Packet assignment under resource constraints with D2D communications

Abstract

D2D communications offer various benefits such as coverage extension and high energy efficiency. For example, a cluster of in-coverage devices can be recruited to relay packets for an out-of-coverage device. As the helper devices can be subject to varying relay costs and resource constraints, a sequence of packets need to be properly assigned among the helpers. In this article, we study the packet assignment problem from a mechanism design's perspective, taking into account the strategic interactions of self-interested helpers. Fundamental concepts in mechanism design are introduced for general readers to understand the background. Auctions are an important application of mechanism design and widely used in many fields. We develop a reverse auction mechanism, consisting of an allocation rule and a payment rule, to assign packets among helpers and determine payments based on their declared costs and resource constraints. As it is NP-hard to find an optimal packet assignment that minimizes the total cost, we use a low-complexity allocation rule, which performs closely to the optimal allocation on average. The payment rule implements the allocation rule in dominant strategies in which helpers disclose their true costs and resource constraints. This truthful revelation in turn justifies the allocation according to the helpers' declaration. Numerical analysis and simulations are conducted to evaluate the performance.