Topology Awareness of Hierarchical Peer-to-Peer Network Based on Network Coding with Feedback
Jun Yang, Bin Dai, Benxiong Huang, Chen Huang · 2010
Network coding is a promising technology to improve throughput and reliability. However, the benefit of network coding is restricted and affected by network topology, especially in peer-to-peer system. Precise topology awareness can be helpful to improve system performance of network coding in terms of throughput and reliability. In this paper, we propose an innovative scheme TANCF (Topology Awareness based on Network Coding with Feedback), to detect the path in hierarchical P2P network based on the feedback from the intermediate nodes and the receiver nodes of network coding system. It is a first attempt to use network coding with feedback to reveal the network connectivity and to construct the network topology. In the TANCF scheme, the path matrix of network can be obtained by the feedback from intermediate nodes and receiver nodes, which has a close relationship with the adjacency matrix of topology, so that it can be used to construct the network topology that lead to optimize the parameters of network coding. Through the matrix analysis and simulation experiment, it can be proved that TANCF scheme can obtain the path matrix correctly with a high probability.