Abstract :
In this paper, we consider the problem of minimizing the delay for data exchange among a group of wireless clients, where each client initially holds a subset of the packets and needs to get all the packets held by other clients. It is assumed that clients cannot communicate with each another directly, they can only exchange packets through a wireless relay. To minimize the total transmission delay during data exchange process, we need to determine at every client, which packets to be uploaded and how to encode the packets. It is also important for the relay node to decide how to encode multiple packets from different clients and select the transmission rate in the downloading process, such that every client can decode all required packets in shortest delay. We first formulate theoretically the above problem of minimizing the total transmission delay as an integer programming, and show that its complexity is NP hard. We then propose an efficient heuristic algorithm, which consists of two processes: uploading process, i.e., how to select and encode the packets from the clients to the relay, and downloading process, i.e., how the relay encode packets and select transmission rate for broadcast to all clients. For each process, theoretical formulation has been derived to minimize their transmission delay, and efficient algorithms are proposed separately. Finally, simulation results demonstrate the effectiveness of the proposed algorithm in reducing the total data exchange delay.
Keywords :
broadcast communication; computational complexity; cooperative communication; integer programming; minimisation; network coding; relay networks (telecommunication); NP hard complexity; client broadcast; data exchange process; delay minimization; heuristic algorithm; integer programming; multiple packets; network coding; packet exchange; relay-based cooperative data exchange; transmission delay; transmission rate; wireless clients; wireless relay; Communication system security; Delays; Encoding; Network coding; Relays; Wireless networks;