Arko Datta
ABSTRACT
This paper explores the feasibility of using Tiny Machine Learning (TinyML) to develop a compact system for inferring environmental pollution levels at a device’s location. Leveraging the correlation between the Air Quality Index (AQI), the localised temperature and humidity, and various other meteorological and temporal factors, we propose a first-of-its-kind device that tries to embed some intelligence into a mobile air quality sensing device that predicts the local AQI using the local temperature and humidity collected through sensors, and internet-sourced meteorological and temporal information collected through a web crawler. A dust particle sensor has been added to calculate the real AQI, for validating the inference. Our contributions culminate in an efficient C/C++ based XGBoost implementation within a 2MB memory constraint, achieving 75.2% accuracy with a 1615μs latency.
- Pedro Andrade, Ivanovitch Silva, Marianne Silva, Thommas Flores, Jordão Cassiano, and Daniel G. Costa. 2022. A TinyML Soft-Sensor Approach for Low-Cost Detection and Monitoring of Vehicular Emissions. Sensors 22, 10 (May 2022), 3838.Google Scholar
Cross Ref
- Ji Lin, Wei-Ming Chen, Yujun Lin, John Cohn, Chuang Gan, and Song Han. 2020. MCUNet: Tiny Deep Learning on IoT Devices.Google Scholar
- Ji Lin, Ligeng Zhu, Wei-Ming Chen, Wei-Chen Wang, Chuang Gan, and Song Han. 2022. On-Device Training Under 256KB Memory.Google Scholar
- Ali Moin, Andy Zhou, Abbas Rahimi, Alisha Menon, Simone Benatti, George Alexandrov, Senam Tamakloe, Jonathan Ting, Natasha Yamamoto, Yasser Khan, Fred Burghardt, Luca Benini, Ana C. Arias, and Jan M. Rabaey. 2020. A wearable biosensing system with in-sensor adaptive machine learning for hand gesture recognition. Nature Electronics 4, 1 (Dec. 2020), 54–63.Google ScholarCross Ref
- Prithviraj Pramanik, Prasenjit Karmakar, Praveen Kumar Sharma, Soumyajit Chatterjee, Abhijit Roy, Santanu Mandal, Subrata Nandi, Sandip Chakraborty, Mousumi Saha, and Sujoy Saha. 2023. AQuaMoHo: Localized Low-cost Outdoor Air Quality Sensing over a Thermo-hygrometer. ACM Transactions on Sensor Networks 19, 3 (March 2023), 1–30.Google ScholarDigital Library
- Praveen Kumar Sharma, Prasenjit Karmakar, Soumyajit Chatterjee, Abhijit Roy, Santanu Mandal, Sandip Chakraborty, Subrata Nandi, and Sujoy Saha. 2021. Can I go for a roof walk today?. In Proceedings of the 8th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation. ACM.Google ScholarDigital Library
Index Terms
- Real-Time Air Quality Predictions for Smart Cities using TinyML
Recommendations
Efficient Air Quality Index Prediction on Resource-Constrained Devices using TinyML: Design, Implementation, and Evaluation
ICDCN '24: Proceedings of the 25th International Conference on Distributed Computing and NetworkingThese days, mobile/handheld air quality monitoring devices primarily sense pollutants and send them to the cloud without performing intelligent processing, such as soft calibration, sensor fault detection, or any onboard predictions. This paper explores ...
Thermal Comfort and Indoor Air Quality of Task Ambient Air Conditioning in Modern Office Buildings
ICIII '09: Proceedings of the 2009 International Conference on Information Management, Innovation Management and Industrial Engineering - Volume 02Energy and environmental sustainability is a major global trend for the 21st century. Thermal discomfort and poor air quality in office buildings can result in loss of productivity, absenteeism and medical problems. The most frequently identified causes ...
Performance simulation of R410A air conditioning system with variable speeds
With the implementation of "Montreal protocol on substances that deplete the ozone layer", HCFCs especially R22 will be phased out. R410A (R32/R125,50/50wt%), as one alternative of R22, is a promising refrigerant for air conditioning system. Comparing ...
Comments