Research On Reliability Based On Network Coding In Wireless Sensor Networks

Data transmission reliability is an important measurement of quality of service(QoS) in wireless sensor networks. Mechanism for reliability has been a key issue inthis field. Traditional reliable mechanisms are Automatic Repeat Request (ARQ),Forward Error Correction (FEC), multi-path transmission and so on. They have defectsin energy consumption and increase lots of redundant data. Therefore, this paperresearches the issue on reliable data transmission based on network coding in wirelesssensor networks. The main work is as follows.Firstly, we introduces the concept of network coding, common network codingmethod and then analyze the effects of network coding for data transmission reliabilityand average energy consumption in transmission. Comparing the reliability betweentraditional transmission and transmission with network coding, it is concluded that in acertain number of redundant, use network coding technology can improve the reliabilitywhich energy consumption caused by node encoding is less. Therefore, it suitable forresource-constrained wireless sensor networks to improve reliability.By analyzing the advantage and disadvantage of ARQ retransmission mechanism,combing ARQ and network coding technology, we propose a reliable transmissionmechanism based on network coding. In this strategy, the encoding nodes in the pathcompute the required number of redundancy according the channel loss rate and thenumber of received packets by this node, and use ARQ mechanism to ensure the nodecan decode packets successfully. At last, simulation experiments shows that thistransmission strategy is superior to traditional transmission in data transmissionreliability.Finally, we propose an encoding nodes selection strategy based on node degree inflooding routing transmission.This strategy selects the nodes in sparse field according totheir degree, and make these nodes add redundancy by encoding to ensure their reliabletransmission. It improves the reliability in flooding routing by using limit encodingnodes. Simulation shows the relationship between the density of network topology andthe number of encoding nodes in the strategy.