Energy-efficient data compression and resource allocation for edge assisted IoT networks

Data compression has the potential to reduce energy consumption for data transmission. However, the optimal data compression ratio is in conjunction with the decision of resources allocation. In the paper, we jointly optimize data compression and resource allocation with the aim of minimizing the energy consumption while guaranteeing the latency requirements for edge assisted Internet of Thing (IoT) networks. Due to the dynamic task packet size and time-varying wireless communication environment, it is challenge to obtain the optimal policy with traditional optimization methods. Therefore, we propose a deep reinforcement learning framework to tackle the joint data compression and resource allocation problem for edge assisted IoT networks. Simulation results show that our proposed scheme has a good coverage performance and can significantly reduce the energy consumption compared to other baselines.