MOCA: Multiobjective Cell Association for Device-to-Device Communications

Abstract

It is widely accepted that device-to-device (D2D) communication is envisaged to become the key enabler of direct localized communication between mobile nodes in future wireless networks. However, little attention has been paid to an important aspect that can potentially affect both the performance of D2D and cellular transmissions, which is that of the D2D cell association. In this paper, a multiobjective cell association (MOCA) optimization framework for orchestrating a large number of D2D links in a multicell network is introduced. To this end, and without loss of generality, a differentiated Fractional Frequency Reuse (FFR) scheme is considered as the interference-limiting method, especially for cell-edge users, and we assume the provision of different resource pools for D2D and cellular users, which can vary according to their location. Under this assumption, we develop a set of integer linear programming optimization formulations for D2D links, part of which fall within the coverage area of different neighboring base stations. The main purpose is to achieve improved network traffic balancing via an efficient cell association scheme. Furthermore, we provide an iterative randomized resource allocation algorithm ( i-RRA ), which roots its logic on the differentiated FFR model in order to increase the overall network throughput. A wide set of numerical investigations demonstrate the benefits offered by MOCA, as well as the throughput gains that can be achieved through i-RRA , compared to existing solutions.