Research On Energy Prediction And Data Cooperative Transmission Of EH-WSN

Wireless sensor network(WSN)has many applications in military,scientific research,livelihood and other fields because of its small size,low power consumption and wide deployment range.In WSN,the sensor node is responsible for data collection and transmission,but the node size is generally small,the battery energy is not enough to maintain the long-term operation of the network.Moreover,the general deployment environment is so bad that the battery can not be replaced.Once these batteries are exhausted,the node cannot work,and eventually the node is forced to sleep and die.Therefore,it is urgent to prolong the life cycle of WSN and solve the problem of energy limitation.In this paper,energy harvesting wireless sensor network(EH-WSN)is studied to solve the self-powering problem in wireless sensor network.Aiming at the problems of poor energy prediction accuracy,low node survival rate and high energy consumption in the transmission process in the existing EH-WSN routing algorithms,this paper proposes ARIMA-LSTM combined prediction algorithm for solar energy,EH-DEEC clustering routing algorithm and Adaptive routing transmission protocol based on EH-DEEC.The specific research work is as follows:1)Aiming at the problem that the inaccuracy of existing energy prediction algorithms,a combined forecasting model based on Autoregressive Integrated Moving Average(ARIMA)and Long Short-Term Memory(LSTM)neural network is proposed.In this model,the historical solar radiation data recorded by the sensor are filtered by the ARIMA model,and then the residual values are incorporated into the LSTM neural network model to modify the final prediction results.The simulation results show that the ARIMA-LSTM combination forecasting model combines the advantages of linear and nonlinear models,and has a better prediction effect than a single model in the case of provided data.The simulation results show that the combined model reduces the error and improves the prediction accuracy,which is beneficial to be extended to the prediction of solar radiation.2)Aiming at the problem that the node survival rate of the classical Distributed EnergyEfficient Clustering scheme(DEEC)algorithm is too low,the ARIMA-LSTM combination prediction model is introduced to optimize the cluster head election model: a new EH-DEEC(Energy Harvesting DEEC)cluster head election model is established based on the four factors of node residual energy,energy collection,energy prediction and base station location.This increases the probability that the node closer to the base station and with large residual energy will be elected as the cluster head.Furthermore,in order to solve the problem of high energy consumption of heterogeneous EH-WSN transmission,an adaptive routing transmission algorithm based on EH-DEEC is proposed.First of all,the algorithm gives the proposition of using single-hop transmission,multi-hop transmission,single-hop cooperative transmission and multi-hop cooperative transmission selection,and proves the sufficient conditions for cluster head nodes to dynamically select transmission mode to achieve self-adaptation.Three lemmas are derived from the proved conditions,which are used as the premise of mode selection for routing transmission.On this basis,the proposed EH-DEEC cluster head election model and adaptive routing transmission algorithm are verified by simulation.Simulation results show that the EH-DEEC cluster head election model and the adaptive routing transmission algorithm can reduce the overall routing energy consumption and increase the throughput,thus prolonging the node life and system life cycle.