Congestion Control Algorithm For Wireless Sensor Networks

In recent years, with the rapid development of the Internet of things and people to focus research on sensor network, wireless sensor network (WSN) as the core technology of the Internet of things is also widely used and researched. The sensor node that limiting by energy, computing and network transmission bandwidth can not transmit or process data and brought the congestion problem, because of the wireless sensor network with low power consumption, low cost, small volume and short distance communication, and people for the demand is higher and higher. Network congestion causes transmission delay increase, the network throughput degradation, even appear lost package phenomenon, which not only has a great influence on the performance of the entire network, also will cause a great loss to users. Therefore, in order to better solve the problem of wireless sensor network congestion, study and research of network congestion control strategy is very important.Currently, The realization of the wireless sensor network congestion control strategy with congestion detection, congestion avoidance, congestion relieving have closely related. Congestion detection is the basis of congestion control strategy, which is based on the network running status to determine whether smooth, when happen congestion or whether the occurrence trend.the timeliness and accuracy of congestion detection has important effects on the performance of congestion control, most of the existing detection methods are based on buffer queue length, rate and channel sampling inspection; Congestion avoidance is a series of measures to avoid the network into congestion state, usually by reducing the rate or changing topology method; However, congestion relieving is that taking speed or flow control method to relieve congestion and make the network into the normal working condition. Aiming at the problem of congestion control in wireless sensor networks, the main research work is as follows:Firstly, this paper introduces the research background, in the background of the rapid development of the Internet,wireless sensor network has been widely used.the basis of a brief overview of the wireless sensor network (WSN), the characteristics, application, technology, research status, such as knowledge, the traditional network congestion control mechanism is not suitable for WSN, mainly introduced the current relatively popular congestion control algorithm in wireless sensor networks, such as CODV, STCP, COMUT, etc. Secondly, aiming at the clustering hierarchy model of congestion control problems in the cluster head, the article proposes a congestion control algorithm based on double cluster Heads (CCBDC), the main innovation of the algorithm are:In the congestion detection stage uses a detection method of buffer occupancy (a) and the congestion of the combination(C(n)) of the two parameters, and determine the occurrence of congestion; in congestion avoidance phase for different possible causes of cluster head congestion, respectively, using the shortest path number SPT and residual energy find the vice cluster head which shares data cluster heads transmit pressure, so as to avoid the occurrence of congestion. Through making simulation experiment on the NS2,the results show that the algorithm at network throughput, packet loss rate, time delay of data transmission have a better performance.Finally, this paper analyzes the minimum-rate based on congestion control in wireless sensor networks algorithm(MR). This algorithm in the variable topology structure has good stability, but there are some defects in the fairness allocation, such the upstream nodes to rate may have serious imbalance. Aiming at this problem, this paper presents a improved algorithm (Δx-MR) on ensuring the stability of algorithm. The original model increase Δx at the network convergence rate, that makes the subtree nodes of upstream node quite size of rate. The simulation experiments prove that, the improved algorithm improves fairness and throughput of the whole wireless sensor network, while ensuring the stability of the variable topology case,but also effectively improves the packet loss in the network.