Optimum Reed-Solomon Erasure Coding in Fault Tolerant Sensor Networks

This paper presents an optimization framework for a wireless sensor network in which the sensor network uses multipath diversity in the network layer and erasure codes. We use a distributed sink where information arrives at the sink via multiple proxy nodes, called "prongs" in [1]. The sender node uses Reed-Solomon erasure coding and splits each packet into multiple fragments and transmits the fragments over multiple disjoint paths. The erasure coding allows the sink to reconstruct the packet if it receives a number of fragments equal to or higher than the number of fragments in the original packet. The total size of fragments and the number of fragments that should be transmitted on each path are design parameters that can be tuned to adapt to the traffic of the network nodes, number of parallel paths and availability probability of each link. The total cost due to packet loss and fragment transmission using such systems is analyzed in the presence of disjoint parallel paths. Based on the total cost, we investigate the trade-off between the system reliability and the cost of fragment transmissions. The problem of optimizing the number of parity fragments under different network situations is treated numerically.