A joint linear-analog network coding based protocol for data transfer in underwater acoustic networks

In underwater acoustic networks, the objective that researchers are pursuing is establishing a network with high reliability, high throughput and low energy consumption. Network coding can enhance reliability and throughput of network. Moreover, the throughput of network will be significantly improved by analog network coding. Accordingly, with higher reliability and higher throughput, the energy consumption will be lower. In this paper, the linear network coding and analog network coding are combined creatively. Besides, a SNR estimation algorithm is used to choose packets instead of choosing packets randomly for decoding. Afterwards, for taking advantage of the redundant packets better, packets are refreshed via bit-by-bit comparison. At last, we present numerical experiments to evaluate the performance of this scheme.