Adaptive and Reliable Multi-Path Transmission in Wireless Sensor Networks using Forward Error Correction and Feedback

Providing end-to-end reliability for data transmission is a major challenge for energy-constrained and bandwidth-limited wireless sensor networks. To meet this challenge, attempts have been made using a multi-path transmission mechanism that encodes an information bitstream using forward error correction (FEC), and sends the obtained packets over a number of paths to the destination. Algorithms that, given the paths statistics (e.g., failure probability and energy cost of data transmission over the paths), find the number of channel packets and their transmission paths that optimize the reliability-energy cost tradeoff have been proposed. Once a transmission strategy is obtained, all packets are transmitted regardless of the actual state of the wireless links, which may result in energy and bandwidth losses due to unnecessary transmissions. This can be more severe when arbitrary transmission strategies have to be used due, for example, to the unavailability of paths statistics at the source node, which is the case in many practical applications. In this paper, we propose a hybrid FEC-feedback mechanism. The feedback scheme is used to send ACK packets from the sink to the source node when the information bitstream is actually recovered. We then derive a measure of the expected energy cost of data transmission for the hybrid mechanism. Simulation results show that, on average, the hybrid mechanism consumes up to about 44% less energy than the FEC-based mechanism for the transmission of a bitstream with the same reliability constraint. The energy savings of the hybrid mechanism are even higher when arbitrary transmission strategies have to be used.