Contract for Joint Communication and Computing in Wireless Caching Systems

Abstract

Wireless network virtualization (WNV) has been approved to be a promising approach to jointly allocate physical resources. In WNV, the mobile virtual network operator (MVNO) works as a controller to schedule the physical resources, e.g., communication, computing, and caching resources. However, due to the selfishness of the user, it may keep its expected quality of service (QoS) as private information, which is not known by the MVNO for lower price or higher QoS. Such asymmetric information will lead to inefficient resource allocation. To solve the aforementioned problems, we give the caching strategy and exploit contract theory to formulate the joint resource allocation problem under asymmetric information in wireless caching systems. Particularly, the MVNO acts as a principal designing contract items for the users who are classified into a finite number of types according to their expected QoS. Since nonlinear and the analytical result is infeasible to obtain, the alternating direction method of multipliers (ADMM) algorithm is further proposed to find the optimized contract with low complexity. With the optimal contract, the MVNO successfully attracts the users to reveal their true requirements and make the efficient resource allocation decisions to maximize its own utility. Simulation results verified the effectiveness of the proposed scheme.