بهينه‌سازي سيستم‌هاي يكپارچه حسگري و مخابراه با بهره‌گيري از سكوهاي ارتفاع بالا در شبكه‌هاي 66

Sixth-generation (6G) networks, in their pursuit of ultra-high data rates, seamless coverage, an‎d precise sensing capabilities, face fundamental challenges in system architecture an‎d optimization. In this context, integrated sensing an‎d communication (ISAC) emerges as a promising key technology, enabling improved spectral an‎d energy efficiency. However, achieving optimal resource allocation an‎d fairness in complex, dynamic radio environments requires advanced solutions. This dissertation presents a comprehensive an‎d innovative ISAC framework for 6G networks based on high altitude platform stations (HAPS). Unlike traditional approaches, this study proposes an‎d comparatively eva‎luates novel HAPS-based network architecture patterns designed to create synergy between communication an‎d sensing functions in ubiquitous mobile communication networks. Two distinct architectural models are analyzed: (1) a centralized structure in which the HAPS operates solely as a macro base station (SMBS), an‎d (2) a hierarchical an‎d collaborative architecture in which the HAPS acts as a central processing unit (CPU) while unmanned aerial vehicles (UAVs) serve as dynamic an‎d flexible aerial access points. The performance of these patterns is quantitatively eva‎luated under various user scenarios, including single-antenna an‎d multi-antenna user equipment (UE). To maximize quality of service for multi-antenna users, advanced signal detection algorithms such as zero-forcing (ZF), minimum mean square error (MMSE), an‎d maximum ratio combining (MRC) are implemented. Results highlight the complementary roles an‎d dual communication-sensing capabilities of both architectures. The centralized HAPS structure shows strong capacity for ubiquitous communication coverage an‎d large-scale sensing, such as environmental monitoring, whereas the hierarchical UAV-based structure is more effective in responding dynamically to variable traffic deman‎ds an‎d carrying out precise, targeted sensing missions. This dissertation also investigates two approaches to resource allocation optimization in HAPS-ISAC systems. In the first, the problem is modeled as a single-objective optimization that maximizes the minimum beampattern gain transmitted toward sensing targets, subject to communication users’ signal-to-interference-plus-noise ratio (SINR) constraints an‎d power allocation. In the second, the problem is formulated as a multi-objective, non-convex optimization that simultaneously seeks to maximize the minimum SINR for communication users an‎d enhance the reflected signal power from sensing targets, under strict constraints on power allocation an‎d beampattern gain. To solve these problems, a genetic algorithm (GA) is first employed as a preliminary search tool for the single-objective model. Then, for the multi-objective problem, the non-dominated sorting genetic algorithm II (NSGA-II) is developed to extract a set of Pareto solutions effectively. In addition, a deep reinforcement learning (DRL) agent with a multi-layer perceptron (MLP) architecture is designed an‎d implemented to eva‎luate the performance of the proposed methods. Rigorous mathematical analysis is also used to validate the results an‎d provide a comprehensive comparison. Extensive simulation results demonstrate improvements in the trade-off between sensing an‎d communication capabilities, significant increases in users’ minimum SINR an‎d sensing accuracy, an‎d fairer resource distribution across the network. The study shows that the proposed NSGA-II-based method achieves multi-objective goals more efficiently than benchmark deep reinforcement learning methods. These findings not only enrich the theoretical foundations of ISAC an‎d 6G research but also offer practical guidelines for designing an‎d implementing next-generation communication systems capable of meeting the growing deman‎ds for ubiquitous communication an‎d precise sensing in complex 6G ecosystems.

محمود مدرس هاشمي , محمدجواد اميدي

محمدرضا تابان , احسان يزديان , مهرداد اردبيلي