摘要
A link quality assessment model named BDCT-SVM based on support vector machine and decision tree is proposed for wireless sensor networks in this paper. The communication area is divided into effective region, transitional region and clear region by experiments in the first place. Then a 4 level link quality decision tree is made to have five grades of link quality. The radial basis function is chosen as the kernel function in BDCT-SVM model and the k-fold cross validation method is used to optimize parameters according to the CCI, RSSI and PRR value by statistics. Experimental results show that the BDCT-SVM model proposed in this paper is reasonable and more accurate.
- 出版日期2012
- 单位南昌航空大学