A New Data Transmission Strategy in Mobile D2D Networks-Deterministic, Greedy, or Planned Opportunistic Routing?

In this paper, we study the problem of delay-constrained data transmission in mobile opportunistic device-to-device networks. In contrast to the deterministic or greedy single-copy single-path (SCSP) and multicopy multipath (MCMP) routing schemes that have been discussed in the literature, we develop a planned opportunistic routing scheme that aims to determine the optimal single-copy multipath (SCMP) transmission strategy that satisfies the delay requirement and, at the same time, minimizes communication cost. We first address the unicast by formulating the optimization problem and developing a distributed routing algorithm under practical network settings. Then, we explore optimal multicast strategies based on the SCMP transmissions. We implement the proposed algorithms on Android tablets and carry out extensive experiments, each with 25 nodes, for a period of two weeks. Moreover, we extract the algorithm codes from our prototype and run simulations based on the Haggle trace to study performance trends under various network settings. The experimental and simulation results show that the proposed protocols achieve significant performance gain in comparison with their counterparts based on SCSP and MCMP transmissions.