Adaptive Sensing for Smart Cities: A Reinforcement Learning Approach

Abstract

Urban sensing networks face the challenge of balancing data quality with energy consumption. We propose an adaptive sensing framework that uses reinforcement learning to dynamically adjust sampling rates based on environmental conditions and application requirements. Our system reduces energy consumption by 60% while maintaining 95% of the data quality compared to fixed-rate sampling approaches.

Citation

@inproceedings{smith2023adaptive,
  title = {Adaptive Sensing for Smart Cities: A Reinforcement Learning Approach},
  author = {Smith, Jane and Garcia, Maria},
  booktitle = {Proceedings of the ACM Conference on Embedded Networked Sensor Systems (SenSys)},
  year = {2023},
  doi = {10.1145/sensys.2023.001}
}