A Tractable Multi-RATs Offloading Scheme on D2D Communications

Device-to-device (D2D) communications achieve a considerable proximate gain by means of single hop, which can improve the link spectrum efficiency of cellular networks by reusing the shortage licensed spectrum. However, co-channel interference will become a serious challenge that will influence the performance of both D2D and cellular links if it cannot be processed appropriately. The existing solutions are limited to single radio access technology (RAT), and they all involve tradeoffs between the available terminal density and the link spectrum efficiency. In this paper, we propose the offloading scheme of D2D communications on multi-RATs, which can achieve high link spectrum efficiency without sacrificing the available terminal density. The multi-RATs offloading scheme will achieve maximum link spectrum efficiency with the retention probability as a parameter that is formulated for a non-convex function. Considering the main factor of different scenarios that refer to high and sparse available terminal densities, we propose the dynamic adjustment offloading algorithm and the sparse-density optimization offloading algorithm to solve the non-convex problem. In addition, because we use the closed form of the stochastic geometry, the optimum offloading algorithms possess predictable features that do not break the established D2D links when a new link is applied. Finally, the simulation results show that the performance of the coverage probability and link spectrum efficiency is both greatly improved compared with the traditional D2D communications without exploiting the offloading scheme. More significantly, the spectrum efficiency of cellular links is also improved effectively due to the offloading effect.