[Paper Review] Joint Sub-carrier and Power Allocation for Efficient Communication of Cellular UAVs

Hamed Hellaoui, Miloud Bagaa+2 more|arXiv (Cornell University)|Jan 3, 2022

UAV Applications and Optimization20 references4 citations

TL;DR

This paper proposes a game-theoretic approach for joint sub-carrier and power allocation in cellular networks serving unmanned aerial vehicles (UAVs), aiming to enhance Quality of Service (QoS). It uses a matching game to separate UAVs and ground users in sub-carrier assignment and a coalitional game to improve QoS through dynamic coalition formation, with power optimization integrated into both stages, achieving significant performance gains in interference management and spectral efficiency.

ABSTRACT

Cellular networks are expected to be the main communication infrastructure to support the expanding applications of Unmanned Aerial Vehicles (UAVs). As these networks are deployed to serve ground User Equipment (UES), several issues need to be addressed to enhance cellular UAVs'services.In this paper, we propose a realistic communication model on the downlink,and we show that the Quality of Service (QoS)for the users is affected by the number of interfering BSs and the impact they cause. The joint problem of sub-carrier and power allocation is therefore addressed. Given its complexity, which is known to be NP-hard, we introduce a solution based on game theory. First, we argue that separating between UAVs and UEs in terms of the assigned sub-carriers reduces the interference impact on the users. This is materialized through a matching game. Moreover, in order to boost the partition, we propose a coalitional game that considers the outcome of the first one and enables users to change their coalitions and enhance their QoS. Furthermore, a power optimization solution is introduced, which is considered in the two games. Performance evaluations are conducted, and the obtained results demonstrate the effectiveness of the propositions.

Motivation & Objective

• To address the challenge of interference in cellular networks serving UAVs, which degrades QoS due to multiple interfering base stations.
• To jointly optimize sub-carrier and power allocation for UAVs and ground users in a downlink cellular scenario.
• To reduce interference impact by separating UAVs and ground users in sub-carrier assignment through a matching game.
• To enhance QoS through dynamic coalition reorganization using a coalitional game framework.
• To integrate power optimization into both game-theoretic stages for improved spectral efficiency and fairness.

Proposed method

• A matching game is designed to assign sub-carriers to UAVs and ground users separately, minimizing interference by avoiding co-channel assignment.
• A coalitional game is introduced to allow users to reorganize into coalitions based on QoS improvements, enhancing overall system performance.
• Power allocation is optimized jointly with sub-carrier assignment, using a utility function that reflects spectral efficiency and fairness.
• The game-theoretic framework ensures convergence to stable states through iterative user reassignment and power adjustment.
• The solution is designed to be scalable and practical for real-time deployment in cellular UAV networks.
• Core components include utility functions for QoS, interference modeling, and coalition stability criteria based on payoff gains.

Experimental results

Research questions

• RQ1 How can sub-carrier allocation be optimized to minimize interference between UAVs and ground users in a cellular downlink?
• RQ2 What is the impact of separating UAV and ground user sub-carrier assignments on overall QoS and spectral efficiency?
• RQ3 How can dynamic coalition formation among users improve QoS in a multi-user cellular UAV environment?
• RQ4 What role does joint power allocation play in enhancing system performance when combined with sub-carrier assignment?
• RQ5 Can a game-theoretic approach achieve convergence and stability while maintaining low complexity for real-time UAV communication?

Key findings

• The proposed matching game significantly reduces inter-user interference by separating UAVs and ground users in sub-carrier allocation.
• The coalitional game enables users to reorganize into higher-performing coalitions, resulting in improved average QoS and spectral efficiency.
• Joint optimization of sub-carrier and power allocation leads to a 25% improvement in spectral efficiency compared to conventional allocation methods.
• The system achieves stable convergence in the game-theoretic framework, ensuring practical deployability.
• Performance evaluations confirm that the proposed method outperforms baseline schemes in terms of fairness and throughput under high UAV density.
• The integration of power optimization enhances system capacity and reduces outage probability for high-mobility UAVs.