Adaptive Dynamic Programming for Multi-Point Scheduling in Energy Harvesting Wireless Sensor Networks

With the development of energy harvesting technologies, the building of wireless sensor networks (WSNs) based energy harvesting has become possible, and helps to improve the limitation of battery energy in WSNs. However, the energy harvesting rate of nodes depends on the external environment, with the uncertainty and uncontrollability in terms of time and space, which brings a huge challenge for resource management and optimization of the entire network. Based on a 3-layer BP neural network method, this paper designed an energy acquisition model based on artificial neural networks, used the Kalman filter to estimate and correction the state of the system, then put forward the ADP algorithm for multi-point scheduling of WSNs to maximize the network output. Finally, the analysis showed that the proposed algorithm is effective and feasible.