Low complexity algorithms for network coding based on singular value decomposition

In this paper, we propose a low complexity algorithm for decoding where network coding is deployed in client-server networks. We consider battery powered clients, so that minimizing their power consumptions is essential. Our focus is thus on developing a decoding algorithm that can reduce the computational complexity. Unlike general decoding algorithms that are based on Gaussian elimination, we propose a decoding algorithm based on the singular value decomposition, as it enables to easily compute an inverse matrix, leading to lower decoding complexity. Our simulation results confirm that proposed algorithm can reduce not only the decoding complexity but also the overall network complexity. While the network efficiency of the proposed strategy is degraded as the network dimension increases, we show that the efficiency converges into a lower bound as the network dimension increases. These are confirmed by the simulation results.