基于BP神经网络的主动队列管理拥塞控制算法研究

ccording to wireless sensor network sink node important status and characteristics of wireless sensor network congestion This paper makes sink node bottleneck linear model and proposes a predictive controller of sink node queue length based on the BP neural network model. This controller can broadcast feedback information to source node to reduce transmitting rate when sink node queue length is over. Using NS-2 to simulating this algorithm and the existing congestion control algorithm, the results showed that our proposed algorithm is better than CODA algorithm on sink node queue length stability, end-to-end delay and delivery rate.