Research On Energy Balance Routing Algorithm Based On Link Quality Estimation

The Wireless Sensor Network is composed of a large number of sensor nodes deployed in the monitoring area,and the sensor nodes complete the monitoring task through the perception of the surrounding environment.Sensor nodes in WSN are usually deployed in harsh environments,and the communication between nodes is easily affected by environmental factors,resulting in a decline in communication quality.Effective link quality assessment methods can not only objectively evaluate the link quality in the network,but also provide a basis for the upper-layer routing protocol to select a suitable link for data transmission,ensure the quality of data transmission,and effectively reduce the energy consumption of sensor nodes and prolong the network life cycle.This paper first proposes a link quality assessment method,establishes a link quality assessment model,and then designs a link quality-based clustering algorithm LQECRP to complete clustering using the link quality estimation value to achieve efficient data transmission in the network.It is mainly reflected in the following aspects:(1)To address the problem that the link state in WSN is complex and not easy to evaluate,this paper proposes a link quality evaluation method based on Light GBM,which integrates hardware layer parameters such as received signal strength indicator,signal-to-noise ratio,and link quality indicator as input parameters for the link quality evaluation model.It seeks the optimal split point for features,generates classification and regression trees using a depth-constrained leaf-wise growth strategy,and calculates the weighted sum of the generated regression trees in each round to obtain the final training model.Finally,the grid search algorithm is used to optimize the hyperparameters of the model to improve the performance of the evaluation model.Using the accuracy rate,precision rate,recall rate and F1 evaluation indicators,the model in this paper is compared with the Support Vector Machine model,Gradient Boosting Decision Tree model and Wavelet Neural Network models for comparative analysis.Experimental results show that the link quality assessment method proposed in this paper has higher accuracy and reliability than other models.(2)To address the problem of uneven network energy consumption in the traditional cluster routing algorithm,this paper proposes the LQECRP algorithm,which determines the optimal number of cluster heads of the network based on the first-order radio model in the cluster formation phase.During the cluster head election,member nodes calculate their own cluster head election parameters based on their link quality,remaining energy,and distance information to the current cluster head node.The current cluster head node selects the member node with the maximum cluster head election parameter as the next round’s cluster head node.Experiments show that the LQECRP algorithm can balance the energy consumption of nodes in the network and effectively prolong the life cycle of the network.(3)For the problem of unstable data transmission in WSNs,this paper proposes an adaptive power transmission strategy,in which the cluster head node communicates directly with the base station if the distance between the cluster head node and the base station is less than a threshold value set in advance during the data transmission phase,and vice versa,multi-hop transmission is required.When multi-hop transmission is required,the distance between the cluster head node and the relay node,the remaining energy of the node and the link quality are used as the selection reference of the relay node to ensure the quality of data transmission.Experimental results show that the algorithm proposed in this paper can guarantee the quality of data transmission in the network and improve the throughput of the network.