IoT-based Real-time Temperature Monitoring in Critical Systems: A Review

Abstract

The accurate monitoring of temperature is critical in a wide range of sectors, including healthcare, pharmaceuticals, food logistics, and data centers, where even minor thermal deviations can compromise safety, quality, and operational efficiency. Over recent years, the evolution of the Internet of Things (IoT) and embedded systems has significantly enhanced real-time temperature monitoring, enabling precise control via interconnected and autonomous systems.

This work aims to review and analyze current temperature measurement systems designed for critical environments. It identifies the strengths and weaknesses of different architectural approaches and their core components—such as sensors, actuators, microcontrollers, and gateways—focusing on aspects like scalability, reliability, energy efficiency, and response time. The study also highlights challenges and gaps in current technologies.

The review was conducted using a traditional literature analysis methodology, drawing from scientific databases including IEEE Xplore, B-on, and Google Scholar. Key research was synthesized to provide a comprehensive and up-to-date perspective on IoT-based temperature monitoring technologies and their applications.

The paper contributes a structured evaluation of the field and supports future research by outlining essential requirements, comparing technologies, and proposing directions for improvement in the development of adaptive, secure, and efficient temperature monitoring systems for high-risk environments.

Introduction

In many critical environments, such as pharmaceutical warehouses, medical facilities, food logistics, and data centers, maintaining precise temperature control is essential to ensure product safety, system efficiency, and regulatory compliance. The failure to detect and respond to thermal deviations in real time can lead to material degradation, system downtime, and even risks to human health. As such, robust and continuous temperature monitoring is essential.

The rise of the Internet of Things (IoT) and advances in embedded systems have enabled the development of smarter, more connected, and autonomous monitoring solutions. These systems support distributed sensor deployment and local data processing via edge computing, and efficient communication between nodes and centralized platforms. This paradigm shift has improved the responsiveness, scalability, and reliability of environmental monitoring infrastructures, making them suitable for complex, real-world deployments.

The primary objective of this work is to explore the current technological landscape of IoT-based temperature monitoring systems in high-precision scenarios. This includes a comparative analysis of various system architectures and core components such as sensors, actuators, microcontrollers, gateways, interconnection protocols, and software platforms. The study also aims to identify existing limitations, practical trade-offs, and open challenges that still hinder broader adoption or long-term reliability in demanding applications.

To support this analysis, a traditional literature review methodology was employed, with relevant publications sourced from reputable scientific databases including IEEE Xplore, B-on, and Google Scholar. The review focused on identifying the most impactful research and technological solutions applied in real-world scenarios.

The structure of this article is as follows: the next section presents typical IoT architectures for temperature monitoring, followed by an in-depth overview of sensing technologies and actuators, along with a discussion of edge node implementation and communication strategies. The paper also evaluates IoT software platforms and orchestration approaches, presents a critical assessment of emerging technologies, and concludes with the main findings and proposed directions for future research.

IoT system architectures for temperature monitoring

IoT-based temperature monitoring systems require well-structured architectures to ensure reliability, scalability, and efficiency across all stages of data collection, processing, and transmission. Several architectural models have emerged over time, including the widely adopted three-layer, five-layer, and seven-layer designs. In addition to layered architectures, these systems can be categorized as centralized or decentralized. The choice of architecture depends on multiple factors such as the number of devices, communication requirements, energy efficiency, and processing capabilities [11Kutseva M. Adaptation of Seven-Layered IoT Architecture for Energy Efficiency Management in Smart House. In: 2022 10th International Scientific Conference on Computer Science, COMSCI 2022 - Proceedings. 2022. doi:10.1109/COMSCI55378.2022.9912604.].

The three-layer architecture, composed of perception, network, and application layers, is the most common and is valued for its simplicity and effectiveness [22Three-layered software architecture and its variability for teleoperated system. IEEE Xplore; 2025 Apr 22[Online]. Available from: https://ieeexplore.ieee.org/document/7347791]. In this model, environmental data from temperature sensors is collected at the perception layer, transmitted via protocols like Wi-Fi, LoRa, or ESP-based communication (e.g., ESP-NOW, ESP-MESH in mesh or star topologies) at the network layer, and then processed and visualized at the application layer through mobile or web interfaces. While this structure is easy to implement and integrate, it may struggle with scalability and centralized processing bottlenecks.

To address these issues, the five-layer architecture introduces two additional layers: processing and management. These layers enable edge-level data filtering, compression, OTA updates, device authentication, and centralized coordination of devices. While this architecture provides greater robustness and is better suited to large-scale or mission-critical applications requiring cloud integration and artificial intelligence, it also increases system complexity and potential latency due to additional processing overhead [11Kutseva M. Adaptation of Seven-Layered IoT Architecture for Energy Efficiency Management in Smart House. In: 2022 10th International Scientific Conference on Computer Science, COMSCI 2022 - Proceedings. 2022. doi:10.1109/COMSCI55378.2022.9912604.].

The seven-layer architecture, inspired by the OSI model, provides a comprehensive structure for critical applications such as medical monitoring or pharmaceutical cold chains [33Sarkar S, Akshatha KS, Saurabh A, Samanvitha B, Sarwar MF. IoT Enabled Cold Supply Chain Monitoring System. In: 2022 IEEE 3rd Global Conference for Advancement in Technology, GCAT 2022. 2022. doi:10.1109/GCAT55367.2022.9972137.,44Al-Awami SH, Al-Aty MM, Al-Najar MF. Comparison of IoT Architectures Based on the Seven Essential Characteristics. In: Proceedings - 2023 IEEE 3rd International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and Computer Engineering, MI-STA 2023. 2023;305–10. doi:10.1109/MI-STA57575.2023.10169289.]. It includes the physical layer (temperature sensors and actuators), the data link layer (communication protocols like ESP-MESH or Zigbee), the network layer (routing), the transport layer (reliable delivery via TCP/UDP), the session layer (persistent connections), the presentation layer (data encryption and compression), and the application layer (user interfaces and data analysis).

Regarding performance metrics, communication protocols differ significantly under varying conditions. LoRaWAN, for example, offers excellent long-range coverage but operates at low data rates (0.3 - 50 kbps) and exhibits uplink latencies between 1 and 3 seconds per transmission, depending on payload size and transmission mode. It is therefore better suited for low-frequency measurements rather than real-time monitoring [55Al-Sammak KA, Al-Gburi SH, Marghescu C, Dragulinescu AM, Suciu G, Abdulqader AG. A Comprehensive Assessment of LoRaWAN and NB-IoT Performance Metrics under Varied Payload Data Sizes. In: Proceedings of the 16th International Conference on Electronics, Computers and Artificial Intelligence, ECAI 2024. 2024. doi:10.1109/ECAI61503.2024.10607481.]. On the other hand, ESP-NOW and ESP-MESH support higher throughput (typically ~200 kbps) and significantly lower latency, often under 100 milliseconds per hop, making them ideal for real-time applications with multiple sensor nodes operating in local environments [66Eridani D, Rochim AF, Cesara FN. Comparative Performance Study of ESP-NOW, Wi-Fi, Bluetooth Protocols based on Range, Transmission Speed, Latency, Energy Usage and Barrier Resistance. In: Proceedings - 2021 International Seminar on Application for Technology of Information and Communication, iSemantic 2021. 2021 Sep;322–8. doi:10.1109/ISEMANTIC52711.2021.9573246.].

In terms of system organization, centralized architectures gather all data in a single server, simplifying management and integration with applications, but introducing a single point of failure and scalability limitations [77Kassim MRM. Applications of IoT and Blockchain in Smart Agriculture: Architectures and Challenges. In: 2022 IEEE International Conference on Computing, ICOCO 2022. 2022;253–8. doi:10.1109/ICOCO56118.2022.10031697.]. Decentralized architectures, by contrast, distribute processing across multiple nodes, allowing devices to communicate directly. This enhances resilience, reduces latency through local processing, and is especially effective with ESP-MESH networks. However, it also increases system complexity and demands more capable edge devices [77Kassim MRM. Applications of IoT and Blockchain in Smart Agriculture: Architectures and Challenges. In: 2022 IEEE International Conference on Computing, ICOCO 2022. 2022;253–8. doi:10.1109/ICOCO56118.2022.10031697.].

For systems based on ESP32 microcontrollers and ESP-MESH communication, a decentralized five-layer architecture has proven particularly effective. This model supports large-scale deployments without overwhelming a central server, takes full advantage of direct device-to-device communication, and includes dedicated layers for device management and security. When combined with cloud integration, it provides a robust, secure, and energy-efficient architecture that is well-suited for high-reliability use cases such as hospital monitoring, cold chain logistics, and industrial environments [88Gül F, Eroğlu H. Low-Cost IoT Mesh Network for Real-Time Indoor Air Quality Monitoring. In: Proceedings of the 9th International Conference on Communication and Electronics Systems, ICCES 2024. 2024;502–7. doi:10.1109/ICCES63552.2024.10860109.].

Sensors and actuators

Sensors and actuators are key components in temperature monitoring systems, enabling precise environmental data acquisition and automated responses to thermal changes. Working as complementary elements in a closed-loop control system, sensors collect temperature data while actuators execute corrective actions based on those inputs, ensuring real-time environmental regulation [99Sehrawat D, Gill NS. Smart sensors: Analysis of different types of IoT sensors. In: Proceedings of the International Conference on Trends in Electronics and Informatics, ICOEI 2019. 2019 Apr;523–8. doi:10.1109/ICOEI.2019.8862778.].

These devices are typically integrated through a microcontroller, such as the ESP32, which processes the sensor data and coordinates actuator responses. This setup enables fast and energy-efficient reactions to temperature fluctuations, maintaining environmental conditions within predefined thresholds [1010Garg R, Kumar A, Singh S, Dayana R. Advanced Air Quality Monitoring System using IoT and Sensor Technology. In: 3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025. 2025;687–92. doi:10.1109/IDCIOT64235.2025.10915106.].

Sensor selection plays a critical role in ensuring precision and reliability, especially in sensitive environments. Among the most suitable sensors are the DS18B20, a robust digital sensor with ± 0.5 °C accuracy and one-wire interface ideal for industrial and cold chain applications; the SHT31, offering high-precision temperature (±0.2 °C) and humidity (±2%) readings with excellent long-term stability, is suitable for scientific and industrial uses; and the BME680, a multi-sensor capable of measuring temperature, humidity, pressure, and air quality (VOCs), ideal for comprehensive environmental monitoring in healthcare and pharmaceutical settings [1010Garg R, Kumar A, Singh S, Dayana R. Advanced Air Quality Monitoring System using IoT and Sensor Technology. In: 3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025. 2025;687–92. doi:10.1109/IDCIOT64235.2025.10915106.,1111Ahmad YA, Gunawan TS, Mansor H, Hamida BA, Hishamudin AF, Arifin F. On the Evaluation of DHT22 Temperature Sensor for IoT Application. In: Proceedings of the 8th International Conference on Computer and Communication Engineering, ICCCE 2021. 2021 Jun;131–4. doi:10.1109/ICCCE50029.2021.9467147.].

The ESP32 facilitates the integration of these sensors through native support for I2C and 1-Wire protocols, ensuring low power consumption and reliable data transmission over wireless interfaces such as Wi-Fi and Bluetooth Low Energy [1212Maier A, Sharp A, Vagapov Y. Comparative analysis and practical implementation of the ESP32 microcontroller module for the internet of things. In: 2017 Internet Technologies and Applications, ITA 2017 - Proceedings of the 7th International Conference. 2017 Nov;143–8. doi:10.1109/ITECHA.2017.8101926.].

Edge nodes

Edge nodes are essential components in IoT architectures for temperature monitoring, acting as intermediaries between sensor devices and centralized cloud infrastructure. These nodes enable not only data collection but also local filtering and preliminary processing, significantly reducing the computational load on central servers and enhancing system efficiency [1313Yacchirema D, Palau C. Internet of things Interoperability: A smart IoT gateway approach. In: Iberian Conference on Information Systems and Technologies, CISTI. 2023 Jun. doi:10.23919/CISTI58278.2023.10211617.]. Edge nodes are especially relevant in industrial and mission-critical environments where latency and autonomous operation are key factors.

In temperature monitoring systems, edge nodes process thermal data locally before transmission to the cloud, minimizing both latency and bandwidth consumption. Hardware selection depends on system requirements: low-power microcontrollers like the ESP32 and ESP8266 are suitable for energy-efficient deployments, while more powerful microcomputers like the Raspberry Pi or Jetson Nano are used for applications requiring advanced processing capabilities [1313Yacchirema D, Palau C. Internet of things Interoperability: A smart IoT gateway approach. In: Iberian Conference on Information Systems and Technologies, CISTI. 2023 Jun. doi:10.23919/CISTI58278.2023.10211617.,1414Kashyap M, Sharma V, Verma G. Implementation and Analysis of IoT Based Automation Using MQTT. In: 2021 IEEE 4th International Conference on Computing, Power and Communication Technologies, GUCON 2021. 2021 Sep. doi:10.1109/GUCON50781.2021.9573857.].

Microcontrollers are ideal for accurate data acquisition and wireless communication via protocols such as Wi-Fi or Bluetooth Low Energy. Microcomputers, on the other hand, support storage and machine learning-based analysis for predictive insights [1515Tondato De Faria B, Tercete GM, Filev Maia R. The effectiveness of IoT and machine learning in Precision Agriculture. In: 2022 Symposium on Internet of Things, SIoT 2022. 2022. doi:10.1109/SIOT56383.2022.10070308.]. This local decision-making capacity enhances system responsiveness and reduces dependence on constant cloud connectivity, which is crucial in applications like pharmaceutical cold chain monitoring or data center thermal management [1313Yacchirema D, Palau C. Internet of things Interoperability: A smart IoT gateway approach. In: Iberian Conference on Information Systems and Technologies, CISTI. 2023 Jun. doi:10.23919/CISTI58278.2023.10211617.].

The proposed architecture follows a hybrid model
(Figure 1): ESP32 microcontrollers are deployed as peripheral sensor nodes, communicating via ESP-MESH—a scalable, low-power mesh protocol [88Gül F, Eroğlu H. Low-Cost IoT Mesh Network for Real-Time Indoor Air Quality Monitoring. In: Proceedings of the 9th International Conference on Communication and Electronics Systems, ICCES 2024. 2024;502–7. doi:10.1109/ICCES63552.2024.10860109.,1616Dzahir MASM, Chia KS. Evaluating the Energy Consumption of ESP32 Microcontroller for Real-Time MQTT IoT-Based Monitoring System. In: 2023 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies, 3ICT 2023. 2023;255–61. doi:10.1109/3ICT60104.2023.10391358.]. The Raspberry Pi 4 acts as the central gateway, aggregating and pre-processing data before cloud transmission, while also acting as a local storage buffer during connectivity interruptions [1313Yacchirema D, Palau C. Internet of things Interoperability: A smart IoT gateway approach. In: Iberian Conference on Information Systems and Technologies, CISTI. 2023 Jun. doi:10.23919/CISTI58278.2023.10211617.].

This architecture offers four main advantages: (1) long-term battery operation through the ESP32’s ultra-low power modes [1616Dzahir MASM, Chia KS. Evaluating the Energy Consumption of ESP32 Microcontroller for Real-Time MQTT IoT-Based Monitoring System. In: 2023 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies, 3ICT 2023. 2023;255–61. doi:10.1109/3ICT60104.2023.10391358.], (2) mesh network scalability without dedicated Wi-Fi infrastructure [88Gül F, Eroğlu H. Low-Cost IoT Mesh Network for Real-Time Indoor Air Quality Monitoring. In: Proceedings of the 9th International Conference on Communication and Electronics Systems, ICCES 2024. 2024;502–7. doi:10.1109/ICCES63552.2024.10860109.], (3) data integrity through local buffering on the Raspberry Pi [88Gül F, Eroğlu H. Low-Cost IoT Mesh Network for Real-Time Indoor Air Quality Monitoring. In: Proceedings of the 9th International Conference on Communication and Electronics Systems, ICCES 2024. 2024;502–7. doi:10.1109/ICCES63552.2024.10860109.], and (4) real-time decision-making with reduced cloud dependency, leveraging local algorithms for basic system operations [1313Yacchirema D, Palau C. Internet of things Interoperability: A smart IoT gateway approach. In: Iberian Conference on Information Systems and Technologies, CISTI. 2023 Jun. doi:10.23919/CISTI58278.2023.10211617.].

Tables 1,2 list the hardware and software components used in the proposed IoT Edge Monitoring System, respectively.

Table 1: Hardware Components Used in the IoT Edge Monitoring System.

Table 2: Software Components Used in the IoT Edge Monitoring System.

Interconnection solution

Efficient interconnection of devices is a foundational pillar in developing robust IoT systems for temperature monitoring, ensuring reliable communication among sensors, actuators, and edge nodes. Selecting the appropriate communication protocol requires a multidimensional analysis that balances coverage, bandwidth, energy consumption, implementation cost, and scalability [1717Ali AI, Partal SZ, Kepke S, Partal HP. ZigBee and LoRa based Wireless Sensors for Smart Environment and IoT Applications. In: Proceedings - 2019 IEEE 1st Global Power, Energy and Communication Conference, GPECOM 2019. 2019 Jun;19–23. doi:10.1109/GPECOM.2019.8778505.].

For continuous thermal monitoring systems, three key features are essential: low energy consumption for prolonged operation, sufficient communication range, and native support for decentralized architectures to improve system resilience. In this context, wireless mesh networks are particularly relevant due to their ability to eliminate single points of failure by allowing direct device-to-device communication [1818Mekki K, Bajic E, Meyer F. Indoor positioning system for IoT device based on BLE technology and MQTT protocol. In: IEEE 5th World Forum on Internet of Things, WF-IoT 2019 - Conference Proceedings. 2019 Apr;787–92. doi:10.1109/WF-IOT.2019.8767287.].

IoT communication technologies fall into two main categories: wired solutions like Ethernet and RS-485, known for reliability but limited in flexibility; and wireless protocols such as Wi-Fi, BLE, LoRaWAN, Zigbee, and ESP-MESH. Each has strengths suited to different use cases, and a comparative evaluation should consider four criteria: range, energy efficiency, data throughput, and resistance to electromagnetic interference.

In traditional centralized IoT architectures, all sensor nodes communicate directly with a single gateway or cloud server, which handles all data processing and routing. While simple to implement, this model introduces a single point of failure and may struggle to scale efficiently in environments with a high density of devices. In contrast, decentralized architectures, such as those based on mesh networking, enable devices to communicate with each other directly, forming a distributed communication network (Figure 1). ESP-MESH, specifically, is a mesh protocol developed by Espressif for ESP32 devices that allows nodes to relay data between each other until it reaches the gateway. This greatly improves coverage, network resilience, and fault tolerance, as the system can adapt dynamically to changes in topology or connectivity. By enabling communication across multiple hops, ESP-MESH reduces the load on individual nodes and eliminates reliance on a single access point, making it especially suitable for large-scale or critical applications where connectivity and robustness are essential [1919Aziz ANAA, Rashid RA, Nasir MNM, Sarijari MA. ESP Mesh Network for Security Application. In: Conference Proceedings - IEEE International Conference on Advanced Telecommunication and Networking Technologies, ATNT 2024. 2024. doi:10.1109/ATNT61688.2024.10719207.,2020Yusof MK, Man M, Wan Hamzah WMAF, Safei S, Tasani MH, Shah NAAA. A Secure Model Based on Symmetric Encryption for REST API in Data Integration. In: 27th International Computer Science and Engineering Conference 2023, ICSEC 2023. 2023;402–6. doi:10.1109/ICSEC59635.2023.10329720.].

Figure 1: System architecture of the proposed IoT-based monitoring platform. Sensor nodes using ESP32 microcontrollers communicate via ESP-MESH, transmitting encrypted data to an ESP32 gateway. The data is routed through an MQTT broker to a Raspberry Pi functioning as an edge device. The Raspberry Pi runs containers for the MQTT broker, backend API (written in Go), and database (MongoDB). Client applications, both web and mobile, access processed data from the backend, while an orchestrator handles containerized service management.

On the software side, the proposed architecture adopts Go (Golang) as the primary development language because of its resource efficiency, native support for concurrency, and compatibility with various communication protocols and APIs. This enables seamless integration from sensor-level data acquisition over ESP-MESH [88Gül F, Eroğlu H. Low-Cost IoT Mesh Network for Real-Time Indoor Air Quality Monitoring. In: Proceedings of the 9th International Conference on Communication and Electronics Systems, ICCES 2024. 2024;502–7. doi:10.1109/ICCES63552.2024.10860109.,1919Aziz ANAA, Rashid RA, Nasir MNM, Sarijari MA. ESP Mesh Network for Security Application. In: Conference Proceedings - IEEE International Conference on Advanced Telecommunication and Networking Technologies, ATNT 2024. 2024. doi:10.1109/ATNT61688.2024.10719207.] to communication with cloud-based analytics platforms via RESTful APIs and WebSockets [2121Garg H, Dave M. Securing IoT Devices and Securely Connecting the Dots Using REST API and Middleware. In: Proceedings - 2019 4th International Conference on Internet of Things: Smart Innovation and Usages, IoT-SIU 2019. 2019 Apr. doi:10.1109/IOT-SIU.2019.8777334.].

As the solution is still under development, one of the proposed security mechanisms includes the design of a dedicated Encryption API, which is responsible for securely managing all encryption and decryption processes. The core encryption algorithm considered is AES-256 in CBC (Cipher Block Chaining) mode, chosen for its balance between strong security and performance in constrained environments. For secure key exchange, RSA-2048 or Elliptic Curve Diffie-Hellman (ECDH) are being evaluated, depending on the trade-off between computational load and cryptographic strength [2222Setiadi dRIM, Najib AF, Rachmawanto EH, Sari CA, Sarker MK, Rijati N. A comparative study MD5 and SHA1 algorithms to encrypt REST API authentication on mobile-based application. In: 2019 International Conference on Information and Communications Technology, ICOIACT 2019. 2019 Jul;206–11. doi:10.1109/ICOIACT46704.2019.8938570.,2323Gunathilake NA, Buchanan WJ, Asif R. Next Generation Lightweight Cryptography for Smart IoT Devices: Implementation, Challenges and Applications. In: IEEE 5th World Forum on Internet of Things, WF-IoT 2019 - Conference Proceedings. 2019 Apr;707–10. doi:10.1109/WF-IOT.2019.8767250.].

Offloading cryptographic operations to an external API is necessary due to the resource constraints of the ESP32, which has restricted RAM (~520 KB) and limited CPU headroom, making it less suitable for handling complex key exchanges and real-time encryption of continuous sensor data. The Encryption API will be hosted on the gateway device (e.g., Raspberry Pi), which has sufficient processing capability to manage cryptographic workloads while maintaining low-latency communication with sensor nodes [2222Setiadi dRIM, Najib AF, Rachmawanto EH, Sari CA, Sarker MK, Rijati N. A comparative study MD5 and SHA1 algorithms to encrypt REST API authentication on mobile-based application. In: 2019 International Conference on Information and Communications Technology, ICOIACT 2019. 2019 Jul;206–11. doi:10.1109/ICOIACT46704.2019.8938570.,2424Dey S, Bera T. Design and Development of a Smart and Multipurpose IoT Embedded System Device Using ESP32 Microcontroller. In: IEEE International Conference on Electrical, Electronics, Communication and Computers, ELEXCOM 2023. 2023. doi:10.1109/ELEXCOM58812.2023.10370327.]

The API will also handle key management strategies, including secure key generation, periodic key rotation, and secure key storage (either in encrypted files or via integration with hardware secure modules if available). Each ESP32 device will be assigned a unique identifier and access token to authenticate requests to the API, ensuring that only authorized nodes can request encryption services. This centralized security approach not only reduces the attack surface on edge nodes but also simplifies updates, revocations, and audits [2020Yusof MK, Man M, Wan Hamzah WMAF, Safei S, Tasani MH, Shah NAAA. A Secure Model Based on Symmetric Encryption for REST API in Data Integration. In: 27th International Computer Science and Engineering Conference 2023, ICSEC 2023. 2023;402–6. doi:10.1109/ICSEC59635.2023.10329720.]

These strategies align with best practices in IoT security and are expected to enhance the overall confidentiality, integrity, and availability of the temperature monitoring system while maintaining high operational efficiency.

This architecture achieves a well-balanced combination of performance, security, and flexibility, supporting deployments ranging from small-scale local networks to enterprise-grade distributed solutions. Such versatility is particularly important in temperature monitoring, where operational requirements vary across industrial, medical, and logistics environments.

Critical assessment of emerging technologies

Recent advancements in IoT-based temperature monitoring have significantly improved metric accuracy, network scalability, and energy efficiency. However, a critical analysis reveals persistent gaps that represent both technical challenges and opportunities for innovation.

Among digital thermal sensors (Figure 2), the Sensirion SHT31 stands out for its ±0.2 °C accuracy and long-term stability, ideal for scientific and medical use. The DS18B20 remains relevant for its simple 1-Wire interface and robustness in industrial environments, despite slower response times and lower accuracy. The Bosch BME680 offers a comprehensive solution by integrating temperature, humidity, barometric pressure, and air quality (VOCs) sensing in a low-power, ESP32-compatible package [2525Jose J, Sasipraba T. Indoor air quality monitors using IOT sensors and LPWAN. In: Proceedings of the International Conference on Trends in Electronics and Informatics, ICOEI 2019. 2019 Apr;633–7. doi:10.1109/ICOEI.2019.8862647.] making it the preferred option for real-time environmental monitoring.

Figure 2: Comparative radar chart of three environmental sensors: BME680, SHT31, and DS18B20. The evaluation is based on five technical criteria—accuracy, range, power consumption, cost, and integration. While the BME680 excels in accuracy, the DS18B20 shows strong performance in cost-efficiency and ease of integration. The SHT31 offers a balanced performance across most categories, with superior range capabilities.

Figure 3 presents a comparative analysis between different communication technologies, evaluating them according to six main dimensions: coverage, bandwidth, power consumption, cost, scalability, and overall suitability for edge computing environments. In summary, ESP-MESH enables decentralized, self-healing networks with extensive coverage and no single point of failure. LoRaWAN offers long-range, low-power communication but lacks data throughput for frequent updates. Wi-Fi and MQTT remain relevant for their cloud integration, but are energy-intensive [1717Ali AI, Partal SZ, Kepke S, Partal HP. ZigBee and LoRa based Wireless Sensors for Smart Environment and IoT Applications. In: Proceedings - 2019 IEEE 1st Global Power, Energy and Communication Conference, GPECOM 2019. 2019 Jun;19–23. doi:10.1109/GPECOM.2019.8778505.].

Figure 3: Radar chart illustrating a comparative analysis of IoT communication technologies, including Wi-Fi (802.11 b/g/n), BLE, ESP-MESH, LoRaWAN, Zigbee, and Ethernet. The comparison evaluates six dimensions: coverage, bandwidth, power consumption, cost, scalability, and overall suitability for edge computing environments. LoRaWAN stands out in coverage and low power consumption, while Ethernet and Wi-Fi provide superior bandwidth. ESP-MESH offers a balanced compromise between scalability and performance in distributed networks.

Figure 4 provides a comparative analysis of five edge computing platforms, evaluated across six technical dimensions: computing power, power efficiency, GPIO availability, connectivity, AI support, and price. The comparison suggests that combining ESP32 (low-power data collection) with Raspberry Pi gateways (local preprocessing and buffering) creates a balanced distributed architecture [11Kutseva M. Adaptation of Seven-Layered IoT Architecture for Energy Efficiency Management in Smart House. In: 2022 10th International Scientific Conference on Computer Science, COMSCI 2022 - Proceedings. 2022. doi:10.1109/COMSCI55378.2022.9912604.,2626Naveen S, Kounte MR. Key Technologies and challenges in IoT Edge Computing. In: Proceedings of the 3rd International Conference on I-SMAC IoT in Social, Mobile, Analytics and Cloud, I-SMAC 2019. 2019 Dec;61–5. doi:10.1109/I-SMAC47947.2019.9032541.]. Docker containerization improves modularity, remote management, and update mechanisms [2727Tansangworn N. Development of IoT Edge Hub for Wireless Sensor Networks based on Docker Container. In: Proceedings - 2020 IEEE International Conference on Smart Internet of Things, SmartIoT 2020. 2020 Aug;356–7. doi:10.1109/SMARTIOT49966.2020.00068.].

Figure 4: Radar chart comparing five edge computing platforms—ESP32, ESP8266, STM32, Raspberry Pi 4, and Jetson Nano—across six technical dimensions: computing power, power efficiency, GPIO availability, connectivity, AI support, and price. Raspberry Pi 4 and Jetson Nano stand out for computational capabilities and AI support, while ESP8266 and ESP32 offer excellent power efficiency and cost-effectiveness for lightweight IoT applications.

A hybrid visualization system includes native mobile apps (Kotlin for Android, Swift for iOS) and a Next.js web dashboard. Secure REST APIs with OAuth 2.0 and TLS 1.3 ensure reliable 15-second updates [2121Garg H, Dave M. Securing IoT Devices and Securely Connecting the Dots Using REST API and Middleware. In: Proceedings - 2019 4th International Conference on Internet of Things: Smart Innovation and Usages, IoT-SIU 2019. 2019 Apr. doi:10.1109/IOT-SIU.2019.8777334.,2828Kamarudin NH, Chung WS, Madon B, Khmag A. Internet of Things (IoT) with Mobile Educational Apps: A Review. In: ICETAS 2019 - 2019 6th IEEE International Conference on Engineering, Technologies and Applied Sciences. 2019 Dec. doi:10.1109/ICETAS48360.2019.9117528.]. Advanced analytics, real-time maps, anomaly detection, and remote configuration are supported via MQTT (QoS 1), WebSockets, and GraphQL APIs [1414Kashyap M, Sharma V, Verma G. Implementation and Analysis of IoT Based Automation Using MQTT. In: 2021 IEEE 4th International Conference on Computing, Power and Communication Technologies, GUCON 2021. 2021 Sep. doi:10.1109/GUCON50781.2021.9573857.]. Data persistence is handled by MongoDB due to its flexible schema, write performance, and native replication [2929Mladenova T, Valova I. Performance Study of MySQL and MongoDB for IoT Data Processing and Storage. In: International Conference Automatics and Informatics, ICAI 2022 - Proceedings. 2022;60–3. doi:10.1109/ICAI55857.2022.9960134.].

Challenges remain in cross-vendor interoperability, energy autonomy in harsh environments, and mesh network cybersecurity. Promising research includes federated learning for distributed sensor calibration and lightweight neural networks (tinyML) for local anomaly detection [3030Yamakami T. A gap analysis framework of IoT-empowered city platform as a service. In: 14th International Conference on Services Systems and Services Management, ICSSSM 2017 - Proceedings. 2017 Jul. doi:10.1109/ICSSSM.2017.7996135.].

ESP32 development can be done via Arduino (C/C++) for high-performance, memory-constrained applications, or MicroPython for rapid prototyping and interactive development [3131Singhvi RK, Lohar RL, Kumar A, Sharma R, Sharma LD, Saraswat RK. IoT Based Smart Waste Management System: India prospective. In: Proceedings - 2019 4th International Conference on Internet of Things: Smart Innovation and Usages, IoT-SIU 2019. 2019 Apr. doi:10.1109/IOT-SIU.2019.8777698.,3232Gaspar G, Fabo P, Kuba M, Flochova J, Dudak J, Florkova Z. Development of IoT applications based on the MicroPython platform for Industry 4.0 implementation. In: Proceedings of the 2020 19th International Conference on Mechatronics - Mechatronika, ME 2020. 2020 Dec. doi:10.1109/ME49197.2020.9286455.]. Each approach offers trade-offs between control and ease of use.

Prototype validation

To evaluate the feasibility and performance of the proposed IoT-based temperature monitoring system, a preliminary prototype validation was conducted under controlled conditions. The test environment consisted of a closed indoor space with ambient temperature ranging from 18 °C to 27 °C, relative humidity between 40% and 80%. Environmental sensors such as the BME680 and LDR were used to monitor temperature and ambient light, respectively.

The experimental setup included a total of two sensor nodes deployed over approximately 10 meters from the central gateway (a Raspberry Pi 4) in two opposite orientations. Each node transmitted temperature data at fixed 30-second intervals over a continuous 24-hour period. All transmitted data was received via MQTT.

This early-stage validation confirms the prototype’s ability to perform accurate and low-latency environmental monitoring with reasonable power efficiency. The results align with similar studies in the field and provide a basis for further optimization and scaling in more demanding environments. Future tests will extend to outdoor deployments and higher node densities to assess performance under more variable conditions.

Results

The system demonstrates high reliability in data transmission, low latency due to edge computing, and scalability across different applications. The security measures, including encryption in both ESP Mesh communication and gateway-to-API transmission, ensure data integrity and confidentiality.

Limitations

Although the proposed system shows promising results, several limitations must be acknowledged. In terms of scalability, the current prototype was tested with only two sensor nodes in an indoor environment with limited spatial distribution. While ESP-MESH supports the creation of larger and more complex mesh networks, increasing the number of nodes introduces challenges such as increased propagation delay, routing instability, and greater energy consumption. Future work should assess the performance of the system under larger-scale deployments with wider coverage areas and higher node densities to validate its scalability in real-world applications.

From a cybersecurity standpoint, mesh-based communication architectures such as ESP-MESH expose additional attack surfaces. If not properly secured, unauthorized devices might join the mesh and potentially intercept or alter data flows. Furthermore, lightweight communication protocols such as MQTT, while efficient, may be vulnerable to spoofing or man-in-the-middle attacks if encryption and authentication are not rigorously implemented. The proposed architecture mitigates these concerns through an Encryption API that centralizes key management and cryptographic operations, but practical deployment scenarios will require robust validation of these security mechanisms.

In addition, environmental constraints can impact the long-term reliability of the system. Sensors like the BME680 and LDR may experience performance degradation or calibration drift when subjected to extreme temperature fluctuations, high humidity, dust exposure, or electromagnetic interference. These factors can lead to reduced accuracy over time, especially in outdoor or industrial settings. To address this, future versions of the system should incorporate regular calibration routines, protective enclosures for hardware, and compensation algorithms to correct environmental bias and sensor aging.

Discussion

The solution proposed in this work is currently under development and has been partially validated through the design and testing of a functional prototype. Preliminary results suggest that low-power microcontrollers such as the ESP32, when combined with decentralized communication models like ESP-MESH and lightweight protocols such as MQTT, can provide an effective foundation for real-time environmental data acquisition. The use of a Raspberry Pi as an edge gateway enables local data buffering and processing, as well as secure transmission to a backend API developed in Go, with data stored in a MongoDB database (Figure 1).

The architecture’s modular and containerized design facilitates flexibility, scalability, and simplified deployment, which are key advantages in scenarios such as agriculture, logistics, and smart infrastructure. These early insights support the findings in the literature review, which indicate that this approach helps overcome limitations of traditional monitoring systems—namely, high cost, poor energy efficiency, and limited interoperability—while enhancing data reliability and operational robustness.

To further enhance system intelligence and autonomy, a concrete pipeline for integrating TinyML is proposed [3333Yuan Z, Law KLE. Distributed TinyML on Resource-Constrained IoT Sensor Networks. In: 2024 IEEE 10th World Forum on Internet of Things, WF-IoT 2024. 2024;457–62. doi:10.1109/WF-IOT62078.2024.10811277.]. In this approach, sensor nodes equipped with ESP32 microcontrollers would perform local data analysis using lightweight models deployed with TensorFlow Lite for Microcontrollers. The typical data flow begins with temperature readings processed on-device to detect anomalies based on predefined thresholds [3434Dokic K, Martinovic M, Mandusic D. Inference speed and quantisation of neural networks with TensorFlow Lite for Microcontrollers framework. In: SEEDA-CECNSM 2020 - 5th South-East Europe Design Automation, Computer Engineering, Computer Networks and Social Media Conference. 2020 Sep. doi:10.1109/SEEDA-CECNSM49515.2020.9221846.]. When anomalies are identified, alerts can be raised locally and transmitted to the gateway. For continuous learning and improvement, nodes participate in federated learning cycles by updating model weights based on locally observed data. These updates are transmitted securely to the edge gateway using the MQTT protocol. The gateway aggregates updates, refines the global model, and redistributes it across the network. This method ensures privacy by avoiding the transmission of raw sensor data while enabling the system to adapt to localized environmental patterns.

Interoperability is also a critical consideration in the system design. The solution leverages OASIS MQTT, a widely adopted standard for IoT messaging that ensures compatibility with multiple platforms and vendors. At the network layer, communication protocols such as IEEE 802.15.4—the basis for Zigbee and Thread—and ESP-MESH are used to support device-to-device communication in heterogeneous deployments. Data formatting follows standard practices using lightweight JSON structures, and all API interfaces are REST-compliant, promoting seamless integration with cloud platforms and third-party services. These decisions aim to ensure that the system remains extensible and interoperable in diverse technological ecosystems [3535IEEE Draft Standard for a Smart Transducer Interface for Sensors, Actuators, and Devices - Message Queue Telemetry Transport for Networked Device Communication. P1451.1.6/D02, Sept 2024. 2025.].

Conclusion

The evolution of temperature monitoring systems has demonstrated that IoT integration significantly enhances precision and efficiency. This project proposes a comprehensive architecture combining state-of-the-art technologies tailored for continuous thermal monitoring.

The core solution employs the BME680 sensor, offering accurate temperature measurements (±0.5 °C) alongside humidity, pressure, and air quality data, surpassing alternatives such as the SHT31 and DS18B20 in terms of multifunctionality. A hybrid communication model leverages ESP-MESH for self-organizing sensor networks and MQTT over TLS for secure cloud transmission, ensuring scalability and energy efficiency compared to conventional Wi-Fi or LoRaWAN approaches.

For data processing, MicroPython enables rapid prototyping on ESP32 nodes, while the Go (Golang) backend ensures high-performance concurrent data handling. MongoDB provides flexible time-series storage, and an interactive Next.js portal with Kotlin/Swift mobile apps delivers real-time thermal data visualization.

Key innovations address persistent challenges: optimized energy use via ESP32 Deep Sleep algorithms, end-to-end encryption, and edge computing to reduce cloud dependency. This architecture balances precision, scalability, and reliability, supporting deployments ranging from small-scale to enterprise systems while overcoming conventional limitations.

Conflict of interest

The authors declare no conflict of interest, financial or otherwise, related to the development or publication of this article.

Acknowledgement

The authors would like to acknowledge the support of the Polytechnic Institute of Coimbra – ISEC. No external funding or grants were used in the development of this work.

Author contributions

José Nabão: Conceptualization, system architecture design, investigation, data curation, sensor selection, electronics integration, development of backend API, integration of containers and orchestration, and original draft preparation.

Acácio M. R. Amaral: Conceptualization, methodology, investigation, resources, review and editing, supervision, project administration, and funding acquisition.

Filipe Sá: Conceptualization, methodology, investigation, resources, review and editing, project administration, and funding acquisition.

Ethics and funding

This study did not involve any experiments with human or animal subjects and therefore did not require ethical approval. All development and testing were conducted in compliance with institutional guidelines for academic and technological research.

No external funding was received for this work. The project was self-funded and partially supported through academic and professional resources provided by the Polytechnic Institute of Coimbra – ISEC.

References

1. Kutseva M. Adaptation of Seven-Layered IoT Architecture for Energy Efficiency Management in Smart House. In: 2022 10th International Scientific Conference on Computer Science, COMSCI 2022 - Proceedings. 2022. doi:10.1109/COMSCI55378.2022.9912604.

2. Three-layered software architecture and its variability for teleoperated system. IEEE Xplore; 2025 Apr 22[Online]. Available from: https://ieeexplore.ieee.org/document/7347791

3. Sarkar S, Akshatha KS, Saurabh A, Samanvitha B, Sarwar MF. IoT Enabled Cold Supply Chain Monitoring System. In: 2022 IEEE 3rd Global Conference for Advancement in Technology, GCAT 2022. 2022. doi:10.1109/GCAT55367.2022.9972137.

4. Al-Awami SH, Al-Aty MM, Al-Najar MF. Comparison of IoT Architectures Based on the Seven Essential Characteristics. In: Proceedings - 2023 IEEE 3rd International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and Computer Engineering, MI-STA 2023. 2023;305–10. doi:10.1109/MI-STA57575.2023.10169289.

5. Al-Sammak KA, Al-Gburi SH, Marghescu C, Dragulinescu AM, Suciu G, Abdulqader AG. A Comprehensive Assessment of LoRaWAN and NB-IoT Performance Metrics under Varied Payload Data Sizes. In: Proceedings of the 16th International Conference on Electronics, Computers and Artificial Intelligence, ECAI 2024. 2024. doi:10.1109/ECAI61503.2024.10607481.

6. Eridani D, Rochim AF, Cesara FN. Comparative Performance Study of ESP-NOW, Wi-Fi, Bluetooth Protocols based on Range, Transmission Speed, Latency, Energy Usage and Barrier Resistance. In: Proceedings - 2021 International Seminar on Application for Technology of Information and Communication, iSemantic 2021. 2021 Sep;322–8. doi:10.1109/ISEMANTIC52711.2021.9573246.

7. Kassim MRM. Applications of IoT and Blockchain in Smart Agriculture: Architectures and Challenges. In: 2022 IEEE International Conference on Computing, ICOCO 2022. 2022;253–8. doi:10.1109/ICOCO56118.2022.10031697.

8. Gül F, Eroğlu H. Low-Cost IoT Mesh Network for Real-Time Indoor Air Quality Monitoring. In: Proceedings of the 9th International Conference on Communication and Electronics Systems, ICCES 2024. 2024;502–7. doi:10.1109/ICCES63552.2024.10860109.

9. Sehrawat D, Gill NS. Smart sensors: Analysis of different types of IoT sensors. In: Proceedings of the International Conference on Trends in Electronics and Informatics, ICOEI 2019. 2019 Apr;523–8. doi:10.1109/ICOEI.2019.8862778.

10. Garg R, Kumar A, Singh S, Dayana R. Advanced Air Quality Monitoring System using IoT and Sensor Technology. In: 3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025. 2025;687–92. doi:10.1109/IDCIOT64235.2025.10915106.

11. Ahmad YA, Gunawan TS, Mansor H, Hamida BA, Hishamudin AF, Arifin F. On the Evaluation of DHT22 Temperature Sensor for IoT Application. In: Proceedings of the 8th International Conference on Computer and Communication Engineering, ICCCE 2021. 2021 Jun;131–4. doi:10.1109/ICCCE50029.2021.9467147.

12. Maier A, Sharp A, Vagapov Y. Comparative analysis and practical implementation of the ESP32 microcontroller module for the internet of things. In: 2017 Internet Technologies and Applications, ITA 2017 - Proceedings of the 7th International Conference. 2017 Nov;143–8. doi:10.1109/ITECHA.2017.8101926.

13. Yacchirema D, Palau C. Internet of things Interoperability: A smart IoT gateway approach. In: Iberian Conference on Information Systems and Technologies, CISTI. 2023 Jun. doi:10.23919/CISTI58278.2023.10211617.

14. Kashyap M, Sharma V, Verma G. Implementation and Analysis of IoT Based Automation Using MQTT. In: 2021 IEEE 4th International Conference on Computing, Power and Communication Technologies, GUCON 2021. 2021 Sep. doi:10.1109/GUCON50781.2021.9573857.

15. Tondato De Faria B, Tercete GM, Filev Maia R. The effectiveness of IoT and machine learning in Precision Agriculture. In: 2022 Symposium on Internet of Things, SIoT 2022. 2022. doi:10.1109/SIOT56383.2022.10070308.

16. Dzahir MASM, Chia KS. Evaluating the Energy Consumption of ESP32 Microcontroller for Real-Time MQTT IoT-Based Monitoring System. In: 2023 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies, 3ICT 2023. 2023;255–61. doi:10.1109/3ICT60104.2023.10391358.

17. Ali AI, Partal SZ, Kepke S, Partal HP. ZigBee and LoRa based Wireless Sensors for Smart Environment and IoT Applications. In: Proceedings - 2019 IEEE 1st Global Power, Energy and Communication Conference, GPECOM 2019. 2019 Jun;19–23. doi:10.1109/GPECOM.2019.8778505.

18. Mekki K, Bajic E, Meyer F. Indoor positioning system for IoT device based on BLE technology and MQTT protocol. In: IEEE 5th World Forum on Internet of Things, WF-IoT 2019 - Conference Proceedings. 2019 Apr;787–92. doi:10.1109/WF-IOT.2019.8767287.

19. Aziz ANAA, Rashid RA, Nasir MNM, Sarijari MA. ESP Mesh Network for Security Application. In: Conference Proceedings - IEEE International Conference on Advanced Telecommunication and Networking Technologies, ATNT 2024. 2024. doi:10.1109/ATNT61688.2024.10719207.

20. Yusof MK, Man M, Wan Hamzah WMAF, Safei S, Tasani MH, Shah NAAA. A Secure Model Based on Symmetric Encryption for REST API in Data Integration. In: 27th International Computer Science and Engineering Conference 2023, ICSEC 2023. 2023;402–6. doi:10.1109/ICSEC59635.2023.10329720.

21. Garg H, Dave M. Securing IoT Devices and Securely Connecting the Dots Using REST API and Middleware. In: Proceedings - 2019 4th International Conference on Internet of Things: Smart Innovation and Usages, IoT-SIU 2019. 2019 Apr. doi:10.1109/IOT-SIU.2019.8777334.

22. Setiadi dRIM, Najib AF, Rachmawanto EH, Sari CA, Sarker MK, Rijati N. A comparative study MD5 and SHA1 algorithms to encrypt REST API authentication on mobile-based application. In: 2019 International Conference on Information and Communications Technology, ICOIACT 2019. 2019 Jul;206–11. doi:10.1109/ICOIACT46704.2019.8938570.

23. Gunathilake NA, Buchanan WJ, Asif R. Next Generation Lightweight Cryptography for Smart IoT Devices: Implementation, Challenges and Applications. In: IEEE 5th World Forum on Internet of Things, WF-IoT 2019 - Conference Proceedings. 2019 Apr;707–10. doi:10.1109/WF-IOT.2019.8767250.

24. Dey S, Bera T. Design and Development of a Smart and Multipurpose IoT Embedded System Device Using ESP32 Microcontroller. In: IEEE International Conference on Electrical, Electronics, Communication and Computers, ELEXCOM 2023. 2023. doi:10.1109/ELEXCOM58812.2023.10370327.

25. Jose J, Sasipraba T. Indoor air quality monitors using IOT sensors and LPWAN. In: Proceedings of the International Conference on Trends in Electronics and Informatics, ICOEI 2019. 2019 Apr;633–7. doi:10.1109/ICOEI.2019.8862647.

26. Naveen S, Kounte MR. Key Technologies and challenges in IoT Edge Computing. In: Proceedings of the 3rd International Conference on I-SMAC IoT in Social, Mobile, Analytics and Cloud, I-SMAC 2019. 2019 Dec;61–5. doi:10.1109/I-SMAC47947.2019.9032541.

27. Tansangworn N. Development of IoT Edge Hub for Wireless Sensor Networks based on Docker Container. In: Proceedings - 2020 IEEE International Conference on Smart Internet of Things, SmartIoT 2020. 2020 Aug;356–7. doi:10.1109/SMARTIOT49966.2020.00068.

28. Kamarudin NH, Chung WS, Madon B, Khmag A. Internet of Things (IoT) with Mobile Educational Apps: A Review. In: ICETAS 2019 - 2019 6th IEEE International Conference on Engineering, Technologies and Applied Sciences. 2019 Dec. doi:10.1109/ICETAS48360.2019.9117528.

29. Mladenova T, Valova I. Performance Study of MySQL and MongoDB for IoT Data Processing and Storage. In: International Conference Automatics and Informatics, ICAI 2022 - Proceedings. 2022;60–3. doi:10.1109/ICAI55857.2022.9960134.

30. Yamakami T. A gap analysis framework of IoT-empowered city platform as a service. In: 14th International Conference on Services Systems and Services Management, ICSSSM 2017 - Proceedings. 2017 Jul. doi:10.1109/ICSSSM.2017.7996135.

31. Singhvi RK, Lohar RL, Kumar A, Sharma R, Sharma LD, Saraswat RK. IoT Based Smart Waste Management System: India prospective. In: Proceedings - 2019 4th International Conference on Internet of Things: Smart Innovation and Usages, IoT-SIU 2019. 2019 Apr. doi:10.1109/IOT-SIU.2019.8777698.

32. Gaspar G, Fabo P, Kuba M, Flochova J, Dudak J, Florkova Z. Development of IoT applications based on the MicroPython platform for Industry 4.0 implementation. In: Proceedings of the 2020 19th International Conference on Mechatronics - Mechatronika, ME 2020. 2020 Dec. doi:10.1109/ME49197.2020.9286455.

33. Yuan Z, Law KLE. Distributed TinyML on Resource-Constrained IoT Sensor Networks. In: 2024 IEEE 10th World Forum on Internet of Things, WF-IoT 2024. 2024;457–62. doi:10.1109/WF-IOT62078.2024.10811277.

34. Dokic K, Martinovic M, Mandusic D. Inference speed and quantisation of neural networks with TensorFlow Lite for Microcontrollers framework. In: SEEDA-CECNSM 2020 - 5th South-East Europe Design Automation, Computer Engineering, Computer Networks and Social Media Conference. 2020 Sep. doi:10.1109/SEEDA-CECNSM49515.2020.9221846.

35. IEEE Draft Standard for a Smart Transducer Interface for Sensors, Actuators, and Devices - Message Queue Telemetry Transport for Networked Device Communication. P1451.1.6/D02, Sept 2024. 2025.

About the Article

Cite this Article

Nabão J, Amaral AMR, Sá F. IoT-based Real-time Temperature Monitoring in Critical Systems: A Review. IgMin Res. May 15, 2025; 3(5): 226-234. IgMin ID: igmin303; DOI:10.61927/igmin303; Available at: igmin.link/p303

• DOI10.61927/igmin303

25 Apr, 2025

Received

14 May, 2025

Accepted

15 May, 2025

Published

Share this Article

Anyone you share the following link with will be able to read this content:

Get shareable link

Topics

Biomedical Engineering