In wireless sensor network (WSN), the existing error recovery mechanisms do not concentrate on physical aspects such as error correction and coding. Also a complete balance needs to be maintained among the performance and energy tradeoff in order to prevent communication overhead. In
this paper, we propose a relay assisted effective loss recovery technique for WSN. In this technique, the relay nodes among the source and destination node is selected based on the parameters such as channel state information and combined score using a decentralized partially observable Markov
decision process. The combined score for each sensor is defined involving the parameters such as queue length, link bandwidth, MAC contention and residual energy. Then the low density parity check (LDPC) codes are used to encode and decode for performing error recovery. By simulation results,
we show that the proposed technique reduces the number of errors and also reduces the computational complexity while selecting the node for retransmission of packets at the destination without fail. Also the decision making can be done locally with low overhead.
Performance Comparison of Leach and GSTEB in Wireless Sensor Network