Multi-Objective Meet-in-the-Middle Optimization for Relay Selection in Cellular User Cooperation: A Focus on Latency-Sensitive Applications

Abstract

Communication in cellular networks faces challenges such as intermittent connectivity, variable packet loss, and high latency, particularly in remote or high-mobility environments. This study proposes a novel user cooperation model for providing non-traditional broadband connections by bandwidth aggregation. It addresses the challenge of relay selection in the proposed model by utilizing a Multi-Objective Meet-in-the-Middle (MitM) optimization technique for selecting relay sets that can support high-bandwidth and latency-sensitive applications in remote regions. The study uses a regression-based simulation model to characterize relay mobility and positional variations,
enabling the evaluation of relay performance under dynamic conditions. This model informs the development and comparison of two relay selection algorithms: GreedySSRS, which prioritizes throughput, and iSSRS, which balances throughput, delay, and packet loss using multi-
objective optimization. Key findings demonstrate that GreedySSRS achieves higher aggregate throughput (775.25 Mbps) compared to iSSRS (721.91 Mbps). However, iSSRS significantly reduces delay by 4.7% and packet loss by 2.5% compared to GreedySSRS, making it more suitable for latency-sensitive applications. The proposed approach offers a practical solution to enhance user cooperation in dynamic wireless networks, providing reliable and efficient communication in scenarios with limited or unreliable infrastructure, such as remote areas and high-mobility environments. These results underscore the potential of multi-objective optimization techniques to improve network performance in real-world applications where stringent quality of service (QoS) requirements must be met.