Protocol Design and Game Theoretic Solutions for Device-to-Device Radio Resource Allocation

Device-to-device (D2D) communication has been proposed to improve the resource efficiency and lighten the heavy load of the base station (BS) in Long-Term Evolution (LTE)-Advanced systems. In a D2D-enabled LTE-A system, the resource efficiency is primarily determined by D2D/cellular mode selection and resource allocation. However, the D2D channel quality, which is the key factor for the BS to allocate resources, cannot be learned directly by the BS, owing to the peculiarity of D2D communication. Rational user equipment (UE) will take advantage of the peculiarity to report their experienced D2D quality untruthfully for their selfish interests and, consequently, degrade system efficiency. This so-called unknown channel quality (UCQ) problem imposes a fatal impact on the resource efficiency and will limit the practicality of D2D communication. To overcome the UCQ problem, we propose to use game theory to analyze the peculiarity of D2D communication. In this paper, two practical D2D resource allocating protocols were investigated, and the system efficiencies were analyzed to show the potential performance degradation when the UCQ problem is not addressed. To circumvent the performance degradation caused by the UCQ problem, a contract-based mechanism and the corresponding algorithms were proposed to eliminate the UE's incentive of reporting untruthfully. Numerical and simulation results validated the feasibility and the effectiveness of our approach.