV-specific taxonomy, and structured guidance for new consensus mechanism design.

below shows a comparative anatomy of the existing literature and the work presented in this systematic review.

3. Materials and Methods

To ensure methodological rigor and reproducibility, this study applied the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 framework during the literature selection and analysis process. PRISMA was used to structure the identification, screening, eligibility, and inclusion phases of the review, ensuring transparent reporting of how studies were selected and excluded [9]. Figure 1 illustrates the PRISMA-based study identification and selection process adopted for this systematic review. An initial database search retrieved 150 publications related to consensus mechanisms within the Internet of Vehicles (IoV) and Internet of Things (IoT) domains. After removing duplicate records, 120 studies remained for preliminary screening. During this stage, 20 papers were excluded due to insufficient technical content or lack of relevance to the study scope.

Subsequently, 100 full-text articles were assessed for eligibility based on clearly defined inclusion and exclusion criteria. Of these, 22 studies were removed because they did not explicitly address consensus mechanisms applicable to IoV or IoT systems. Finally, 78 studies met the inclusion requirements and were retained for in-depth qualitative and comparative analysis. This rigorous screening procedure ensured the inclusion of only peer-reviewed, high-quality publications that directly contribute to understanding and evaluating consensus mechanisms in IoV environments. This widely recognized approach enables a comprehensive and objective evaluation of the research topic. This section explains our search strategy, eligibility criteria and information sources, providing a roadmap for integrating the current state of knowledge in this rapidly evolving field.

3.1. Eligibility Criteria

The focus of this systematic review was on consensus mechanisms in the context of the IoV and the broader IoT. IoT papers were included because of their significant contributions to the connectivity and data exchange required for efficient IoV operations. These papers offer insights into novel applications and frameworks that can improve vehicle safety, transportation efficiency, and communication, making them essential for a comprehensive understanding of the IoV landscape. Additionally, since IoV is a subfield of IoT, potential solutions designed for IoT may also be applicable to IoV but may not have been explored specifically in an IoV context. Including IoT studies helps uncover such solutions that could be adapted to meet the unique challenges of IoV. Also, many IoT consensus algorithms, particularly lightweight ones designed for resource-constrained devices, can be effectively adapted for IoV environments where similar constraints exist.

For example, algorithms that prioritize low computational overhead and energy efficiency, such as Practical Byzantine Fault Tolerance (PBFT) or Lightweight Consensus Protocols, can be utilized in vehicular networks to ensure timely decision-making while accommodating the high mobility and dynamic conditions characteristic of IoV. Peer-reviewed journal or conference papers (2010–June 2025) written in English and focusing on consensus mechanisms in IoV or IoT were included. Publications were excluded during the full-text review stage for various reasons, including non-peer-reviewed sources, papers without technical discussion of consensus, or studies outside IoV context.

3.2. Information Sources

To ensure comprehensive coverage of relevant research, we searched four major electronic publishing databases: IEEE Xplore (IEEE, Piscataway, NJ, USA), ACM Digital Library (ACM, New York, NY, USA), Scopus (Elsevier B.V., Amsterdam, The Netherlands), and Science Direct (Elsevier B.V., Amsterdam, The Netherlands). These databases were chosen for their broad coverage of fields contributing to IoV research, such as computer science, engineering, and transportation technology. While Scopus and ScienceDirect offer multidisciplinary perspectives, IEEE Xplore and ACM Digital Library emphasize technical publications in computer science and engineering, making them crucial for this review.

3.3. Search Strategy

The systematic review employed a comprehensive search strategy using the following keywords combined with Boolean search terms: (“IoV” OR “Vehicle” OR “IoT” or “Internet of Things”) AND (“Consensus” OR “Blockchain” OR “block chain”). The base search was (“IoV” OR “IoT”) AND (“Consensus”), reflecting our core focus on consensus mechanisms in IoV and IoT. To account for variations in terminology, “Vehicle” and “Internet of Things” were added to ensure that studies using different terms are captured. Similarly, both “Blockchain” and “block chain” were included to accommodate the different ways blockchain is referenced in the literature. Blockchain has emerged as dominant applications of consensus in decentralized systems, making it a critical keyword for identifying consensus-related studies. This comprehensive keyword design allows for a thorough exploration of relevant literature across both IoV and IoT, ensuring no significant work is overlooked.

3.4. Papers from Each Database

Our search across four databases yielded a total of 150 relevant papers. The highest number of papers came from Scopus (50), followed by ACM Digital Library (40), IEEE Xplore (40), and ScienceDirect (20). This shows an increasing interest in IoV-specific research on consensus mechanisms, especially in more recent literature. The large number of articles highlights the rapid development and growing significance of consensus protocols in related sectors, offering a solid foundation for examining the present trends, challenges, and future directions in IoV consensus mechanisms.

4. Results

This section presents the different results derived from the analysis and synthesis of the studies included in this systematic review. The reviewed papers were examined to identify key patterns, performance trends, and design characteristics of consensus mechanisms applied within the Internet of Vehicles (IoV). The results are organized according to the main evaluation dimensions such as scalability, latency, fault tolerance, and energy efficiency to provide a clear understanding of how various consensus approaches address the unique challenges of IoV environments.

4.1. Consensus Mechanisms

Consensus mechanisms are protocols used in distributed systems, including but not limited to blockchain and distributed ledger technologies, to achieve agreement on a single data value among distributed processes or systems [10]. They guarantee that all participants in a network agree on the validity of transactions and the state of the system [11]. In IoV environments, consensus mechanisms are essential for maintaining trust, coordinating actions among vehicles, and ensuring the integrity of data shared across the network without relying on a central authority [12]. From our systematic literature review, a variety of consensus mechanisms were identified as key to maintaining the integrity of IoV networks, particularly in the context of blockchain and distributed ledger technologies. The following is a summary of the consensus mechanisms most frequently cited in the reviewed literature, each offering unique advantages and challenges relevant to IoV applications.

4.1.1. Proof of Work (PoW)

Proof of Work (PoW) is a consensus mechanism where miners compete to solve computational puzzles [13], with the first to find a valid solution earning the right to add a new block to the blockchain and receive a reward, typically in cryptocurrency [14]. The process involves finding a nonce that, when hashed with the block’s data, produces a hash value below a target difficulty threshold, ensuring that block creation requires significant computational effort and thus maintaining network integrity. Although PoW offers strong security and resistance to attacks—even tolerating up to 51% of nodes being compromised [3]—it is limited by low throughput (≈7 transactions per second) and high latency (≈10 min per block). Moreover, its substantial energy consumption, estimated at around 100 terawatt-hours annually, raises environmental and cost concerns [15].

Additionally, the competitive nature of mining can also lead to centralization, where a few large mining pools dominate the process, undermining decentralization and increasing vulnerability to collusion [3]. While PoW benefits from high-bandwidth communication technologies such as 5G and Dedicated Short-Range Communication (DSRC), its heavy computational and energy demands make it less suitable for Internet of Vehicles (IoV) environments, which require rapid, energy-efficient, and scalable consensus [18]. In vehicular networks, the high latency of PoW can disrupt time-critical tasks such as collision avoidance and cooperative driving. The frequent changes in network topology and intermittent connectivity further challenge the continuous synchronization required by this mechanism [15].

4.1.2. Proof of Stake (PoS)

In PoS, validators are chosen to create new blocks based on the number of coins they hold and are willing to “stake” as collateral. This method is more energy-efficient than PoW, as it does not require extensive computational resources [19]. PoS could be a better fit for IoV networks due to its lower energy demands(<10 W per node), but it also raises questions about decentralization and fairness in highly dynamic environments like vehicular networks [20].The reliance on wealth as a determining factor for validation can lead to centralization where wealthier participants dominate the network [21]. Additionally, the “nothing at stake” problem can arise, where validators may have little incentive to act honestly since they can vote on multiple forks without incurring costs [15].Though PoS improves energy efficiency, it may compromise security if a small number of validators control a significant portion of the stake. This centralization can lead to potential collusion, undermining the integrity of the network.

Proof of Stake (PoS) offers a much higher potential throughput, with some implementations aiming for over 100,000 TPS [21]. PoS also boasts lower latency, with finality achieved in just a few seconds and minimal energy consumption often less than 1% of what PoW requires [22]. It can tolerate up to one-third of nodes being faulty, making it a robust alternative. PoS mechanisms benefit from the low latency (<50 ms) and high throughput offered by 5G and LTE-V, which are critical for dynamic environments like IoV [23]. These protocols ensure that consensus can be reached quickly, enhancing the efficiency of the system. For IoV environments, PoS offers the advantage of energy efficiency and fast block generation, which are critical for edge-based vehicular networks where power and resources are constrained [24]. However, PoS assumes relatively stable validator participation, making it less suited for highly dynamic IoV networks where vehicles frequently join or leave. Despite this, hybrid models that combine PoS with permissioned or reputation-based approaches can provide a balance between scalability, decentralization, and security [25]. Therefore, PoS contributes to IoV by enabling sustainable blockchain integration while paving the way for adaptive and resource-aware consensus designs in vehicular systems.

4.1.3. Delegated Proof of Stake (DPoS)

DPoS improves PoS by allowing coin holders to vote for a small number of delegates who are responsible for validating transactions and maintaining the blockchain [26]. This system improves efficiency and scalability in comparison with PoS, making it potentially useful for IoV applications that require faster transaction processing and consensus among a limited number of trusted nodes [27]. The delegation process can lead to a concentration of power among a few delegates, which may result in governance issues and reduced accountability [28]. Additionally, the voting process may be susceptible to manipulation by large stakeholders who can exert undue influence over the selection of delegates. Although DPoS enhances transaction speeds, the potential for centralization may pose risks to network security, as a small group of delegates could collude to censor transactions or act against the interests of the broader community [29].

Delegated Proof of Stake (DPoS) can achieve around 4000 TPS and has low latency, with block times averaging half a second. Its energy consumption is low due to reduced computational needs [30]. DPoS relies on efficient communication among delegates, making 5G and LTE-V ideal choices for supporting its consensus process. Low latency is crucial in this model to ensure timely responses and effective governance [31].

4.1.4. Practical Byzantine Fault Tolerance (PBFT)

In PBFT, network nodes reach consensus through a voting mechanism that tolerates a limited number of faulty or malicious nodes [32]. This mechanism ensures that the network can still function correctly, even under adversarial conditions. PBFT’s focus on fault tolerance and fast decision-making aligns well with the demands of IoV, where reliability and real-time decision-making are crucial [33].Practical Byzantine Fault Tolerance (PBFT) can process over 1000 TPS in controlled environments, with low latency of about one to two seconds. Its energy consumption is also lower compared to PoW, and it can tolerate up to one-third of nodes being faulty [34]. PBFT requires a high level of communication between nodes, which can lead to scalability challenges as the number of nodes increases [35].

The protocol can become inefficient in larger networks due to the increased message complexity required for consensus [5]. Although PBFT provides strong safety guarantees, its performance may degrade in large-scale deployments, potentially hindering the responsiveness needed in IoV applications. The need for extensive communication can also introduce latency, affecting real-time decision-making [32]. PBFT thrives in environments with low latency and high throughput, making 5G and LTE-V ideal communication protocols [36]. These technologies enable quick consensus among nodes, which is crucial for real-time applications in the Internet of Vehicles.

4.1.5. Proof of Authority (PoA)

PoA relies on a limited number of pre-approved validators who are trusted to create new blocks [21]. This method is often used in private or consortium blockchains and emphasizes trust and reputation. In IoV contexts, PoA could be applicable where central authorities (e.g., government or regulatory bodies) are in charge of validation, though it sacrifices some decentralization. The main concern with PoA is the reliance on a small number of validators, which can create single points of failure [36]. If a validator is compromised, it can jeopardize the entire network’s integrity.

At the same time PoA can enhance transaction speeds due to fewer validators, the lack of decentralization may increase vulnerability to attacks and reduce the overall trust in the system, particularly in environments where transparency, scalability and high mobility is essential [3]. Proof of Authority (PoA) shines with very high throughput, exceeding 1000 TPS, and low block times of around three seconds. It also has minimal energy consumption(<5 W per node) but relies on the trustworthiness of a limited number of nodes, which can pose centralization risks [37]. PoA benefits from reliable communication protocols such as 5G and LTE-V, which facilitate efficient interactions among known authority nodes. This ensures that consensus can be achieved swiftly, enhancing the overall performance of IoV systems [38].

4.1.6. Tendermint

Tendermint is a Byzantine Fault Tolerant consensus algorithm that combines Proof of Stake mechanism with a gossip protocol [39]. It allows for fast transaction finality, making it well-suited for applications that require quick confirmation (≈1–2 s), such as vehicle-to-vehicle communication in IoV networks. Tendermint’s reliance on a fixed set of validators may lead to centralization risks similar to PoS and PoA. Additionally, the gossip protocol can become bandwidth-intensive, potentially straining network resources [40]. Safety and efficiency can as well be impacted as Tendermint offers rapid finality, its centralization risks could undermine the trustworthiness of the network [29]. The bandwidth demands may also limit scalability in scenarios with a high volume of transactions. Tendermint requires low latency (<50 ms) and high throughput (≥1 Gbps) for effective block finalization, making it well-suited for 5G and LTE-V networks. These communication protocols are essential for supporting the real-time demands of IoV applications.

4.1.7. Raft

Raft is a consensus algorithm designed for managing a replicated log through leader election and log replication [41]. It ensures that all nodes in a cluster agree on the same sequence of operations. Its simple design makes it suitable for distributed systems, including IoV, though its use may be limited to smaller networks or applications requiring predictable leader-based consensus [42]. Raft’s leader election process can introduce latency, especially in scenarios where the leader fails or needs to be replaced [43].

Additionally, the algorithm’s performance may degrade as the cluster size increases, leading to longer consensus times [44]. Raft is efficient for smaller networks; its scalability limitations may hinder its application in larger IoV environments [45]. The reliance on a single leader can also create vulnerabilities, as a compromised leader could manipulate the log. The Raft consensus algorithm is designed for low-latency environments (<10 ms), making 5G and LTE-V favorable communication options. These protocols support fast leader election and log replication, which are critical for maintaining consistency in distributed systems.

4.1.8. Directed Acyclic Graph (DAG)

DAG structures transactions in a graph rather than a traditional blockchain. Each transaction confirms previous ones, allowing for parallel transaction processing and increased capability [16]. DAG-based protocols, such as IOTA (Internet of Things Application), are particularly well-suited to IoT applications where high throughput (>10,000 TPS) and scalability are essential. DAG could also be beneficial for IoV systems [10]. The complexity of DAG structures can lead to challenges in ensuring consistency and validating transactions [41]. Additionally, the lack of a linear structure may complicate the implementation of certain security features. DAG’s ability to facilitate high throughput can significantly enhance efficiency in IoV applications.

However, the unique architecture may introduce new security vulnerabilities, and ensuring consensus in a decentralized environment can be more challenging than in traditional blockchain systems [46]. DAG-based consensus mechanisms thrive in high-throughput environments, with 5G and LTE-V providing the necessary communication capabilities [47]. Their ability to handle sporadic communication makes them particularly suitable for mobile networks.

4.2. Features of Consensus Mechanisms

Through our systematic literature review, we identified several key features of consensus mechanisms that are critical to the operation of distributed systems, particularly in IoV environments [29]. These features, which include safety, permissioning, energy efficiency, and fault tolerance, among others, emerged as central themes in the reviewed literature. Our analysis highlights how these features play a pivotal role in addressing the unique challenges of IoV systems, such as real-time decision-making, security, and scalability [32]. The following sections provide a detailed breakdown of each feature, explaining their relevance, how they are implemented across different consensus mechanisms, and their specific importance in IoV contexts.

4.2.1. Safety

Safety is a fundamental property of consensus mechanisms, ensuring that all honest network participants agree on the state of the system [44]. It guarantees the consistency and integrity of transactions, preventing fraudulent activities such as double-spending [15]. In the context of IoV, safety is specifically critical due to the real-time nature of vehicle-to-vehicle (V2V) communication and the high stakes of decision-making in traffic management or autonomous driving [28]. Incorrect or delayed consensus in these settings can lead to serious consequences, including accidents or system failures. Therefore, high-safety consensus mechanisms are essential to ensure secure and reliable data exchange in IoV networks [34].

Consensus protocols achieve safety through various mechanisms. In Proof of Work (PoW), such as in Bitcoin, safety is ensured by computational difficulty and the longest chain rule, which reduce the probability of conflicting transactions. Proof of Stake (PoS), on the other hand, achieves safety by requiring validators to stake their tokens as collateral, which can be forfeited if they act maliciously [48]. Byzantine Fault Tolerance (BFT) protocols provide safety by tolerating up to one- third of defective or malicious nodes, ensuring that the network can still reach consensus under adversarial conditions. Safety in consensus mechanisms is typically measured by their resilience against attacks and inconsistent behavior. This can be categorized into three levels: low, medium, and high.

• Low safety: Systems with low safety are vulnerable to attacks and inconsistent behaviors. They may fail to reach a consistent consensus and allow invalid transactions to be confirmed.
• Medium safety: These systems offer moderate protection against faulty transactions and network errors by medium safety systems [49].
• High safety: Mechanisms with high safety, such as BFT and PoS, provide strong assurance against invalid transactions and can withstand a variety of assaults, including Byzantine faults. They generally introduce heavy penalties to deter misbehaviours [50]. However, high safety does not equal perfect safety. Measuring the resilience of a protocol against specific attacks, such the well-known 51% attack in PoW, is a key metric for evaluating the strength of safety [51].

4.2.2. Permissioned/Permissionless

Consensus mechanisms can be categorized as either permissioned or permissionless, depending on how they regulate node participation in the network. In permissioned systems, participation is restricted to a predetermined group of nodes, providing greater security and control [52]. This is crucial for dealing with sensitive data, such as corporate IoV networks, where the authenticity of validators must be ensured [40]. Permissionless systems, on the other hand, allow any node to join or leave the network freely, making them more decentralized and open. In the IoV context, the choice between permissioned and permissionless chains impacts the balance between security, scalability, and openness [53].

4.2.3. Leader Based

Consensus mechanisms may rely on designating a single node (the leader) to propose new transactions or blocks, which helps streamline communication and decision-making among distributed nodes [54]. By reducing the number of nodes participating in decision-making, leader-based systems can significantly speed up the consensus process [55]. In IoV applications, where low-latency decision-making is critical, such as in vehicle-to-vehicle communication or real-time traffic management, leader-based systems provide an efficient approach to maintain consistent and timely data propagation [56].

This feature relies on the leader to coordinate the replication and validation of transactions across the network. For example, in Raft, nodes elect a leader through a lightweight voting process [57]. The elected leader then collects client requests, adds them to its log, and replicates these logs to follower nodes, ensuring that all valid transactions are consistently recorded [58].

4.2.4. Communication Based

Consensus mechanisms may achieve agreement through message exchanges between participating nodes [59]. This communication-based feature emphasizes the importance of efficient communication to maintain synchronization across the network. In IoV systems, where nodes (e.g., vehicles, infrastructure, and roadside units) must exchange data continuously, communication-based mechanisms are critical for ensuring accurate and timely consensus [60].

Communication-based consensus relies on message exchange protocols to validate and confirm transactions. For instance, in Practical Byzantine Fault Tolerance (PBFT), nodes communicate in multiple rounds to agree on the validity of a transaction [61]. The number of messages exchanged per transaction and the time taken to complete the communication rounds and finalize a transaction can be used to quantify the performance [62].

4.2.5. Communication Overhead

Communication overhead refers to the amount of data exchanged among nodes during the consensus process. It directly impacts the efficiency, scalability, and responsiveness of consensus mechanisms [63]. Communication overhead can be categorized into three levels:

• Low communication overhead: Few messages are exchanged between a limited number of participants. For example, DPoS minimizes message exchange by limiting participation to a small group of elected validators, reducing the number of messages exchanged [11].
• Medium communication overhead: More messages are exchanged among a moderate number of participants. For example, in PoS, validators need to share messages among a larger but still manageable group, balancing efficiency with decentralization [64].
• High communication overhead: A large number of messages are exchanged among all network participants, which can lead to bottlenecks and delays, particularly as the network size grows. For example, in PoW, each miner needs to broadcast their discoveries to the entire network, which leads to substantial communication overhead, particularly during periods of high activity [65].

4.2.6. Trust Based

Trust-based consensus mechanisms evaluate the reliability of nodes based on their past actions and reputation [66]. Nodes with a history of consistent and honest behaviors are assigned higher trust ratings, which influence their participation in transaction validation and consensus processes. In IoV environments, where nodes (e.g., vehicles, infrastructure, and roadside units) may frequently join and leave the network, trust-based mechanisms can help mitigate risks posed by untrusted or malicious nodes [67].

4.2.7. Fault Tolerance

Fault tolerance refers to the ability of a consensus mechanism to maintain system functionality and reach consensus even in the presence of malicious activity, node failures, or network disruptions [46]. This feature is essential for ensuring the reliability and availability of IoV systems, where vehicles and infrastructure nodes operate in dynamic, decentralized, and often unpredictable environments [68].

This feature can be measured by the percentage of participants that can be tolerated during consensus processes [69]. For example, Byzantine Fault Tolerance (BFT) protocols, such as Practical Byzantine Fault Tolerance (PBFT), allow consensus to be achieved as long as fewer than one-third of the nodes are defective or malicious [70].

4.2.8. Latency

Latency measures the time it takes for a consensus mechanism to validate and finalize a transaction or decision across the network. It is a critical component of system performance, directly influencing user experience and system responsiveness [71]. Latency can be categorized into three levels:

• Low latency: Transactions are confirmed in seconds or milliseconds, enabling real-time responsiveness. For example, Tendermint and Raft can achieve fast transaction confirmation through streamlined communication protocols and efficient leader-based processes [72].
• Medium latency: Transaction confirmation times range from a few seconds to several minutes [73]. For example, BFT-based solutions involve slightly longer transaction times due to the need for message exchanges among a group of participants.
• High latency: Confirmation times exceed several minutes. This may arise from intricate consensus procedures or the necessity for many validators to achieve to a consensus [29]. For example, PoW in Bitcoin often experience significant delays due to computationally intensive mining processes and long block intervals, leading to confirmation times of ten minutes or more.

4.2.9. Energy Efficiency

Energy efficiency measures the amount of energy consumed during the process of achieving consensus in a distributed system [74]. It directly impacts the environmental sustainability and operational costs of the system. High energy consumption can reduce the viability of IoV networks, especially in scenarios requiring long-term or large-scale operation. As IoV systems aim to contribute to sustainable smart cities, prioritizing energy-efficient solutions are essential.

Energy efficiency in consensus mechanisms can be categorized into three levels:

• Low efficiency: Systems that consume large amounts of energy relative to their output [75]. For example, Proof of Work (PoW) mechanisms, such as Bitcoin, rely on computationally intensive mining processes, which consume significant amounts of electricity to solve complex puzzles [37].
• Medium efficiency: Systems with moderate energy consumption balanced against their performance. These systems may employ certain cutting-edge techniques to lower energy consumption, but they still depend on somewhat inefficient processes.
• High efficiency: Systems optimized to minimize energy use without sacrificing security or functionality. Proof of Stake (PoS) mechanisms, such as Ethereum 2.0, select validators based on their stake in the network, drastically reducing energy consumption by eliminating the need for resource-intensive activities.

4.3. Comparison of Existing Consensus Mechanisms

After summarizing the consensus mechanisms and their features based on existing literature, we now present a comprehensive comparison of these mechanisms. Each mechanism comes with its unique set of characteristics, advantages, and drawbacks, influencing critical factors such as transaction speed, energy efficiency, scalability, and decentralization. This comparison aims to provide a deeper understanding of the trade-offs involved in selecting a consensus mechanism, particularly for IoV applications. The evaluation framework considers energy efficiency, latency, safety, fault tolerance, communication overhead, scalability, Leader based, permissioned or permissionenless and IoV/IoT compatibility, which collectively determine suitability for vehicular applications.

Table A1. Comparison of Consensus Mechanisms.

Protocol	TPS (Transactions/sec)	Average Latency (ms)	Energy Overhead (J/Tx)	Security Level	Fault Tolerance (%)	Permission Type	Leader-Based	Communication Rounds	Trust Mechanism	IoT Suitability	IoV Suitability	Key References
PoW	7–15	6000–10000	100–150	Very High	<50% nodes	Permissionless	No	1	Cryptographic	Low	Moderate (niche use)	[9,47,48,49,50]
PoS	50–100	1000–3000	5–15	High	<50% stakes	Both	No	Variable	Stake-based	High	High	[35,49,51,52]
DPoS	1000–3000	200–800	2–8	High	<50% validators	Both	Yes	2	Delegate voting	High	High	[9,35,51,52,53]
PoA	2000–5000	50–200	1–5	High	<33% validators	Both	Yes	1	Authority nodes	High	Very High	[1,9,18,44]
PBFT	1000–1500	100–300	3–10	High	<33% nodes	Permissioned	Yes	3	Node agreement	High	Very High	[7,23,30,39,47,55]
R-PBFT	1800–2000	80–200	2–6	High	<33% nodes	Permissioned	Yes	3	Adaptive voting	High	Very High	[36,38]
SG-PBFT	2000–2500	70–150	2–5	High	<33% nodes	Permissioned	Yes	3	Subgroup-based	High	Very High	[52,56]
CPBFT	2200–2500	70–120	2–4	High	<33% nodes	Permissioned	Yes	3	Cluster-based	High	Very High	[14,35,36,37]
G-PBFT	2500–3000	60–100	2–4	High	<33% nodes	Permissioned	Yes	3	Gossip-based	High	Very High	[42,43,44]
DBFT	2200–2800	80–120	2–5	High	<33% nodes	Permissioned	Yes	3	Distributed ledger	High	Very High	[58,59]
Paxos	50–100	2000–5000	10–20	Moderate	<50% nodes	Permissioned	No	2–3	Majority vote	Moderate	Moderate	[44,56]
Raft	300–800	500–1500	5–10	High	<50% nodes	Permissioned	Yes	2	Leader election	High	High	[27,33,36,60,61]
Tendermint	1000–1500	100–300	3–8	High	<33% nodes	Both	Yes	2–3	BFT-based	High	Very High	[55,62]
DAG (e.g., IOTA, Tangle)	3000–5000	50–150	1–3	Medium	Dynamic	Permissionless	No	Variable	Transaction graph	High	Moderate	[3,7,63]

(Appendix A.1) is the main contribution of this paper, offering a structured comparison of several prominent consensus mechanisms, including Proof of Work (PoW), Proof of Stake (PoS), Delegated Proof of Stake (DPoS), and others. The table organizes consensus mechanisms as rows and features as columns, providing a clear and concise overview of how each mechanism performs across the identified features such as safety, latency, fault tolerance, energy efficiency, and communication overhead.

This analysis not only highlights the strengths of specific mechanisms but also identifies their limitations, offering insights into their suitability for IoV scenarios. In the subsequent discussion, we elaborate on the data presented in the table, emphasizing the trade-offs and identifying mechanisms that excel in particular areas while pointing out potential areas for improvement or future research. Table A1 presents an evaluation of the different consensus mechanisms based on the aspects such as energy efficiency, communication overhead, permissioned/permissionless leader-based, trust- based, fault tolerance, latency and IoT/IoV compatibility.

4.4. Applying the Comparison for Consensus Algorithm Design

Building on the results of Table A1, this section demonstrates how the comparative analysis presented in the previous section can be utilized to guide the design and selection of consensus algorithms for specific scenarios. By analyzing the strengths, weaknesses, and trade-offs of various consensus mechanisms across key features, practitioners can make informed decisions about which mechanisms are best suited for their unique requirements. This section uses practical examples to illustrate how the comparison table can serve as a decision-making tool, helping researchers and developers tailor consensus algorithms to meet the demands of diverse IoV applications, such as real-time traffic management, autonomous vehicle coordination, or resource-constrained networks. Through these examples, we highlight the flexibility and utility of the comparative framework in addressing real-world challenges in IoV systems.

4.5. Application of Proof of Work (PoW) in IoV Scenarios

Although PoW has demonstrated robust performance in cryptocurrencies, its high computational demands, energy consumption, and latency make it impractical for many resource-constrained environments such as IoT and IoV systems [30]. However, PoW may have niche applications in IoV scenarios where security is paramount, and the frequency of transactions is low, such as the transfer of vehicle ownership or critical software upgrades. PoW remains the most secure yet computationally intensive consensus algorithm. While unsuitable for large-scale, high-frequency IoV operations due to energy and latency costs, it can still serve low-frequency, high-security use cases for example, vehicle ownership transfers or firmware validation. In such scenarios, security and fault tolerance take precedence over speed and efficiency [48].

According to the comparative framework, PoW’s strength in ensuring immutability and tamper-proof validation makes it appropriate where transaction frequency is limited and the cost of computation is acceptable.

• Specifically, to address the critical requirements for high safety, permissionless participation, low trust priority, and high fault tolerance, PoW emerges as the best candidate for this application scenario according to Table A1.

4.6. Application of Practical Byzantine Fault Tolerance (PBFT) in IoV Scenarios

PBFT addresses the Byzantine Generals’ Problem in distributed systems [76]. It operates in rounds, where a leader node proposes a value, and other nodes vote through multiple stages to reach consensus. PBFT is resistant to various network issues, tolerating up to one-third of nodes being malicious or defective [77]. With its low latency and energy efficiency, PBFT is particularly well-suited for IoV scenarios, especially those requiring real-time decision-making. For instance, PBFT can be effectively applied in vehicle-to-infrastructure (V2I) connections where a group of reliable, well-known roadside units serves as consensus nodes [78].In the comparative framework, PBFT offers an excellent balance of low latency, high safety, and strong fault tolerance, making it well-suited for real-time V2I coordination or traffic-signal optimization where participating nodes (e.g., roadside units) are trusted and limited in number [8]. Although its communication overhead restricts scalability, this limitation is manageable in permissioned vehicular networks with predefined participants.

Specifically, to address the critical requirements for high energy efficiency, high safety, permissioned participation, low latency, and high fault tolerance, PBFT emerges as a strong candidate for real-time traffic management in IoV systems according to Table A1. Byzantine Fault Tolerance (PBFT) works in medium to low-latency Internet of Vehicles (IoV) scenarios. Out of the 78 studies analyzed, 60% identified PBFT as suitable for medium to low-latency applications like IoV. Its ability to operate effectively in permissioned networks with immediate transaction finality aligns with the scenario’s requirements. While its communication overhead poses a scalability challenge, this is mitigated in networks with a limited number of reliable nodes. Non-surprisingly, PBFT and its variants are the most widely studied in IoT and IoV.

4.7. Designing a New Consensus Mechanism for IoV Scenarios

The existing consensus mechanisms, such as PBFT, PoW, and PoS, excel in specific domains, there are IoV scenarios where no single mechanism can adequately address all requirements. In this case, Table A1 can also be used as a guide for developing new consensus protocols. For instance, consider a large-scale, dynamic IoV system for inter-city traffic coordination. This scenario involves a vast number of vehicles, infrastructure nodes, and sensors across multiple jurisdictions, requiring a consensus mechanism that balances scalability, energy efficiency, fault tolerance, and low latency. No existing mechanism fully satisfies these combined demands. Guided by the comparative framework, a hybrid consensus model can merge complementary strengths of multiple mechanisms.

In the Internet of Vehicles (IoV), a hybrid consensus mechanism integrating Proof of Authority (PoA) and Practical Byzantine Fault Tolerance (PBFT) can ensures an efficient, secure, and fault-tolerant consensus mechanism, tailored to the dynamic and safety-critical nature of vehicular networks. PoA is a deterministic consensus mechanism that selects a leader node from a predefined set of trusted authorities, which minimizes computational overhead, enabling rapid leader selection and enhancing scalability for real-time applications. Once the leader is selected, the system transitions to the PBFT consensus protocol. PBFT is designed to tolerate Byzantine faults, where nodes may act maliciously or fail unpredictably. This combination helps overcome weaknesses in both systems; Proof of Authority (PoA) depends on trusted nodes, but it becomes stronger with Practical Byzantine Fault Tolerance (PBFT), which is good at handling faults. At the same time, PBFT’s heavy computing needs are balanced out by PoA’s quick leader selection. Together, they create a consensus mechanism that is both scalable and reliable, making it perfect for Internet of Vehicles (IoV) settings. Additionally, the system incorporates a View Change mechanism, activated if the leader fails to achieve consensus (e.g., insufficient votes). This feature dynamically replaces the leader, ensuring continuous operation and resilience against faults or attacks, such as attempts to disrupt emergency vehicle communications. The flow chart in Figure 2 shows a combined flow of activities when we combine the strengths of PoA and PBFT for a hybrid consensus mechanism.

No existing consensus mechanisms can meet all the requirements. However, Table A1, which shows the features of existing mechanisms, can guide the design of a new consensus mechanism. We can easily first find the most relevant mechanism, such as PBFT, and then propose extensions to address the missing features. With the help of Table A1, we can instantly propose the following solutions:

• Hybrid permission model: Incorporate PBFT for permissioned, localized traffic control nodes (e.g., within a city) and a PoA-like mechanism for vehicles interacting with the system in a permissionless manner.
• Optimized communication overhead: To improve scalability, reduce PBFT’s communication overhead by adopting DPoS-style elected validators for inter-city coordination, limiting the number of participating nodes in each consensus round.
• Fault tolerance enhancement with trust metrics: Integrate trust-based features to dynamically evaluate participants based on node reliability, uptime, and performance, thus enhancing PBFT’s fault tolerance by detecting and isolating malicious nodes faster.

By leveraging the comparison table, potential solutions for a new hybrid consensus mechanism can be quickly identified to meet the unique demands of inter-city traffic coordination. However, further in-depth research and analysis are required to validate and refine these solutions. The purpose of this paper is not to design such mechanisms but to demonstrate how our comparison framework (as shown in Table A1) can assist in guiding the development of novel consensus mechanisms tailored to complex IoV scenarios.

5. Discussion

The evolution of consensus protocols in IoV reflects a shift from adaptations of blockchain mechanisms to solutions specifically tailored to the unique requirements of vehicular networks. Recent research (2020–2025) highlights a growing focus on low-latency, highly scalable protocols tailored to the dynamic and decentralized nature of IoV systems. Energy efficiency has emerged as an important factor, with Proof-of-Stake variants gaining popularity due to their lower computational demands. Security remains a priority, as evidenced by the development of hybrid consensus models that balance robust protection with performance optimization. Additionally, there is a noticeable trend toward the integration of machine learning techniques, enabling adaptive consensus mechanisms that dynamically optimize performance based on real- time network conditions.

Despite notable advancements in consensus protocols, significant limitations remain in the current body of research. One key challenge is the lack of large-scale, real-world implementations. Most studies are confined to theoretical models or simulations, with minimal testing conducted in urban or real-world IoV environments. This gap restricts our understanding of how consensus mechanisms perform under conditions such as varying traffic patterns, environmental factors, and network disruptions. Another major issue is the lack of standardized evaluation methods across studies. Researchers employ a variety of metrics and methodologies, making it difficult to directly compare the performance of different consensus mechanisms. This lack of standardization creates barriers to identifying universally superior approaches and complicates the synthesis of findings. Establishing a unified evaluation framework is essential to advance the field and facilitate the adoption of best practices.

Our work contributes to bridging some of these gaps by providing a comprehensive review and systematic comparison of existing consensus mechanisms. The comparison framework we developed offers a structured approach to evaluate consensus mechanisms across key features such as safety, energy efficiency, fault tolerance, and latency. Moreover, we demonstrate how this framework can guide the design of novel consensus mechanisms tailored to IoV-specific scenarios; mobility, latency, scalability and fault tolerance. This framework not only synthesizes existing work but also provides a roadmap for future IoV consensus design. Figure 3 illustrates the evolution of consensus research in IoV from 2010 to June 2025.

Research Roadmap for IoV Consensus

Building on the identified gaps, this roadmap outlines strategic priorities for advancing IoV consensus mechanisms. In the short term, research should focus on developing ultra-low-latency protocols (<50 ms) for real-time vehicular coordination and establishing quantifiable trust metrics to secure highly dynamic networks. Mid-term efforts should emphasize hybrid and AI-driven adaptive consensus models that balance scalability, safety, and energy efficiency. In the long term, integration with 6G communication, edge intelligence, and regulatory safety standards such as ISO 26262 (Functional Safety for Road Vehicles, ISO, Geneva, Switzerland) [79] and UNECE WP.29 (World Forum for Vehicle Regulations, United Nations Economic Commission for Europe, Geneva, Switzerland) [80] will be essential to achieve deployable, secure, and interoperable IoV consensus ecosystems.

6. Conclusions

This study presented a systematic and analytical review of consensus mechanisms in the Internet of Vehicles, identifying current gaps, performance trade-offs, and emerging directions that can guide the design of next-generation vehicular consensus mechanisms. Consensus mechanisms for the Internet of Vehicles (IoV) have made significant progress in recent years, but notable challenges still remain. Top among these are filling the gap between theoretical models and practical deployment, establishing standardized evaluation methods, and addressing IoV-specific requirements such as high mobility, intermittent connectivity, and heterogeneous communication environments. This paper contributes by providing a systematic review and comparative analysis of existing consensus mechanisms, alongside a framework to evaluate and guide the development of novel solutions. By demonstrating how this framework can be applied to IoV-specific scenarios, the study offers a practical tool for researchers and practitioners aiming to design more robust, efficient, and secure consensus protocols. Ultimately, advancing IoV consensus is not just a technical challenge but a critical enabler for safer, smarter, and more resilient transportation systems.

Although numerous consensus mechanisms have been proposed for the Internet of Vehicles (IoV), most remain limited to theoretical or simulation-based studies, lacking real-world validation under dynamic mobility and heterogeneous communication conditions. There is also no standardized framework for evaluating consensus performance, making it difficult to compare scalability, latency, and energy efficiency across studies. Furthermore, issues such as trust management, privacy preservation, and adaptive response to network fluctuations remain insufficiently explored.

Future research should focus on developing adaptive, AI-driven, and energy-efficient consensus protocols that can dynamically adjust to network conditions, mobility patterns, low latency and high-reliability solutions capable of achieving (<50 ms) a response time for safety-critical tasks such as autonomous merging or collision avoidance. Integrating edge and fog computing can reduce latency, while 5G and upcoming 6G technologies offer opportunities for ultra-reliable low-latency communication. Emerging research also can explore the integration of Federated Learning (FL) with IoV consensus frameworks to enhance privacy-preserving decision-making and anomaly detection. FL can enable distributed model training across vehicles without requiring raw data exchange, thereby improving both data confidentiality and network trust. Combining FL with blockchain-based consensus could lead to intelligent, adaptive, and secure IoV ecosystems capable of resisting cyberattacks and ensuring reliable real-time coordination. Establishing standardized benchmarks and real-world IoV testbeds will also be critical for validating performance and accelerating the deployment of secure, scalable, and intelligent vehicular consensus systems.