Distributed pricing-based resource allocation for dense device-to-device communications in beyond 5G networks

TL;DR

This paper introduces a low-complexity distributed pricing-based resource allocation algorithm for dense D2D communications in beyond 5G networks, improving spectral efficiency and sum-rate performance.

Contribution

It proposes a novel distributed power allocation method with a utility function that manages D2D admission and enhances spectral efficiency in dense scenarios.

Findings

Effective resource allocation for dense D2D in beyond 5G

Higher sum-rate performance compared to other algorithms

Fast closed-form solution for power allocation

Abstract

Considering the dramatic increase of data rate demand in beyond fifth generation (5G) networks due to the numerous transmitting nodes, dense device-to-device (D2D) communications where multiple D2D pairs can simultaneously reuse a cellular link can be considered as a communication paradigm for future wireless networks. Since distributed methods can be more practical compared to complex centralized schemes, we propose a low-complexity distributed pricing-based resource allocation algorithm to allocate power to cellular user equipments (CUEs) and D2D pairs constrained to the minimum quality-of-service (QoS) requirements of CUEs and D2D pairs in two phases. The price of each link is set by the base station (BS). CUEs and D2D pairs maximize their rate-and-interference-dependent utility function by adjusting their power in the first phase, while the price of each link is updated at the BS based on the minimum QoS requirements of CUEs and D2D pairs in the second phase. The proposed utility function controls D2D admission, hence it is suitable for dense scenarios. The proposed method is fast due to the closed-form solution for the first phase power allocation. Numerical results verify the effectiveness of the proposed method for practical dense beyond 5G scenarios by allocating resources to multiple D2D pairs and taking advantage of the spatial reuse gain of D2D pairs. Furthermore, the utility function definition is discussed to illustrate its effectiveness for SE maximization. Finally, the SE of the proposed method is compared to other centralized and distributed algorithms to demonstrate the higher sum-rate performance of the proposed algorithm.