System Design And Implementation Of Energy Harvesting Wireless Sensor Network

In recent years,the rapid development of electronics,communication and embedded technology has promoted the development of Energy Harvesting Wireless Sensor Network(EH-WSN),which is mostly converted to solar power and has significant advantages for outdoor environment monitoring due to the high density of solar energy.EH-WSNs have many nodes and need to introduce efficient node scheduling algorithms,such as charging prediction algorithms and energy-aware routing strategies,in order to achieve maintenance-free and long-term stable operation.However,in practical applications,the actual deployment of existing node scheduling algorithms is not very satisfactory due to factors such as weather changes and the geographical environment where the nodes are deployed.Analyzing the log information,we get the following two reasons:(1)The existing charging prediction algorithm only relies on uniform environmental change characteristics to predict the charging of nodes,ignoring the influence of the individual deployment environment of nodes whose light is blocked by shadows on the charging,which makes the charging prediction less accurate.(2)There is room to improve the environmental adaptability of existing energy-aware routing strategies.The nodes' charging energy changes in real time with the lighting conditions,and the existing energy-aware routing strategies ignore the differences in routing policy claims under different lighting conditions,resulting in high node mortality and serious network fragmentation in the case of darkness or long overcast days.In order to solve the above problems,this paper intends to investigate both charging prediction and energy-aware routing,and propose an optimization strategy that can allow EH-WSN to perform low-cost maintenance and long-term stable operation.The main work of this paper is as follows.To address the problem of degraded accuracy of existing charging prediction algorithms when actually deployed for applications in outdoor environments(problem(1)above),this paper combines geographic and geometric theories to propose an individual feature-based charging prediction adaptation strategy(Shadow Judgment Method,SJM),and,introduces a step-by-step prediction approach,first,without considering shadow occlusion Firstly,the charging situation of small sensors without shadow occlusion is predicted based on the existing charging prediction algorithm.Then,based on the geographical location of each node,SJM is used to calculate the time range of the node being shaded and to correct the predicted value of the node's charging energy.The experimental results show that SJM can improve the prediction accuracy by 63.4%?67.5%.To address the problem of improving the adaptability of energy-aware routing strategies to the environment(problem(2)above),this paper proposes a routing strategy with adjustable optimization objectives based on convex optimization(Solar-aware Routing Strategy,SAR),which considers the differences in routing strategy claims under different light intensities.When the light is good,the optimization objective is to balance the energy consumption of the network;when the light condition is poor,the optimization objective is to save energy and allow some nodes to die temporarily to extend the working time of the network as a whole.The experimental results show that SAR can effectively reduce the number of topological oscillations of EH-WSN and extend the network lifetime.