Research On Congestion Control Strategy Of Wireless Sensor Network In Human Motion Capture System

The human motion capture system is now becoming mainstream and mature in the fields of film making,game interaction,medical rehabilitation and so on.However,the human motion capture system based on wireless inertial sensor networks has not been widely used because of the many problems existing in the wireless network.Among them,the congestion problem of wireless network is an important factor that cannot expand the application scale.Congestion control for wireless networks is one of the key technologies in wireless sensor networks.Short time congestion may lead to the deviation of system detection results due to data packet loss.Long time congestion will greatly affect the network performance,even cause network collapse,and the system cannot operate properly.Aiming at the network characteristics of human motion capture system,a global congestion control strategy is proposed after the background and research status of congestion control strategy are elaborated in this paper.This paper studies the topology of the wireless network nodes,the active management queue mechanism of the network layer,and the TCP congestion control algorithm of the transmission control layer.The star topology is used as the network of the system.In view of the characteristics of the topology,a self-adaptive active queue management mechanism,ARED algorithm,is adopted in the network layer of the network center node,the mechanism can more flexible management transmit data from different nodes.Then,an improved TCP congestion control algorithm,TCP-NewIllinois algorithm,is proposed in the sending nodes of the network.The algorithm detects the congestion level in real time to adjust the sending rate of the sending nodes,so as to avoid the occurrence of network congestion.In order to verify the effectiveness of the proposed congestion control strategy,this paper builds a point to multipoint communication experiment through NS-3 network simulation platform,and compares it with the performance index of wireless network nodes that have not deployed the strategy.The experiments prove that the strategy of strong robustness,high throughput and high transmission efficiency,can fully meet the requirements of network system.Finally,through the simulation of human motion capture system of wireless sensor network node structure,building a real network environment in the NS-3 platform,network throughput,delay,jitter and other indicators of the system would be outputted in real-time.It provides a basic basis for the implementation of this strategy on the hardware platform of human motion capture system,and has good theoretical and practical value.