基于网络编码姿态监控体域网的容错性

Fault-Tolerance Based on Network Coding in Wireless Body Area Networks for Posture Detection

摘要: 由加速度传感器构成的姿态监控体域网被广泛应用于医学、运动等领域，其需要具备较高的通信可靠性，因此，姿态监控体域网的容错性是需要解决的关键问题之一。网络编码的冗余可以作为一种非常有效的错误恢复机制，在非常低的计算和硬件成本下大大提高姿态监控体域网的可靠性。本文提出在源节点和在中继节点进行网络编码的两种错误恢复机制。仿真实验结果表明采用网络编码后的网络纠错能力能提高90%左右，并且在单个中继出现失效的情况下，采用中继节点编码的网络仍然具有较好的错误恢复能力。

Abstract: Wireless body area networks consisting of accelerometer for posture detection have been widely used in the field of medicine, sports, etc. It needs to have a high reliability of communication. Therefore, the fault-tolerance of body area networks for posture detection is one of the key issues to be resolved. Network coding along with redundancy can be used as a very efficient error recovery mechanism that greatly improves body area networks for posture detection reliability at very low computational and hardware cost. Two error recovery mechanisms based on network coding are proposed, i.e. the source node coding and the relay node coding. Simulation results show that the error recovery capability of the network using network coding can improve about 90 percent. And when failure occurs in a single relay, the network using relay node coding still has better error recovery capabilities.