Sheikh Salman Hassan

6G non-terrestrial networks (NTNs) will strengthen global Earth observation (EO) by linking low Earth orbit (LEO) satellite constellations, high altitude platforms (HAPs), and unmanned aerial vehicles (UAVs). However, long paths, intermittent links, and tight spectrum make traditional bit-centric pipelines brittle and bandwidth-heavy. Traditional semantic communication (SC) approaches also fall short because they rely on a single-source processing and degrade severely under low signal-to-noise ratio (SNR), multi-hop, and noisy orbital conditions. Therefore, we present a unified multi-source SC with an intelligent denoising framework that transmits meaning rather than raw symbols, fuses complementary features from multiple satellites, and restores semantics in the presence of noise. We first discuss the fundamentals of the unified multi-source SC with an intelligent denoising framework. Then, we explain our proposed framework in detail. Our framework couples the multi-hop inter-satellite delivery of compact semantic features with transformer-based denoising to maintain fidelity at low SNR and receiver-side image merging and large language model (LLM)-assisted semantic refinement to assemble overlapping segments into a coherent EO product aligned with a shared knowledge base. As a proof of concept, we present a case study on EO image delivery over LEO satellite, which shows improved robustness and bandwidth efficiency relative to conventional SC without denoising and traditional communications, with higher peak SNR (PSNR) and multi-scale structural similarity index (MS-SSIM), and reduced loss across SNR regimes. Finally, we provide several open challenges and conclude the paper.

Low Earth orbit (LEO) satellites with reconfigurable holographic surfaces (RHS) enable agile aperture control for joint communication and sensing. Fast orbital motion (i.e., 7.5 km/s) produces Doppler shifts exceeding ≈ 25 kHz at 20 GHz, which destabilize OFDM links and challenge real-time scheduling. We propose an orthogonal time-frequency space (OTFS)based integrated sensing and communication (ISAC) framework that provides Doppler resilience and jointly optimizes beam power allocation, remote user (RU)-beam assignment, and sensing-communication resource sharing under regulatory constraints. Because the resulting mixed-integer nonlinear program (MINLP) is highly non-convex and NP-hard, we design a transformer-based scheduler with feasibility-preserving projections and train it via imitation learning on small-scale instances. Simulations under realistic LEO conditions demonstrate (i) a controllable communication-sensing Pareto frontier, (ii) ITU-Rcompliant operation with PFD violations below 0.5% (vs. over 40% for baselines), and (iii) sub-millisecond inference latency (≈ 0.8 ms per epoch). These results confirm that our approach achieves up to 40% higher balanced feasible utility than baseline methods under ITU-R constraints, maintains ITU-R compliance with < 0.5% violation, and enables real-time deployment in nextgeneration LEO satellite systems.

Non-terrestrial networks (NTNs) in 6G enable persistent Earth observation (EO), yet satellite downlinks remain bandwidth-limited and vulnerable to path loss, Doppler, and smallscale fading. We propose a semantic communication (SC) pipeline for satellite EO that deploys a lightweight Swin Transformer encoder onboard and a Swin-based decoder with an optional plug-and-play denoiser at the gateway. Training uses channeldiverse pairs of the same scene to promote channel-invariant representations, while inference operates on a single image for reliability and low latency. Across representative operating points, the gateway denoiser consistently improves reconstruction quality while keeping the transmitted feature budget unchanged. On average, it recovers about 60% of the gap in peak signal-to-noise ratio (PSNR) and roughly 25% of the gap in multi-scale structural similarity index (MS-SSIM) relative to a higher-bit-rate baseline. These results show that relocating refinement to the gateway shifts the communication-computation balance toward lower downlink load at comparable quality.

6G non-terrestrial networks (NTNs) will strengthen global earth observation by linking low Earth orbit (LEO) satellite constellations, high altitude platforms (HAPs), and unmanned aerial vehicles (UAVs). However, long paths, intermittent links, and tight spectrum make traditional bit-centric pipelines brittle and bandwidth-heavy. Traditional semantic communication (SC) approaches also fall short because they rely on a single-source processing and degrade severely under low-SNR, multi-hop, and noisy orbital conditions. Therefore, we present a unified multisource SC with an intelligent denoising framework that transmits meaning rather than raw symbols, fuses complementary features from multiple satellites, and restores semantics in the presence of noise. We first discuss the fundamentals of the unified multisource SC with an intelligent denoising framework. Then, we explain our proposed framework in detail. Our framework couples the multi-hop inter-satellite delivery of compact semantic features with transformer-based denoising to maintain fidelity at low signal-to-noise ratios (SINR) and receiver-side image merging and LLM-assisted semantic refinement to assemble overlapping segments into a coherent Earth observation (EO) product aligned with a shared knowledge base. As a proof of concept, we present a case study on EO image delivery over LEO satellite, which shows improved robustness and bandwidth efficiency relative to conventional SC without denoising and traditional communications, with higher peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), and reduced loss across SNR regimes. Finally, we provide several open challenges and conclude the paper.

The advent of the sixth-generation (6G) wireless networks, enhanced by artificial intelligence, promises ubiquitous connectivity through Low Earth Orbit (LEO) satellites. These satellites are capable of collecting vast amounts of geographically diverse and real-time data, which can be immensely valuable for training intelligent models. However, limited inter-satellite communication and data privacy constraints hinder data collection on a single server for training. Therefore, we propose SemSpaceFL, a novel hierarchical federated learning (HFL) framework for LEO satellite networks, with integrated semantic communication capabilities. Our framework introduces a two-tier aggregation architecture where satellite models are first aggregated at regional gateways before final consolidation at a cloud server, which explicitly accounts for satellite mobility patterns and energy constraints. The key innovation lies in our novel aggregation approach, which dynamically adjusts the contribution of each satellite based on its trajectory and association with different gateways, which ensures stable model convergence despite the highly dynamic nature of LEO constellations. To further enhance communication efficiency, we incorporate semantic encoding-decoding techniques trained through the proposed HFL framework, which enables intelligent data compression while maintaining signal integrity. Our experimental results demonstrate that the proposed aggregation strategy achieves superior performance and faster convergence compared to existing benchmarks, while effectively managing the challenges of satellite mobility and energy limitations in dynamic LEO networks.

Space-air-ground integrated networks (SAGINs) are emerging as a fundamental architecture for 6G systems to enable massive connectivity, novel applications, extreme data rates, ultra-low latency, and multi-dimensional networking. The complex nature of SAGINs has driven researchers to explore classical machine learning (CML) as a powerful tool for modeling and optimizing these multi-layer networks. Although CML offers many benefits, it will suffer from exponential complexity for SAGIN design, mainly due to high-dimensional data and problem spaces. Therefore, the NP-hard problems are challenging for CML to enable SAGINs efficiently. To address this critical challenge, we present a comprehensive survey examining the integration of quantum ML (QML) with SAGINs. We systematically evaluate quantum-enhanced SAGIN technologies through detailed performance metrics and create a new classification system for hybrid network architectures. We thoroughly examine key integration strategies and analyze major technical challenges, which include how quantum states degrade in space, how to distribute resources optimally, and how to develop security protocols. We show how QML can solve the growing complexity of 6G networks by improving network capacity, intelligence, and security. The research also creates a comprehensive framework for quantum-enhanced SAGINs that researchers and industry professionals can use as a guide when developing next-generation communication systems. Our findings examine the transformative potential of QML in reshaping global communications infrastructure, which presents a structured approach to navigating the intricate challenges of advanced networking technologies and also provides a critical pathway for addressing the exponentially growing computational demands of future communication ecosystems.

The sixth-generation (6G) non-terrestrial networks (NTNs) are crucial for real-time monitoring in critical applications like disaster relief. However, limited bandwidth, latency, rain attenuation, long propagation delays, and co-channel interference pose challenges to efficient satellite communication. Therefore, semantic communication (SC) has emerged as a promising solution to improve transmission efficiency and address these issues. In this paper, we explore the potential of SC as a bandwidth-efficient, latency-minimizing strategy specifically suited to 6G satellite communications. The existing SC methods have demonstrated efficacy in direct satellite-terrestrial transmissions; however, they still encounter certain limitations. Specifically, some ground users (GUs) experience poor signal-to-noise ratios (SNR), making direct satellite communication challenging. To address these issues, we propose a novel framework that optimizes gateway hop-relay selection for GUs with low SNR and integrates gateway-based denoising mechanisms to ensure high-quality-of-service (QoS) in satellite-based SC networks. This approach directly mitigates distortion, leading to significant improvements in satellite service performance by delivering customized services tailored to the unique signal conditions of each GU. Our findings represent a critical advancement in reliable and efficient data transmission from the Earth observation satellites, thereby enabling fast and effective responses to urgent events. Simulation results demonstrate that our proposed strategy significantly enhances overall network performance, outperforming conventional methods by offering tailored communication services based on specific GU conditions.

In the evolving landscape of 6G satellite-based nonterrestrial networking, optimizing network resources presents significant challenges, particularly due to the limitations of conventional reinforcement learning (RL) methods. Existing RL approaches often rely on manually designed reward functions, which require extensive parameter tuning that fails to adapt effectively to dynamic network conditions. To overcome this challenge, we propose a novel framework leveraging generative adversarial imitation learning (GAIL) with inverse reinforcement learning (IRL) to automate reward function design and enhance policy learning for beamforming and spectrum allocation. Our approach addresses the non-convex optimization challenge of maximizing spectrum efficiency (SE) while ensuring minimum information rate requirements for remote user equipment (RUEs). By integrating the Whale optimization algorithm (WOA) to generate expert policies, we provide high-quality training data for the GAIL framework, which enables intelligent, data-driven adaptation to network dynamics. This innovative methodology allows satellite agents to autonomously learn optimal strategies through iterative interactions with the environment that significantly advance traditional network optimization paradigms. The simulation results validate the effectiveness of the proposed approach which demonstrates a notable 14.6% improvement in convergence speed and reward values over traditional RL technique.

We address the challenge of delivering high-quality wireless connectivity to remote and underserved areas in the sixth-generation (6G) networks. The absence of terrestrial infrastructure, combined with severe signal attenuation and dynamic environmental conditions, often degrades the quality of service (QoS) for remote users (RUs). To overcome these limitations, we propose a novel 6G communication framework that integrates unmanned aerial vehicles (UAVs) with hybrid reconfigurable intelligent surfaces (RISs) featuring both active and passive components. Our UAV-mounted hybrid RIS system enhances mobility, environmental adaptability, and signal propagation through intelligent phase and gain control. We aim to maximize RU communication rates, ensure reliable connectivity, and optimize energy efficiency (EE) while preserving network QoS. We formulate a mixed-integer non-linear programming (MINLP) problem that jointly considers UAV trajectory planning, channel assignment, and RIS configuration. To solve this, we decompose the problem into three interdependent subproblems and apply a sequential approach using proximal policy optimization (PPO)based deep reinforcement learning, matching theory, and the Whale optimization algorithm (WOA). Our RIS configuration jointly optimizes passive phase shifts and active gain control to adaptively enhance weak links. We demonstrate by simulation results that our framework outperforms existing approaches in terms of communication reliability, energy-aware operation, and overall QoS.