[Paper Review] Robust Underlay Device-to-Device Communications on Multiple Channels

Mohamed Elnourani, Siddharth Deshmukh+2 more|arXiv (Cornell University)|Feb 26, 2020

Advanced MIMO Systems Optimization34 references4 citations

TL;DR

This paper proposes a joint uplink and downlink resource allocation framework for robust underlay device-to-device (D2D) communications on multiple channels, optimizing both power and channel assignment under imperfect channel state information (CSI). Using convex relaxation, fractional programming, and alternating optimization, it develops centralized and decentralized algorithms with theoretical convergence guarantees and experimental validation showing superior performance over state-of-the-art methods.

ABSTRACT

Most recent works in device-to-device (D2D) underlay communications focus on the optimization of either power or channel allocation to improve the spectral efficiency, and typically consider uplink and downlink separately. Further, several of them also assume perfect knowledge of channel-stateinformation (CSI). In this paper, we formulate a joint uplink and downlink resource allocation scheme, which assigns both power and channel resources to D2D pairs and cellular users in an underlay network scenario. The objective is to maximize the overall network rate while maintaining fairness among the D2D pairs. In addition, we also consider imperfect CSI, where we guarantee a certain outage probability to maintain the desired quality-of-service (QoS). The resulting problem is a mixed integer non-convex optimization problem and we propose both centralized and decentralized algorithms to solve it, using convex relaxation, fractional programming, and alternating optimization. In the decentralized setting, the computational load is distributed among the D2D pairs and the base station, keeping also a low communication overhead. Moreover, we also provide a theoretical convergence analysis, including also the rate of convergence to stationary points. The proposed algorithms have been experimentally tested in a simulation environment, showing their favorable performance, as compared with the state-of-the-art alternatives.

Motivation & Objective

• To address the limitations of prior works that optimize power or channel allocation in isolation, particularly under imperfect CSI.
• To design a joint uplink and downlink resource allocation scheme that maximizes network spectral efficiency while ensuring fairness among D2D pairs.
• To maintain quality-of-service (QoS) by guaranteeing a target outage probability under imperfect CSI.
• To develop computationally efficient, low-overhead decentralized algorithms with provable convergence for practical deployment.

Proposed method

• Formulates a mixed-integer non-convex optimization problem for joint power and channel allocation in underlay D2D networks.
• Applies convex relaxation to handle the non-convexity of the original problem, enabling tractable solution methods.
• Employs fractional programming to transform the rate maximization objective into a more amenable form.
• Uses alternating optimization to iteratively update power, channel, and auxiliary variables, improving convergence.
• Designs a decentralized algorithm distributing computational load across D2D pairs and the base station to reduce signaling overhead.
• Establishes theoretical convergence with rate guarantees using the Kurdyka–Łojasiewicz (KL) property, proving convergence to stationary points.

Experimental results

Research questions

• RQ1 How can joint uplink and downlink power and channel allocation be optimized in underlay D2D networks to maximize spectral efficiency under imperfect CSI?
• RQ2 What is the impact of imperfect CSI on D2D performance, and how can QoS be guaranteed despite channel uncertainty?
• RQ3 Can decentralized algorithms achieve comparable performance to centralized ones while minimizing communication overhead and computational load?
• RQ4 What theoretical convergence guarantees can be provided for iterative optimization algorithms in this non-convex setting?
• RQ5 How does the proposed method outperform existing state-of-the-art approaches in fairness, spectral efficiency, and robustness?

Key findings

• The proposed centralized and decentralized algorithms achieve higher network spectral efficiency compared to state-of-the-art alternatives in simulation environments.
• The decentralized algorithm maintains low communication overhead by distributing computation among D2D pairs and the base station.
• Theoretical convergence analysis proves that the alternating optimization sequence converges to a stationary point of the objective function.
• The convergence rate is bounded by $ | abla F_2| ightarrow 0 $ with a rate of $ O(k^{-(1- heta)/(2 heta-1)}) $, where $ heta eq 1/2 $, indicating superlinear convergence under the KL property.
• The use of real analytic functions and the KL property ensures that the optimization process is stable and converges to meaningful solutions even in non-convex settings.
• The framework successfully maintains fairness among D2D pairs and guarantees a target outage probability, ensuring robust QoS under imperfect CSI.