Node Dynamic Priority Control Based On Wireless Sensor Network Congestion

At present, wireless sensor networks (WSN) is a short-range communications sensor network, as one of the fastest developing and most promising technology. However, due to some of its inherent characteristics, such as multi-hop and gathering way of communication, the poor ability of the WSN node itself in calculation and communication, and the limitation of the WSN node energy, those all cause that it is easy to occur the WSN congestion problem. The WSN congestion control aspect also has made a lot of research results, but these studies are not perfect, there are many shortcomings, particularly in the energy-friendly congestion control and anticipation of congestion. In this paper, based on previous studies,several aspects of congestion control technology was studied, specific work as follows:(1) Considering the low energy consumption of buffer queue length congestion detection method and the high accuracy of detection congestion method based on the buffer queue rate, I will combine these two methods for congestion detection. Only when the queue length exceeds a certain threshold, will we calculate the rate change of buffer and accurately determine whether congestion occurs.(2) Aimed at the shortcomings of some existing congestion control algorithm in terms of congestion anticipation, I come up with a congestion method named a congestion control method based on node dynamic priority(NDPCC). NDPCC doesn’t rely on congestion detection, and always follows the rule of the most prone to congestion node given the highest priority of sending data.It start running when the protocol stack starting, and can certainly prejudge and treat the congestion.(3) For the continuous network congestion problem beyond NDPCC processing capacity, this paper presents a combined NDPCC and rate control method. When there is no congestion detected, it can use NDPCC congestion for prevention, once congestion occurs, immediately starting the rate control at the same time, it can eliminate congestion by adjusting the source node rate.Finally, with realizing and simulating the above algorithms on NS2, and comparing the performance of algorithm with CODA and null congestion control with the above algorithms, we verify that the NDPCC algorithm can reduce the dropout rate and delays and improve network throughput in congestion control..