Link Quality Estimation Base on CatBoost

To estimate link quality for wireless sensor networks (WSN) accurately and rapidly, an approach of link quality estimation is proposed, which is based on Category Boosting (CatBoost). Received signal strength indicator mean, link quality indicator mean and the signal to noise ratio mean are selected as the link quality parameters. The K-means++ algorithm optimized by elbow method is used to divide the link quality level as the estimation index. The link quality estimation model is constructed based on CatBoost, and the parameters of the model are optimized by grid search method. Experiments in three scenarios of laboratory, corridor and parking show that the proposed method has higher estimation accuracy than F-LQE, SVM, LFI-LQE and RF, and can effectively estimate the link quality of WSN.