Energy Efficient Two-hop D2D Communications Underlay 5G Networks: A Stackelberg Game Approach

Abstract

Although coverage and capacity are the key elements of the 5G user experience, a dominant part of the population living in rural areas still experience inferior connectivity. Several solutions have been proposed to address this issue. They include deploying small cells, increasing the number of sectors per eNodeB, and reusing signal repetition. However, most of them require complex deployment and expensive fees. Accordingly, many efforts have been deployed on coverage extension software. Even so, many critical issues related to public safety, relay capacity, and devices power constraints are still challenging. As a contribution, we propose in this paper a spectral and energy-efficient two-hop device to device (D2D) relay selection algorithm. Our main goal is to extend the connectivity to the out-of-coverage (OOC) devices. Contrarily to previous solutions in which the relay is selected centrally or individually, we propose a distributed two-stage algorithm based on the Stackelberg game to involve all the competing devices. In the first stage, the OOC devices (OCDUs) are matched with the relays maximizing their spectral efficiency, and the required bandwidth for each one is determined. Then, a power control stage is investigated to calculate the optimal transmission power. The numerical and simulation analysis shows that the proposed schema outperforms the former solutions in total system capacity, spectral efficiency (SE), and energy efficiency (EE) while reducing the complexity.

Agent-Based Simulations for Coverage Extensions in 5G Networks and Beyond

“…We will mainly be interested here in the connectivity of the all network coverage probability improvement due to minimization of congestion in the base station (BS) thanks to D2D. An energy-efficient two-hop D2D relay selection algorithm was proposed in [7] to extend connectivity to the out-ofcoverage devices. The proposed algorithm is distributed and based on the Stackelberg game in order to involve all the competing devices.…”

A New Accurate Approximation of the Gaussian Q-Function with Relative Error Less Than 1 Thousandth in a Significant Domain

“…On the other hand, the analytical problems associated with it have raised a great interest in finding its closed-form bounds or approximations for decades [2]- [9], in order to simplify the handling of the mathematical expressions involving it [5], [6]. In fact, exponential-type bounds or approximations are often useful in the bit-error probability evaluation for the most common communication and information theory problems, such those involving coding (see, e.g., [10]- [12], addressing low density parity check (LDPC) codes), fading (see, e.g., [13], considering the FPGA implementation of a burst error and burst erasure channel emulator, being the error burst the result of a temporary reduction in the power of the received signal (fading), leading to a demodulation failure of a certain number of symbols), and multichannel reception (see, e.g., [14], considering a joint spectrum and energy efficient resource allocation algorithm for D2D communications). These approximations for the Gaussian Q-function have been developed with the objective of obtaining high estimation accuracies, to derive the error probability for digital modulation schemes.…”