Joint beamwidth and resource optimization in ultra-dense MmWave D2D communications

The symbiosis of millimeter wave (mmWave) transmission and Device-to-Device (D2D) communication has become a a promising candidate for temporary or spontaneous short-distance communications. Although the use of directional antennas in mmWave communications can effectively reduce the interference between users, as the network becomes much denser, the mmWave D2D communication network tends to be interference-limited rather than noise-limited. Significantly, the inter-user interference from the line-of-sight path should be carefully circumvented to boost the achievable rate further. We first identify that beamwidth, channel state, and resource allocation jointly affect the interference level and achievable rate. To effectively manage the interference, a joint beamwidth selection and resource optimization problem are established to maximize the sum rate. We propose a two-stage algorithm to solve this non-convex mixed integer programming with low complexity. First, a feasibility-aware particle swarm optimization based beamwidth selection algorithm is proposed to find the proper beamwidth. Second, a constrained concave-convex procedure based joint power optimization and time slots scheduling algorithm is proposed to facilitate the concurrent transmission and further improve the sum rate. Numerical results demonstrate that the proposed two-stage algorithm can effectively reduce interference and improve the sum rate with low complexity.