Research On Key Energy-Efficient Technologies In EH-WSN

Wireless sensor networks(WSN)are playing an important role in many fields such as military,scientific research,and people's livelihood.In WSN,the sensor node is responsible for collecting and transmitting data,however,because of the small size of the node,the battery energy in the node is not enough.In a short time,the energy of a node battery will be exhausted.Moreover,since the deployment environment of WSN is usually in the wild or other places where it is difficult to replace the empty battery,it is more urgent to solve the energy limitation problem in WSN.At present,many researches on this problem at home and abroad consider reducing energy consumption or making load balancing,such as node sleeping,energy hole avoidance,and moving SINK to extending WSN life.However,these methods do not fundamentally handle the energy limitation problem,and WSN will still die after a short period of time.In addition,due to the lack of consideration of network performance,these methods are likely to cause network time delays and throughput degradation.Energy Harvesting-WSN(EH-WSN)was proposed as a new solution to completely solve the energy limitation problem.There are many renewable energy sources in the environment around WSN,such as solar energy,geothermal energy and wind energy.Solar energy is used as a major supplementary energy source due to the availability of solar energy and the maturity of photoelectric conversion technologies.However,solar energy is unstable and vulnerable to change because of weather condition,and EH-WSN may still die.How to predict the solar light situation in order to properly allocate the collected energy,and reduce network energy consumption become two maj or problems.In this paper,a prediction algorithm based on variable coefficient adaptive filter is proposed.The historical data is used to predict the future energy collection,and the recent illumination changes are fully considered.Experiments show that the proposed method reduces the error by 13.2%compared with the constant coefficient adaptive filter algorithm and 25%less than the weather condition weighted moving average algorithm(WCMA).In addition,this paper also proposes an improved algorithm EH-DEEC based on RDEEC clustering algorithm,which is more suitable in energy harvesting environment.The experimental results show that EH-DEEC has 17.1%more live nodes than RDEEC in the 230th round.