Research On Congestion Control Algorithm For Healthcare Wireless Sensor Networks

Wireless Sensor Networks (WSNs) is a multi-hop self-organized network formed by a large number of small, low-power sensor nodes through wireless communication technology. With the rapid development of wireless sensor technology, WSNs has been used in many fields, especially in the field of medical treatment. Healthcare Wireless Sensor Networks(HWSNs) is a special category of WSNs, the main function is data collection for medical diagnosis and continuous health assessment. Healthcare Wireless Sensor Networks plays an important role in medical care, reliable and real-time transmission of data is essential in the application of HWSNs, However, medical emergencies will be produced a lot of suddenly flow, network will suffer unpredictable load and causing network congestion. Unlike traditional wired networks, due to the WSNs is a many to one communication mode, the mutual interference of wireless links, the dynamic changes of the network topology, the network resource constraints and other characteristics, WSNs is more prone to congestion. Congestion not only leads to packet loss, increases the end-to-end delay, but also excessive consumption the valuable energy of the network due to retransmission of lost packets. However, it will directly affect the patient's health due to key signal loss, delay or node death in Healthcare Wireless Sensor Network. Therefore, the processing of congestion is a necessary means to save the energy of WSNs an, improve the network QoS, and it is also an important challenge for WSNs.Due to the inherent characteristics of WSNs, the congestion control mechanism of traditional network can't directly apply to WSNs. With the further research of WSNs, the congestion control algorithm of WSNs is more and more used in different scenarios. At present, WSNs congestion control usually includes congestion detection, congestion feedback and congestion control. Although there have been some research results in the congestion control of WSNs, but most of the congestion control algorithms are based on the simple congestion detection and rate regulation mechanism to deal with congestion. Most existing congestion detection methods are based on single metric, such as queue length, channel sampling and the ratio between packet service time and arrival time interval, it does not take into account the change tendency of the flow and congestion of the gradual process. Congestion processing mostly through the speed of adjustment or traffic scheduling, seldom consider the fairness, reliability and real-time performance of the network transmission, especially in HWSNs, the network performance is particularly important. In view of these problems, this paper proposes virtual queue based congestion control algorithm (Congestion Control Algorithm based on virtual queue, CCVQ). The specific research contents and results are as follows:First of all, This paper introduces the research background and significance in HWSNs and also introduces the current situation of WSNs congestion control research, expounds the causes and types of WSNs congestion and the difficulty of the WSNs congestion control is analyzed. This paper introduces the key technology and schedule of the current congestion control, then analyzes and studies the current common congestion control strategy and then summarizes the advantages and defects and congestion control for HWSNs of problems to be solved.Secondly, aiming at the existing problems of the WSNs congestion control algorithm and the characteristics of the data in HWSNs, such as reliability, delay and fairness, this paper proposes CCVQ. CCVQ algorithm includes two steps:congestion avoidance and congestion elimination. In this paper, firstly, we construct the multipath and QoS aware routing to avoid congestion, which can improve the transmission reliability and real-time performance of the key information and achieve the goal of energy consumption balance. However, due to the WSNs is a many to one communication mode, there congestion will be occurs near the sink node. In this paper, the rate adjustment mechanism and the active packet loss strategy are used to eliminate the congestion, which can guarantee Low priority data reliability, reduce end to end delay and improves network throughput.CCVQ detect congestion based on the rate of change of queue and the occupancy of cache. If congestion occurs in the network, in order to relieve congestion, in this paper, we reduce the sending rate of the child's and source rate of the node based to the degree of congestion and these priority index; If network congestion is going to occur, which indicates that the sending rate of the node is less than the receiving rate, in order to reduce the receiving rate of the node, this paper adjust the sending rate of the child's and source rate of the node according to the virtual queue change rate, when the utilization rate of the cache area is reduced, the utilization ratio of network resources is low, In order to increase the throughput of the network, this paper will increase the sending rate of the child's and source rate of the node according to queue length and these priority. At the same time, in order to prevent the buffer overflow caused by congestion, this paper adopts the active packet loss policy management mechanism, when the packet loss probability is greater than a threshold, discard packets with minimal residual value in the cache, thereby saving the network energy consumption and improve the network QoS.Finally, we use NS2 simulation platform to simulate CCVQ algorithm and comparative algorithm HOCA and REEP. Simulation results show that the proposed CCVQ algorithm has obvious advantages in terms of packet loss rate, network throughput, end to end delay and Normalized fairness over energy.