Immersive, Secure, and Collaborative Air Quality Monitoring

Abstract

Air pollution poses a serious threat to both public health and the environment, contributing to millions of premature deaths worldwide each year. The integration of augmented reality (AR), blockchain, and the Internet of Things (IoT) technologies can provide a transformative approach to collaborative air quality monitoring (AQM), enabling real-time, transparent, and intuitive access to environmental data for community awareness, behavioural change, informed decision-making, and proactive responses to pollution challenges. This article presents a unified vision of the key elements and technologies to consider when designing such AQM systems, allowing dynamic and user-friendly immersive air quality data visualization interfaces, secure and trusted data storage, fine-grained data collection through crowdsourcing, and active community learning and participation. It serves as a conceptual basis for any design and implementation of such systems.

1. Introduction

Public health and environmental integrity can be seriously affected by air pollution, being responsible for millions of premature deaths each year, especially in low- and middle-income countries [1]. Beyond human concerns, air pollution globally impacts ecosystems and contributes to climate change through greenhouse gas emissions.

Therefore, air quality must absolutely be improved and preserved in order to promote healthier communities and the long-term preservation of environments, which requires environmental awareness and proactive action by governments, industry, and citizens. Air quality monitoring (AQM) is essential to support informed decision-making in this context. However, challenges still need to be addressed in several areas to realize its full potential, such as increasing transparency, reliability, accuracy, and accountability concerning the collection, storage, processing, and access to pollution data. For example, data sharing systems are often vulnerable to attacks and often lack the required trust, transparency, and interoperability with other data systems [2,3,4]. Furthermore, approaches to AQM, such as those based on the use of stationary pollution monitoring systems, often lack the granularity and flexibility required for fine-grained data collection, leaving gaps in the understanding of air quality within some areas and neighbourhoods [2].

Based on the challenges and limitations mentioned, it can be stated that the development of novel AQM solutions that are decentralized, accessible, collaborative, and interactive remains a relevant open issue that needs to be addressed. The goal is to stimulate participation in reliable and widespread monitoring efforts, even in resource-constrained areas, and to effectively use the data.

This article explores how augmented reality (AR), blockchain, and the Internet of Things (IoT) can be integrated into the design of such novel AQM solutions, particularly those that empower individuals to both contribute to and make use of air quality data. The paradigm of these new AQM systems should shift from merely informing to actively increasing awareness, promoting behavioural change, and raising engagement through intuitive, immersive, and geolocated visualizations. Rather than presenting a design specification or an implementation study, this article offers a conceptual and theoretical exploration that addresses the current fragmentation of air quality monitoring systems, in which diverse technologies are often considered in isolation. It proposes a unified vision of these technologies and lays the foundation for future interdisciplinary research and system development in immersive environmental monitoring enabled by community-driven trust via blockchain verification, immersive engagement through AR overlays mapped to geolocated data, and modular extensibility for combining diverse IoT sensing capabilities. To the best of our knowledge, after a full review of the relevant literature, no existing work has proposed integrating these three dimensions into a single AQM model.

The remainder of this article is organized as follows: Section 2 reviews related work, Section 3 presents the conceptual approach for the targeted type of collaborative AQM system, and, finally, Section 4 draws some conclusions.

2. Related Work

AQM is a critical component of environmental surveillance and public health protection. Over the past two decades, there has been a notable shift from traditional stationary monitoring stations to more distributed and scalable approaches, enabled by advances in sensor miniaturization, wireless communication, and data analytics.

Building on the shift toward more distributed and technology-enabled AQM systems, the combination of AR, blockchain, and the Internet of Things (IoT) can be considered to develop new AQM systems that successfully address the challenges and limitations identified in the previous section. Blockchain supports secure and transparent data storage and sharing, AR provides engaging and intuitive ways to visualize the context data, and the IoT enables the deployment of low-cost sensors for fine-grained collection of air quality data with the participation of different stakeholders. This section presents a review of concepts, research, and existing solutions, following a bottom-up approach, highlighting key contributions and identifying gaps that the current work aims to address by combining the aforementioned technologies.

2.1. Air Quality Monitoring

According to [5], AQM approaches can be divided into sensor-based techniques and image-based techniques. The first type consists of sensors being deployed to measure various pollutants, and involves actions such as collecting and transmitting data, calibrating the sensors for accuracy, processing and analyzing the data, and providing real-time monitoring and visualization. Diverse types of sensors can be used and combined in AQM scenarios. Sensor-based methods are considered reliable for continuous monitoring of air quality parameters. However, they often lack flexibility and real-time monitoring capabilities and saw peak popularity between 2018 and 2020 [5]. Image-based AQM involves capturing images, processing, and analyzing them using advanced algorithms to estimate air quality parameters based on visual indicators (e.g., haze detection and visibility reduction). Image-based techniques are innovative and offer wide-area coverage at low hardware cost. However, these techniques are limited by the size and diversity of datasets. They are primarily effective during daylight hours and are limited by weather conditions and lighting variability. Additionally, they require complex model training. In [5], the authors noted a rising interest in image-based techniques starting around 2015, with significant developments in deep learning applications noted around 2018. To address the limitations of both monitoring techniques, they also proposed a hybrid approach that integrates sensor and image-based methods, aiming to enhance the overall effectiveness of AQM. Sensor-based systems are better suited for near-real-time participatory sensing, while image-based systems provide additional coverage for regional-scale monitoring. Depending on the use case, both could be integrated into flexible, collaborative, and trusted AQM systems.

A distinction between stationary and portable/mobile approaches for AQM can also be made [6]. Currently, the environmental data is mostly collected by meteorological government stations, for data such as temperature, humidity, atmospheric pressure, precipitation, wind direction and speed, and solar radiation, and fixed AQM stations for air quality parameters. However, according to some authors, such as in [3], traditional stationary pollution monitoring systems, while accurate, are not the most appropriate approach for fine-grained data collection due to their excessive installation and maintenance costs and their lack of flexibility. Although the data is collected automatically several times a day, it covers only the specific locations where these devices are located. To address this, a growing body of research has explored the deployment of portable/mobile sensor networks, characterized by their low cost and small size, allowing flexible measurements and high spatial and temporal resolution.

Several studies highlight a trend towards low-cost sensor technologies in air quality monitoring. In [7], the authors provided one of the first comprehensive evaluations of how low-cost sensors could complement regulatory monitoring networks, particularly in urban areas where fine-scale spatial resolution is necessary to capture pollution hotspots. The authors in [8] report that these sensors, costing up to three orders of magnitude less than standard instruments, have enabled broader participation in air quality monitoring. The same study notes that the performance of these sensors can be satisfactory under specific conditions, although there is a lack of consensus regarding recommended end-use and associated minimal performance targets. In [3], a mobile monitoring system uses air quality sensors mounted on moving vehicles to collect data across space and time. Regarding the geographical and environmental context of AQM, monitoring systems are deployed in a variety of settings, including urban and rural areas, indoor and outdoor environments, and specific locations such as cities or regions [8].

It is important to note, however, that while low-cost sensors can provide relatively accurate measurements of particulate matter (e.g., PM_2.5), they are significantly less reliable for detecting and quantifying gaseous pollutants such as NO₂, CO, and O₃. These sensors often exhibit cross-sensitivity, drift, and lack of robustness under variable environmental conditions. As such, their data should be interpreted with caution and should ideally be validated against professional-grade reference instruments. Their main value lies in their ability to provide spatial and temporal granularity rather than strict accuracy.

AQM system architectures span a wide spectrum, from large-scale urban networks to compact exposure devices. Many studies emphasize the development of flexible, mobile, and intelligent systems. Key architectural trends include the integration of IoT technologies, with [9] reporting that 53% of indoor AQM systems incorporate IoT; the use of wireless sensor networks (WSN) [10]; the adoption of cloud-based platforms, with [11] demonstrating the benefits of using open-source cloud services for data processing and analytics; reducing the processing load on sensor nodes; the application of artificial intelligence (AI) and machine learning techniques, with [12] presenting the effectiveness of AI and geospatial analysis in predicting air quality parameters with high accuracy and with [13] presenting an IoT-enabled system using AI techniques for air pollution monitoring; and the deployment of mobile and non-stationary sensing devices, with [14] discussing the development of non-stationary instruments that are more sensitive, selective, precise, and accurate and with [15] describing a portable device integrating nanotechnology-based sensors with the Global Positioning System (GPS) for mobile monitoring. The calibration and validation techniques necessary to improve the reliability of low-cost AQM systems are particularly important. In [16], the authors evaluated various sensor models under field conditions and highlighted the importance of local calibration strategies.

Participatory AQM has gained momentum with the rise in low-cost, mobile, and open-source technologies. Projects such as AirSensEUR [17] have demonstrated the potential of open-source platforms and mobile sensing units to enhance monitoring coverage while engaging communities in data collection efforts. Complementing this, several initiatives emphasize the role of citizen science in democratizing AQM. For example, Sensor Community project [18] (formerly Luftdaten.info) has facilitated the global deployment of thousands of “Do-It-Yourself” sensor nodes by individuals, and the CitiSense initiative [19] explored the use of smartphones paired with portable sensors to provide users with personalized air quality feedback. Together, these efforts display how participatory sensing can contribute to more granular, transparent, and inclusive environmental monitoring systems.

Beyond the sensors themselves, significant attention has been given to the integration of AQM data with Geographic Information Systems (GIS) and models for spatial interpolation, as seen in [20]. More recent efforts, such as those described in [21], have focused on real-time data assimilation and visualization, incorporating contextual variables such as meteorological conditions, traffic patterns, and urban topology.

The use of IoT in AQM and management for smart cities is growing, offering new opportunities. The sensors can be strategically placed to continuously collect real-time data on relevant pollutants and other important parameters for assessing air quality [3,22,23,24]. For instance, [11] demonstrates the effectiveness of integrating commercial gas sensors with the Raspberry Pi microcontroller and cloud services for real-time monitoring and alerting. Similarly, [10] discusses the integration of low-cost sensors in low-power systems for WSN for urban AQM, offering high spatial and temporal resolution. However, the lack of standardized performance evaluation methods across studies makes it challenging to compare the effectiveness of different sensor technologies directly. Most AQM studies do not provide detailed quantitative performance data, highlighting a need for more rigorous testing and validation of these emerging technologies. Some of the challenges of using IoT sensors that need further research are related to limited resources (e.g., battery life), potential attacks, and the need to evaluate their accuracy by comparing measurements with stationary reference stations [2]. Providing an appropriate communication environment is another critical issue that needs to be addressed [25].

All these aspects highlight the multidimensional nature of AQM research, encompassing sensor design, network deployment, data processing, public engagement, and policy impact.

2.2. Communication Technologies for IoT Sensors

Battery-powered IoT sensor devices mostly transmit (and/or receive) small amounts of data at predefined intervals and have very low power consumption and long-range requirements. Therefore, using them to collect environmental data is not the typical use case of traditional communication technologies, such as Wi-Fi and cellular networks to name a few, which give priority to transmission at high data rates at the expense of power consumption and, in the case of cellular networks, require a contract with a network operator. Hence, these technologies are not ideal for these new smart devices, although they are integrated into many solutions proposed in the literature. For instance, [26] proposes an IoT-based environmental pollution monitoring system in which sensing units utilize Wi-Fi modules to transmit data to a backend server. Similarly, [27] presents an air quality sensing and reporting system that also employs Wi-Fi for connectivity. In [28], which describes an IoT solution for monitoring both air quality and UV radiation, the sensing device connects to a laptop or PC via Wi-Fi for data transmission.

Several low-power, wide-area network (LPWAN) technologies have emerged with the purpose of addressing the requirements of new IoT devices, i.e., long-range coverage (several kilometres) with low bandwidth and low power consumption, such as Long Range/Long Range Wide Area Network (LoRa/LoRaWAN) [29], Narrowband IoT (NB-IoT) [30], and Sigfox [31]. Globally, the goal is to form the backbone of IoT ecosystems with battery lifetimes of up to several years, infrequent maintenance requirements, and reduced communication infrastructure costs.

LoRa/LoRaWAN is a wireless transmission technology that offers advantages such as long-range wireless communication with low-power consumption, resulting in longer battery life, simplicity, the ability to create private networks, security, and reduced maintenance requirements and operating costs [29]. LoRaWAN defines the medium access control protocol and system architecture for the network, while the LoRa physical layer enables long-range and low power communication using the unlicensed radio spectrum. IoT devices communicate with LoRa gateways, which then forward the data to a central network server via a standard IP connection. Multiple gateways may receive the same message, which improves reliability. The Things Network (TTN) is worth mentioning since it is a community-based initiative that provides such LoRaWAN gateways and an open-source network server for collecting and managing this data [32]. Data uploaded to the TTN server can be integrated with other platforms for visualization and analysis. LoRa technology supports a maximum data rate of up to 50 kbps, a maximum payload of 243 bytes, and a transmission range of more than 10 km in rural areas and 5 km in urban areas [33]. LoRa/LoRaWAN is an effective solution for collecting air pollution data over large areas, enabling real-time monitoring without requiring expensive network and power infrastructure. The main challenges are the installation of gateways, concerns about the security of transmitted data, and limitations in terms of bandwidth and quality of service.

NB-IoT is an LPWAN technology designed for IoT scenarios, which is based on the cellular network and, as a complement to these networks, supports the connectivity of devices with low power consumption, which increases battery life while maintaining the security and robustness of the cellular network [30]. With this technology, devices connect directly to cell towers in the same way as mobile phones, using the licenced radio spectrum and infrastructure operated by mobile network operators. In general, NB-IoT has lower latency than LoRa/LoRaWAN and can provide better quality of service, supporting up to 250 kbps maximum data rate, 1600 bytes maximum payload, and up to 25 km range [33]. However, it is more suitable for scenarios with static or low-mobility devices.

Sigfox [31] is another example of LPWAN technology designed specifically for IoT applications, offering ultra-narrowband communications in the unlicensed radio spectrum. Unlike NB-IoT, it does not rely on cellular infrastructure, instead using a proprietary global network operated by Sigfox or its partners. The IoT devices send messages directly to the nearest Sigfox base station (access point), which then forwards them to the cloud. Sigfox supports large-scale deployments of battery-powered sensors, but offers very limited bandwidth, making it suitable only for use cases with devices that transmit small amounts of data infrequently (e.g., environmental monitoring). The maximum data rate is a modest 100 bps, with transmission ranges up to 30–50 km in rural areas and 3–10 km in urban areas [33].

There are a few solutions in the literature that propose LPWAN-based IoT AQM systems. For example, in [34], the authors describe the design, fabrication, implementation, and validation of a portable LoRaWAN-based IoT AQM standalone system (AQMS) for outdoor use, which is referred to as LoRaWAN-IoT-AQMS. It uses several sensors for collecting different air quality indicators and the users can access this information in real time via a web-based dashboard and a mobile application. Similarly, [32] presents a low-cost, portable IoT-based AQM system that integrates low-cost sensors, LoRa communication, and cloud-based data storage. The device serves a dual purpose; it functions as a personal alert system, notifying the user when pollution levels are excessive, and simultaneously acts as an IoT device, transmitting environmental data to the cloud for storage, analysis, and remote access. The IoT solution proposed by [35] is another example of an air pollution monitoring prototype that uses LoRaWAN-compliant sensors for measuring CO₂, CO, and PM_2.5 levels. Ref [36] presents a novel air quality assessment system integrating WSN with NB-IoT, and LoRa communication, while another IoT system, presented in [37], also based on NB-IoT, was developed and assessed in Bangkok, Thailand, for air pollution detection and monitoring for smart cities. The use of Sigfox-based connectivity for portable and mobile IoT sensors in an urban environment was studied in Cagliari, Italy, and compared with Wi-Fi and cellular-based approaches in terms of packet loss and energy consumption [38]. The tests were conducted by collecting and sending air quality data, and the authors concluded that Sigfox improves the analyzed parameters in the mobile urban scenario addressed.

2.3. Blockchain Technology

Blockchain technology consists of a distributed database/ledger and was popularized by Bitcoin [39]. It is based on principles such as security, decentralization, and the absence of intermediaries. Blockchain ledgers consist of a chain of interconnected blocks, each one linked to its predecessor, which contain a set of transactions executed in a decentralized manner. Recorded transactions cannot be deleted or modified. There are several types of blockchain networks, which can be classified based on whether they are public, private, consortium, or hybrid [40]. Public blockchains allow anyone to join the network, private blockchains are permissioned networks controlled by a single organization, consortium blockchains are permissioned networks controlled by a group of organizations, community blockchains are governed by a specific community, and hybrid blockchains combine elements of several blockchain types.

The physical infrastructure of a blockchain consists of multiple interconnected computers, called nodes, which operate as a peer-to-peer (P2P) network. Secure data sharing and privacy mechanisms are based on cryptography, consensus algorithms, and digital signatures, and have the potential to provide effective transparency and traceability (i.e., transactions are publicly and permanently recorded, making them auditable) in a wide range of areas such as AQM, food, health, and waste management [25,41]. It is possible to store self-executing programs called smart contracts on the blockchain, which automate operations. These contracts are tamper-proof, once deployed, and can reduce costs, by eliminating intermediaries and delays [42], and be combined for creating decentralized applications that are not controlled by any single entity [43], reducing the risk of censorship or manipulation.

Currently, there are several blockchain platforms available, with distinct characteristics, strengths, and weaknesses that need to be considered when deciding which one to use in each specific scenario. For example, Avalanche, Ethereum, Hyperledger Fabric, Polygon, and Solana are five popular solutions that were studied and discussed in [44], an article that proposed a blockchain-based loyalty management system. The authors considered several key factors to analyze the various blockchain alternatives, such as the number of transactions per second, the cost per transaction, the ability to support smart contracts, and scalability. Polygon was selected as being the preferred blockchain network for the proposed system. According to the authors, this preference was due to its growing popularity, its ability to provide a fair price and a good volume of transactions per second, and its compatibility with Ethereum, which allows for easy transfer of smart contracts.

However, despite its strengths, the blockchain technology still has some limitations that need to be addressed to exploit its full potential. For example, scalability, network bandwidth, transaction processing speed and cost, and integration with other technologies are aspects that need to be improved in novel blockchain platforms [4,45]. According to [45], smart contracts, which implement the core logic layer of blockchain-based solutions, are also vulnerable to attacks when executed in public environments. Being aware of and managing the fast evolution of blockchain technology when investigating its application is a major issue, since the relevance of findings may decrease as novel solutions emerge [41].

Since blockchain technology is not efficient in storing large files due to its high storage costs, it can, for example, be used in conjunction with InterPlanetary File System (IPFS), a distributed P2P file system, composed of multiple nodes, that allows users to store and retrieve files by identifying them using hash codes based on their content, rather than typical identifiers such as file paths or uniform resource locators [46]. These codes, named unique content identifiers (CID), are unique to each file and generated using cryptographic algorithms, acting as permanent tamper-proof links to the files. This means that any modification to the content of a file after it has been stored will be detected, as its current CID becomes invalid. Therefore, using blockchain and IPFS technologies together, which essentially consists of storing large amounts of data off-chain in a decentralized, scalable, and resilient system and recording the small unique CID references and any other metadata on the blockchain, reduces storage costs and scalability limitations inherent to blockchain technology while preserving its benefits.

2.4. Augmented Reality Applied to Air Quality Monitoring

AR provides a visual overlay of environmental information on the user’s surroundings, such as air quality or traffic data, through camera-equipped devices, enabling an intuitive and immersive awareness of environmental issues that can help improve decision-making and promote behavioural and policy changes [47,48,49]. AR can be seen as making visible the invisible, by combining virtual objects and software-generated information (e.g., 3D models, contextual data, text, labels) with the real environment, which increases information dissemination, as well as user’s perception and engagement [50]. In other words, AR devices sense the environment and then render virtual objects into that environment [51]. In general, mobile AR applications use the device’s GPS to determine the user’s current location and the camera to enable the AR visualization. Mobile AR experiences are facilitated by powerful 3D graphics and fast processing systems [51]. Besides smartphones, AR can also use smart glasses and head-mounted displays that eliminate the gap between the user’s eyes and the display to which the data is transmitted [52]. Data visualized in AR applications can be sourced from a variety of open data platforms. For example, real-time visualization of air quality and traffic data in different European regions can be based on environmental sensor data that is publicly available through various web interfaces and platforms [48].

There is little work related to AQM based solely on AR. For instance, a search in the Web of Science online research database, conducted in December 2024, with the query “augmented reality” AND “air quality”, designed to target occurrences of these terms in the title, abstract, or keywords of publications, result in only fifty-eight (58) articles. Many of these articles were removed because the publications were either outside the intended area of knowledge or the full text was unavailable. The main results of the search are the following. The solution proposed by [47] aims to engage citizen communities by combining AR interfaces, IoT, and serious gaming concepts to raise awareness about air pollution issues in a fun and educational way, and to stimulate discussions with relevant authorities. The Situated Pollution application, proposed in [53], was implemented to visualize air quality, enhance public awareness, and help participants follow the least polluted route. The authors in [54] developed an educational game to teach children about air quality. This game uses AR and tangible interaction with sensors, enhancing the learning experience in terms of engagement and effectiveness. Through tangible objects and real-world scenarios, the game aims to make the abstract concept of air quality more concrete and relatable for children. Additionally, AR aids in visualizing invisible air pollutants, facilitating comprehension of their sources and effects on air pollution. In [55], the authors present GoNature AR, an innovative multimodal experience that integrates interactive narration, gesture and hand recognition, and voice commands through a head-worn AR display. This system combines AR technologies with a sensor network to promote environmental awareness by providing citizens with real-time data on air and noise pollution. Based on the premise that visualizing air quality in a person’s living area could help bring awareness, [56] presents AiR, a mobile application based on Android and AR technology. This application allows the public to visualize quantitative information about various air pollutants in their location (based on GPS devices) or in a selected area. The data is from the Central Pollution Control Board of India. Another interesting work on the user of AR in environmental context aimed at raising awareness and promoting sustainability, although it is not specifically about air pollution, is the survey in [57]. Finally, [58] introduces Aire, an application that educates users about air pollution and the contaminants in their environment. The application uses existing AR technology to provide accessible and engaging visualizations of air quality, raising understanding of this complex issue. Aire offers both single and multi-user experiences, encouraging collaboration and communication about the state of air quality. By creating visual representations, it highlights air quality conditions and serves as a catalyst for discussions about causes, consequences, and potential solutions.

2.5. Combining Blockchain, IoT, and AR for Air Quality Monitoring

Some authors consider that the combination of blockchain with IoT sensors enables the development of air and water pollution monitoring systems that are more transparent, efficient, and secure, which helps environmental regulations to be met and promotes accountability in AQM [4,22,23,24]. For example, the solution proposed by [24] uses low-cost sensors to collect air quality data, low-power communication protocols to transmit it, and blockchain technology to process and securely and reliably store it. Globally, it aims to create an automated, decentralized, and transparent system that addresses the limitations of traditional AQM solutions. The blockchain used in this solution also includes a smart contract to ensure data integrity and provide the basis for data analysis and access control. A web application can then serve as the front-end for data access and analysis, allowing users to extract and monitor air quality data from the blockchain in a user-friendly format.

Combining blockchain and AR in AQM is also suggested by some authors [23,49,50]. In this context, blockchain ensures data integrity and transparency, without the need for a central authority, by recording every event related to AQM in its decentralized, immutable, and tamper-proof ledger [22,23]. Due to its decentralized nature, it also makes the data accessible, which facilitates secure collaboration [22,23,24,51]. AR, on the other hand, improves the accessibility of this secured data, which expands the user experience and awareness, especially in urban environments, and promotes behavioural change and increased public engagement [48,49]. AR enables real-time data presentation in a user-friendly format, enabling immediate insight into local air quality and proactive response.

There are few works in the literature that combine these three technologies (AR, blockchain, and IoT). The known examples include studies in healthcare and medical education [59], product design and development [60], and sustainable tourism [61]. However, only the last one links all these technologies to air quality. Therefore, the discussion in the current and previous subsections suggests that this approach needs to be explored, as these technologies are likely to be key components of a flexible, low-cost, publicly accessible, transparent, and engaging system for AQM.

This discussion of related work highlights the absence of proposals that combine AR, blockchain, and IoT technologies into a single AQM model. The conceptual framework introduced in the next section aims to address this gap by offering a unified vision of these technologies, aiming to inspire any implementation of flexible, collaborative, transparent, immersive, and engaging AQM systems.

3. Conceptual Collaborative Air Quality Monitoring System

Building on the previous discussion, the exercise proposed in this section is to define the key components of a robust conceptual collaborative AQM system, in which individuals concerned about air quality can collect and share environmental data. In this framework, those individuals (the users of the system) must be associated with the area they choose to monitor, contributing to an increasingly comprehensive map as new users join. The data, once collected, should be analyzed, processed, stored, and made publicly available. By combining this information with contextual data such as traffic and weather conditions, an ample understanding of air quality dynamics can be achieved, supporting informed decision-making and increasing environmental awareness.

3.1. Architecture

An AQM system with the intended characteristics can be represented as a layered architecture that integrates the technologies for data collection, transmission, storage, processing, and visualization. Thus, its key components, shown in Figure 1, are as follows:

• Sensing layer (data collection):

  ▪

  Crowdsourced IoT sensors carried or deployed by individuals;

  ▪

  Mobile units attached to vehicles or personal devices to collect data;

  ▪

  Stationary monitoring stations;

  ▪

  Image-based methods through satellite imagery or street-level cameras can complement sensor data by providing spatial and temporal insight into air quality variations, increasing data granularity and filling gaps in areas where physical sensors may be sparse.

• Communication layer (data transmission):

  ▪

  The data collected can be transmitted in real time if it comes from a secure and validated source or stored for later transmission and processing if the data source has not yet been validated;

  ▪

  Wireless transmission via technologies such as Wi-Fi for short-range coverage, LoRaWAN for long-range, low-power coverage, or cellular networks for long-range coverage.

• Processing and storage layer:

  ▪

  Secure storage on local servers or cloud platforms;

  ▪

  Blockchain-based, cloud-based, and hybrid solutions to ensure data integrity and transparency, building trust.

• Application layer (data visualization and interaction):

  ▪

  User-friendly AR interfaces for immersive and geolocated insights, and as a fundamental means for committed engagement of individuals and communities in air quality awareness;

  ▪

  Interactive data with intelligent analytical tools.

The proposed architecture integrates IoT, blockchain, and AR technologies, each of which contributes specific strengths to a transparent, inclusive and interactive AQM system. IoT technologies enable cost-effective, scalable, distributed, fine-grained environmental sensing. Blockchain ensures the trustworthiness and immutability of collected data, thereby addressing transparency and accountability issues. On the other hand, AR promotes user engagement by providing intuitive, geolocated visualization of air quality data, helping to translate abstract metrics into understandable insights. Any implementation of the proposed architecture may follow a structured, modular development strategy. Each phase can be developed and validated independently before full system integration, enabling iterative prototyping with manageable complexity.

3.2. System Components

The IoT sensors to be used should be low-cost and capable of accurately measuring key air pollutants such as carbon dioxide (CO₂), nitrogen oxides (NO_x), particulate matter (PM_2.5), and ozone (O₃). The choice of low-cost sensors is driven by the need for wide-spread adoption, flexibility, and scalability, enabling more comprehensive data collection by involving a larger number of users or communities. Ensuring accuracy in measuring pollutants is essential for reliable data, as these pollutants have significant health and environmental impacts. Therefore, balancing affordability with precision is critical to make the system effective and practical. The work of [32] lists several sensor modules as typical low-cost devices that meet the defined criteria (e.g., accuracy between ±15%) for monitoring pollutants such as PM_2.5, O₃, CO, and NO₂ in the low-cost, IoT-based personal air quality monitor prototype it describes. In the same article, the authors also note that low-cost sensors require frequent calibration due to drift-related issues and can have slow response times. In [62], a comprehensive table reviewing various low-cost sensors used in air quality monitoring systems is provided.

Programmable boards (i.e., microcontrollers) are widely used to develop, evaluate, and demonstrate the feasibility of IoT systems. They function as IoT nodes or end devices, enabling sensor integration, data management, and communication handling. Their choice often depends on factors like cost, processing needs, and connectivity requirements. For example, [32] compares three programmable boards based on several key factors to design the prototype of a personal air quality monitor. The boards compared were Pycom FiPy, Arduino MKR WAN 1310, and AVR-IOT. Among other key factors for comparison, the authors used Wi-Fi, Lora, and Sigfox support, built-in GPS, deep sleep mode, and programming language. They chose the Pycom FiPy device due to the specific requirements of their project, including its integrated support for multiple communication protocols on-chip such as Lora, Sigfox, and Wi-Fi. Arduino Uno and Raspberry Pi must also be mentioned as boards commonly used in prototyping and implementing IoT air quality monitoring systems [34,63].

Although this article presents the system conceptually, any implementation and deployment would require consideration of specific issues and performance benchmarks. For example, gas sensors (e.g., for NO₂ or CO) often require frequent calibration due to drift, are more sensitive to noise, and have slower response times [32]. Additionally, data fusion techniques such as temporal and/or spatial averaging can enhance signal robustness. Outlier detection and statistical smoothing can compensate for sensor inaccuracies and anomalies. Communication latency depends on the network technology. For example, while LoRaWAN offers long-range, low-bandwidth transmission with latencies in the hundreds of milliseconds, NB-IoT and Wi-Fi support higher throughput, but at the cost of increased power consumption. Energy usage is a limiting factor for mobile AR devices and IoT sensors, with practical deployments needing to balance measurement frequency, transmission rates, and battery life. These trade-offs must guide the design of future flexible, collaborative, and engaging AQM systems.

The system will also require a public blockchain network to store and validate the collected air quality data, ensuring transparency, immutability, and data integrity. Smart contracts will automate data validation and access control. To further enhance data integrity, cryptographic protocols will secure the transmission of data to the blockchain. However, blockchain is not suitable for directly storing large datasets. Therefore, decentralized data storage solutions like IPFS or a hybrid cloud approach will store large datasets, while blockchain will manage transaction data and reliably store metadata, hashes, or content identifiers. Additionally, blockchain will ensure support to verify the authenticity of devices and users. In summary, any implementation should adopt a hybrid architecture where sensor data is stored off-chain in distributed systems like IPFS or cloud infrastructure, while reference metadata is stored on-chain. It must be noted that effective management of the entire data lifecycle, encompassing storage, retrieval, archiving and deletion, is essential to ensure the long-term availability and integrity of system data. For instance, IPFS, being a decentralized and content-addressable file system, does not inherently guarantee data persistence. Stored content may be subject to local garbage collection unless it is explicitly pinned (i.e., excluded from garbage collection) by a hosting node or maintained through a dedicated pinning service. Without pinning, files may be removed from the network over time and become inaccessible.

While blockchain provides a tamper-proof ledger to validate and store sensor data, it does not protect against IoT vulnerabilities such as device spoofing, GPS spoofing, and sensor tampering. These remain open challenges in collaborative environmental sensing and must be addressed in any implementation through secure hardware, device authentication, and anomaly detection mechanisms. A threat model will be necessary to ensure resilience and trustworthiness in real-world deployments.

Although the proposed approach offers transparency and decentralization, its scalability depends on the performance of the underlying technologies. Public blockchains typically support limited transactions per second, which may create a bottleneck as the number of contributors or sensor updates increases. Similarly, IoT networks, especially those based on LPWAN protocols like LoRaWAN, may experience congestion or latency as the number of devices increases. Therefore, any implementation must consider scalability solutions such as the previously mentioned off-chain storage approach.

It should be noted that any implementation of community-driven AQM systems based on the proposed conceptual architecture must address critical aspects beyond technology, such as data governance, user consent, liability, and ethical data usage. Key issues include data ownership, responsibility for errors or misuse, and ensuring fair access. Transparent data-sharing policies, informed consent, anonymization, and governance frameworks will be crucial for public trust. Although blockchain can ensure traceability, broader institutional and legal mechanisms are needed to manage liability and oversight in real-world applications.

Following the data flow of a generalized architecture for distributed, collaborative, and trusted environmental sensing (Figure 2), sensor devices collect air quality data and transmit it over a communication network (e.g., LoRa, NB-IoT, Wi-Fi, GSM). The data is then routed through intermediate servers, either local or cloud-based, which perform some pre-processing tasks such as validation and filtering. Finally, it is stored in a decentralized/cloud database system with a unique content identifier (e.g., CID if IPFS is used) that is permanently recorded on the chosen blockchain platform to ensure integrity and traceability. The air quality data stored in the system can then be accessed by different stakeholders through appropriate API endpoints, including AR-based user interfaces for visualization. This architecture supports different types of sensors, network technologies, and processing infrastructures, resulting in flexibility combined with data integrity and trustworthiness. It supports transparent environmental monitoring with a verifiable data source, allowing access to historical air quality data via API endpoints while maintaining the integrity and auditability of the information.

Figure 2 provides a conceptual, high-level view of the architecture’s data flow, abstracting implementation-level details. However, it is important to note that relevant performance parameters, such as timing constraints, communication bottlenecks, transaction throughput, data rates, and latency, depend on the specific implementation choices, including the communication protocol, hardware platform, blockchain solution, and AR rendering pipelines. While these aspects are beyond the scope of the article’s theoretical contribution, they are key areas for any research and implementation.

As a concrete implementation, low-cost sensors can be deployed using LoRa/LoRaWAN to transmit measurements over long distances with minimal power consumption. These sensors can connect to TTN, which manages device authentication, message routing, and application layer integration via MQTT (Message Queuing Telemetry Transport), which is a lightweight, publish–subscribe messaging protocol designed for low-bandwidth, high-latency, or unreliable networks, or via webhook endpoints that allow external services to push data to an application via HTTP as it becomes available. A backend ingestion service receives the TTN data, filters and formats it, stores the payloads in IPFS, and registers their CID and other relevant metadata on the selected blockchain platform.

For AR, the system should use smartphones or tablets as the primary hardware to display air quality data through AR overlays, though AR glasses and other wearable devices could provide a more immersive and handless experience, allowing users to monitor air quality while moving through their environment. It is important to note that different AR hardware platforms demand fundamentally distinct approaches to presentation, interaction, and visualization. Therefore, using multiple platforms simultaneously requires a coherent integration of these varied approaches.

The AR application should enable geolocated data visualization. To accomplish that, it must employ spatial computing to anchor AR pollutant representations more precisely in the user’s environment. To implement AR spatial anchoring (i.e., attaching virtual objects to fixed physical locations in the real world so that they appear consistently positioned even as the user moves around), it is essential to combine hardware such as depth sensors (e.g., LiDAR—Light Detection and Ranging), SLAM (Simultaneous Localization And Mapping)-capable cameras, and GPS (for outdoor use) with software frameworks like ARKit (iOS), ARCore (Android), Unity AR Foundation, or cloud-based solutions like Azure Spatial Anchors. These technologies rely on SLAM algorithms, plane detection (i.e., the detection of flat surfaces in the physical environment), and mesh reconstruction (i.e., creating a 3D model of the real-world environment by scanning and mapping its geometry) to understand and persistently map the physical environment. For cross-session and multi-user scenarios, cloud services such as ARCore Cloud Anchors, Azure Spatial Anchors, or Niantic’s VPS enable persistent, shareable anchors. Accurate anchoring also depends on spatial understanding techniques and may involve local serialization of environment maps or cloud-stored anchor IDs with associated metadata.

By using spatial mapping, the AR visualizations become more realistic and interactive, improving the user’s sense of immersion and engagement. Integration with GIS tools could also be used to ensure accurate mapping of air quality data in real time, with overlays showing pollutant levels and contextual information such as traffic and weather conditions. Three-dimensional models and heatmaps can be used to visually represent air quality, while GPS integration will ensure data is correctly located. For example, 3D models representing pollutant molecules, scaled and positioned in relation to their concentration levels, will provide spatial and conceptual context. Concurrently, heatmaps can display real-time pollutant concentrations, overlaid onto the user’s environment. These visualizations should be dynamically updated.

Although AR offers promising opportunities for immersive and geolocated data visualization, its real-world application, particularly outdoors, faces several challenges. These include device heterogeneity (e.g., varying sensor capabilities and display resolutions), high power consumption on mobile devices, and environmental factors, such as lighting conditions, obstructions (e.g., buildings and trees), and GPS inaccuracies, which can interfere with spatial anchoring and rendering stability. Any development and deployment of AR-enhanced AQM systems must carefully address these limitations.

The AR user interface must be intuitive and display clear air quality indicators to ensure ease of interaction, as a user-friendly and valuable experience is essential to prevent adoption resistance from the users. Therefore, usability principles (e.g., Nielsen’s Heuristics [64]) and user experience factors, such as attention overload and interaction simplicity, must be considered during development since they are essential for effective adoption and comprehension.

The system should include AR-based educational modules to teach users about air quality, pollution sources, and mitigation strategies, making abstract concepts more concrete and relatable to improve learning experience. Educational modules will be tailored for general-public audiences, according to their age, incorporating AR visualizations to explain the effects of different pollutants. Interactive scenarios, such as simulating the impact of reducing car traffic or planting trees, will be used to teach concepts of air quality management. Stakeholders include schools, environmental non-governmental organizations, municipal authorities, and citizen science groups. While the primary target users are environmentally conscious citizens, the system should be designed with scalability in mind, supporting broader applications such as community-driven environmental awareness campaigns and municipal planning initiatives.

The system should support multi-user AR experiences, enabling communities to collaboratively monitor and discuss air quality. Multiple users can view and interact with the same AR visualizations simultaneously, raising collective awareness and action. This feature is particularly useful for educational purposes or community-led environmental initiatives. Real-world adoption of sensor-carrying or AR-interacting behaviour may suffer from participation fatigue or novelty decay. Therefore, the system should incorporate motivation strategies, such as gamification elements, into the AR experience to raise user engagement (Figure 3). For example, users can earn points or badges for monitoring air quality regularly, sharing data, or taking actions to reduce pollution in their area, encouraging sustained participation, and making the monitoring process more enjoyable.

Since one of the goals of the system is to drive behavioural changes by making air quality data more tangible and actionable, it must have the ability to show the immediate impact of the users’ actions on local air quality. To assess the impact of user actions, the system must include in-app surveys and usage logs to monitor behavioural changes over time. For example, users who choose eco-friendly transportation options could receive feedback through reduced pollution indicators in the AR view. A combination of pre- and post-intervention questionnaires and semi-structured interviews should be used to evaluate learning outcomes and behavioural shifts. The AR interface should include AI algorithms based on air quality data and user preferences, location, and health data.

3.3. Implementation Considerations

This subsection outlines possible technological solutions and platforms that can be used to implement immersive, secure, and collaborative AQM systems based on the proposed conceptual framework. Future implementations should follow a structured, modular development strategy with the following global phases: conceptualization (e.g., environmental research, stakeholder analysis), system design, implementation, deployment (e.g., pilot testing), and evaluation (e.g., performance assessment, usability).

As shown in Figure 1, the conceptual layers of the proposed framework are as follows: sensing, communication, processing and data storage, and application. To some extent, each layer can be modelled, developed, and validated independently before full system integration, enabling iterative prototyping with manageable complexity.

For the sensing layer, low-cost sensors such as the MQ-135 (for gases like NO₂ and CO₂), SDS011 (for PM_2.5), and MiCS-5524 (for multiple gases) can be integrated with microcontroller boards like Arduino MKR WAN 1310, Pycom FiPy, or Raspberry Pi. Many of these boards support important features such as low-power operation and built-in GPS, which can be critical for mobile sensing units. Programming can be carried out in diverse languages, such as C++ (Arduino IDE) and Python (Raspberry Pi). Data calibration and compensation algorithms can be implemented to address sensor drift and improve measurement accuracy.

For communication, LoRa/LoRaWAN and NB-IoT are examples of technologies that provide long-range, low-power data transmission, essential for outdoor deployments. MQTT or HTTP webhooks can be used for lightweight and scalable data streaming to backend systems. Other technologies, such as Wi-Fi, can complement these solutions in environments where high-bandwidth connectivity is feasible. For intermediate processing (see Figure 2), cloud-based IoT platforms such as Amazon’s AWS IoT Core and Microsoft’s Azure IoT Hub can offer device management and data ingestion capabilities.

The processing and data storage layer will adopt a hybrid approach; blockchain platforms such as Ethereum or Polygon will store immutable metadata and content hashes, while sensor data will be stored on the IPFS to ensure scalability and efficient data retrieval. Smart contracts (e.g., written in Solidity, a programming language) can automate data validation and access control. For development, frameworks like Truffle Suite and Hardhat (Ethereum) can be employed. For testing, solutions such as Rinkeby (Ethereum) and Polygon Mumbai Testnet can be used. Web interfaces for the processing and data storage layer can be developed using frameworks such as Flask, Django, and Spring Boot, while common database management systems can be used to store and manage data such as user account information.

To evaluate feasibility, tools such as ns-3 (Network Simulator 3) and OMNeT++ (Objective Modular Network Testbed in C++) can simulate communication latency and energy consumption across IoT networks, while blockchain testnets, a type of dedicated blockchain environment for development and testing purposes, can enable validation of smart contract execution and transaction throughput. Any pilot implementation will need to assess key factors such as latency in data transmission, energy consumption for battery-powered devices, and blockchain transaction throughput limitations. Scalability modelling should also explore how the system behaves with increasing numbers of participants and sensors, employing simulation tools or small-scale prototypes to benchmark performance under realistic conditions.

Finally, as previously mentioned, AR visualization and interaction in the application layer can be implemented using ARCore or ARKit for native AR development, or Unity’s AR Foundation for cross-platform AR experiences. These frameworks support spatial anchoring and SLAM for 3D overlays of air quality data. GIS data can be integrated through APIs like Google Maps or OpenStreetMap. Supported programming languages include Java/Kotlin for Android, Swift for iOS, and C# for Unity. Tools such as Unity’s Profiler will assist in measuring AR rendering performance.

4. Conclusions

The critical issue of air pollution was addressed in this article, highlighting its threat to public health and the environment, and the necessity of improved AQM for informed decision-making. It identified several key challenges with traditional AQM methods, including the lack of transparency, reliability, accuracy, accountability, and vulnerability in data sharing systems. Furthermore, conventional stationary monitoring stations are noted as being costly, lacking flexibility, and not providing fine-grained data across specific areas.

Recognizing the identified limitations, the article established the need for novel AQM solutions that are decentralized, accessible, collaborative, and interactive to stimulate wider participation and effective data utilization. Then, it explored how AR, blockchain, and IoT technologies can be used to offer an approach to developing such systems. This exploration includes a review of existing AQM techniques (sensor-based, image-based, stationary, and mobile), relevant communication technologies for IoT sensors (Wi-Fi, cellular, LPWAN like LoRa/LoRaWAN, NB-IoT, Sigfox), blockchain technology and its characteristics (security, decentralization, smart contracts, scalability challenges, use with IPFS), and applications of AR in environmental contexts. The limited existing work combining these three technologies for AQM was also noted.

Building on this review, the article presented a conceptual framework for a collaborative AQM system, detailing its layered architecture for data collection, transmission, processing, storage, and visualization/interaction. In the proposed approach, IoT sensors, particularly low-cost and mobile units, are leveraged for fine-grained, real-time, and flexible data collection, supporting crowdsourcing and enabling wider adoption and scalability. Blockchain technology is fundamental to ensuring the transparency, immutability, security, and trustworthiness of the collected data. AR provides a user-friendly and immersive interface for visualizing air quality data in a geolocated context, effectively enhancing user awareness, decision-making, and potential behavioural change. This integrated approach aims to empower individuals and communities, fostering greater awareness and active participation in environmental monitoring.

As a theoretical contribution, this article provides a unified conceptual basis upon which any implementation and validation can be built. Future work should extend it with detailed system design specifications, performance benchmarks, threat models, and mitigation strategies to address challenges related to the integrated technologies, such as inherent vulnerabilities and scalability limitations. Pilot deployments and user-centred studies will also be required to evaluate performance, usability, and impact in real-world contexts.

In conclusion, the implementation of systems based on the proposed approach can represent a significant step forward in the field of AQM. By combining the strengths of AR, blockchain, and IoT, the issue of air pollution can be effectively addressed, contributing to healthier communities and a more sustainable environment. As these technologies continue to evolve, their integration holds great promise for transforming how we monitor, understand, and respond to environmental challenges.