Syed Muhammad Waqas

Syed Muhammad Waqas

Auto-generated profile, based on publicly available information

Syed Muhammad Waqas

Yango University

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 July 2023 in Expert Systems with Applications
Verified
Journal Article
Closed access
SWSyed Muhammad Waqas
Syed Muhammad Waqas
YTYazhe Tang
Yazhe Tang
Fakhar AbbasFakhar Abbas
Fakhar Abbas
HCHongyang Chen
Hongyang Chen
MHMehboob Hussain
Mehboob Hussain

26 July 2023 in Computer Communications
Verified
Journal Article
Closed access
SWSyed Muhammad Waqas
Syed Muhammad Waqas
YTYazhe Tang
Yazhe Tang
LYLisu Yu
Lisu Yu
Fakhar AbbasFakhar Abbas
Fakhar Abbas

6 November 2025 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
SWSyed Muhammad Waqas
Syed Muhammad Waqas
ALAnhui Liang
Anhui Liang
Xingsi XueXingsi Xue
Xingsi Xue
FHFenghua Huang
Fenghua Huang
WLWenxi Liu
Wenxi Liu
Fakhar AbbasFakhar Abbas
Fakhar Abbas
The integration of space, air, and ground segments into unified space–air–ground integrated networks (SAGINs) enables low-latency, ubiquitous, and scalable computing. However, such systems face critical challenges: ground terminals suffer from weak satellite links, unmanned aerial vehicle (UAV)-based edge nodes have limited resources, and highly dynamic environments make it difficult to make efficient offloading and resource allocation decisions. Prior approaches often optimize either communication or computation in isolation and lack adaptability to real-time environmental feedback. This article presents a novel perception-aware hybrid-action deep reinforcement learning (DRL) framework for joint optimization of task offloading, beamforming, and resource allocation in SAGINs. To improve tractability, the original nonconvex problem is first decomposed using block coordinate descent (BCD) and approximated with successive convex approximation (SCA), generating a structured feasible action space. A soft actor–critic (SAC) agent then learns policies over this space, informed by real-time UAV perception via millimeter-wave radar and vision sensors that detect user density, link quality, and environmental blockages. The DRL agent operates over a hybrid action space, combining discrete offloading decisions with continuous controls such as beamforming weights, CPU frequency, and transmission power. We employ a constraint-aware action masking mechanism that prunes infeasible hybrid actions violating delay, power, or signal-to-noise ratio limits, thereby accelerating learning while respecting SAGIN-specific constraints. Extensive simulations show that the proposed framework significantly outperforms greedy, no-perception DRL, and state-of-the-art DRL offloading algorithms in reducing latency and energy consumption, while improving offloading success and resource stability. These results highlight the effectiveness of combining analytical optimization structure with adaptive perception-driven learning for robust and scalable control in future SAGINs.

14 May 2026 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
Xingsi XueXingsi Xue
Xingsi Xue
SWSyed Muhammad Waqas
Syed Muhammad Waqas
Mohammed AlenaziMohammed Alenazi
Mohammed Alenazi
SKSaru Kumari
Saru Kumari
CCChien-Ming Chen
Chien-Ming Chen
NLNan Li
Nan Li
Multimodal remote sensing has become an essential paradigm for land-cover classification by exploiting complementary information from heterogeneous sensors. However, in practical applications, one or more modalities are often unavailable at inference time due to sensor failure, acquisition constraints, or environmental conditions. Such missing-modality scenarios introduce a systematic form of data quality degradation, which fundamentally undermines the reliability of conventional multimodal models and limits the effectiveness of manually designed compensation strategies. To address this challenge, this paper proposes PE-MMRSS, an automated encoder architecture design framework based on evolutionary neural architecture search for multimodal remote sensing classification under missing-modality conditions. Instead of relying on handcrafted hallucination architectures, PE-MMRSS constructs a unified and generic search space that jointly accommodates encoders associated with both available and missing modalities. A dual-part encoding scheme is designed to represent coupled teacher-student architectures, enabling effective evolutionary optimization. To further improve search efficiency, a lightweight LSTM-based Structural Compatibility Predictor is introduced to estimate the relative effectiveness of candidate architectures using early-stop validation signals, thereby significantly reducing the computational cost of architecture evaluation. Experimental evaluations across a diverse set of multimodal remote sensing benchmarks (Huston2013 and Augsburg) show that the proposed method achieves superior performance compared with the manually designed baselines in a wide range of missing-modality scenarios. PE-MMRSS achieves improvements of up to 0.5%-1.6% in overall accuracy and Kappa coefficient compared with the strongest handcrafted baseline, while reducing the architecture search cost to approximately 1.4 GPU-days. These results indicate that automated neural architecture search provides an effective and efficient solution for mitigating data quality degradation in multimodal remote sensing systems.

1 August 2026 in IEEE Transactions on Emerging Topics in Computational Intelligence
Verified
Journal Article
Closed access
Xingsi XueXingsi Xue
Xingsi Xue
SWSyed Muhammad Waqas
Syed Muhammad Waqas
WLWenxi Liu
Wenxi Liu
Jerry Chun-Wei LinJerry Chun-Wei Lin
Jerry Chun-Wei Lin
Knowledge Graphs (KGs) enable structured representation of domain-specific knowledge, but their integration is challenged by ontology heterogeneity due to varying terminologies and levels of granularity. KG alignment addresses this issue by identifying correspondences between semantically similar entities across diverse KGs, relying heavily on the effective combination of Similarity Features (SFs). However, manually determining the optimal combination of SFs is inefficient and often inaccurate, especially given the complex interactions among entities in different domains. To address this problem, this paper develops an Adaptive-Designed Particle Swarm Optimization (ADPSO) to automate SF combination for KG alignment. First, an adaptive framework is developed to dynamically adjust solution movement and evaluation rules, mitigating premature convergence. Second, a novel evaluation rule construction method is proposed, leveraging Genetic Programming (GP) to refine the fitness function, enhancing solution evaluation accuracy. Finally, a new velocity update mechanism is designed using a success update archive and GP, improving search efficiency. Experiments on diverse KG alignment tasks demonstrate that ADPSO significantly enhances matching accuracy and efficiency, achieving superior performance compared to state-of-the-art methods with an average f-measure of 0.90.

1 January 2026 in IEEE Transactions on Wireless Communications
Verified
Journal Article
Closed access
SWSyed Muhammad Waqas
Syed Muhammad Waqas
FHFenghua Huang
Fenghua Huang
Fakhar AbbasFakhar Abbas
Fakhar Abbas
SRSalman Raza
Salman Raza
WLWenxi Liu
Wenxi Liu
Xingsi XueXingsi Xue
Xingsi Xue