Xiaonan Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Xiaonan Liu has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Yansha Deng
King's College London
7 shared publications - Arumugam Nallanathan
Queen Mary University of London
5 shared publications - Haipeng Yao
Beijing University of Posts and Telecommunications
3 shared publications - Tharmalingam Ratnarajah
San Diego State University
3 shared publications - Yongkang Gong
School of Computer Science and Technology, Shandong University (SDU), Qingdao, China
3 shared publications - Liyana Adilla Burhanuddin
Universiti Teknologi Malaysia
2 shared publications
Latest Publications
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25, 18669-18684
Federated Dropout: Convergence Analysis and Resource Allocation
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25, 18669-18684
Federated Dropout: Convergence Analysis and Resource Allocation
- 26 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network Science and Engineering
- Vol. 13, 327-343
Adaptive Model Pruning in Decentralized Federated Learning
Federated Learning (FL) has emerged as a promising decentralized learning framework that facilitates efficient model training across multiple devices, leveraging insights from a variety of data sources while preserving the privacy of devices. However, the distributed architecture of FL introduces high communication cost between the central server and mobile devices, especially for mobile devices far away from the server. Meanwhile, the computation and communication latency can be high on devices with limited computation capability and wireless resources when updating large-scale learning models. Therefore, in this paper, we consider a decentralized distributed FL with model pruning over wireless networks. In each global communication round, the K-means algorithm is deployed to perform dynamic clustering of devices based on their geographical proximity and device-to-device (D2D) communication technique is used to share the updated model weights. Then, we mathematically analyze the computation and communication latency and convergence of the proposed system. To improve convergence rate, guarantee learning accuracy, and decrease computation and communication latency, Karush-Kuhn-Tucker (KKT) conditions are employed to jointly optimize the pruning ratio and bandwidth allocation. Simulation results demonstrate that the proposed FL algorithm necessitates only 88% of the training time required by the system employing equal pruning and 80% of the training time needed by the system without pruning, while achieving a testing accuracy comparable to that of the latter.
- 26 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network Science and Engineering
- Vol. 13, 327-343
Adaptive Model Pruning in Decentralized Federated Learning
Federated Learning (FL) has emerged as a promising decentralized learning framework that facilitates efficient model training across multiple devices, leveraging insights from a variety of data sources while preserving the privacy of devices. However, the distributed architecture of FL introduces high communication cost between the central server and mobile devices, especially for mobile devices far away from the server. Meanwhile, the computation and communication latency can be high on devices with limited computation capability and wireless resources when updating large-scale learning models. Therefore, in this paper, we consider a decentralized distributed FL with model pruning over wireless networks. In each global communication round, the K-means algorithm is deployed to perform dynamic clustering of devices based on their geographical proximity and device-to-device (D2D) communication technique is used to share the updated model weights. Then, we mathematically analyze the computation and communication latency and convergence of the proposed system. To improve convergence rate, guarantee learning accuracy, and decrease computation and communication latency, Karush-Kuhn-Tucker (KKT) conditions are employed to jointly optimize the pruning ratio and bandwidth allocation. Simulation results demonstrate that the proposed FL algorithm necessitates only 88% of the training time required by the system employing equal pruning and 80% of the training time needed by the system without pruning, while achieving a testing accuracy comparable to that of the latter.
- 23 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (14) , 26782-26798
Precoding Optimization for Rate Splitting Enabled Internet of Underwater Things Over Optical Wireless Underwater Turbulent Channels
With the booming development of the underwater Internet of Things, underwater devices and data volume are explosively increasing. Thanks to the advantages of high efficiency and low complexity, the emerging rate splitting multiple access (RSMA) can be integrated with underwater optical wireless communication (UOWC), which can help improve resource utilization and system performance. However, the oceanic characteristics are distinctive and pose great challenges for RSMA-enabled UOWC networks. Precoding design is crucial for benefitting from RSMA-enabled systems, and thus, we explore the precoding optimization for RSMA-enabled UOWC networks over underwater turbulent channels. Specifically, the RSMA-enabled UOWC network is modeled, where a combined turbulent channel involving the effects of absorption, scattering, and underwater turbulence is considered, and a message transmission scheme with rate splitting design is analyzed. A precoding problem is formulated to maximize the ergodic sum rate (ESR) under the constraint of total transmitted power. Since the accurate expression of ESR is a nonclosed form and intractable, we first propose a consecutive Fenton-Wilkinson (CFW)-based ergodic rate approximation approach to solve this formulation problem. Fenton-Wilkinson moment matching is consecutively employed to achieve a closed-form and low-complexity expression of the ergodic rate, as only Fenton-Wilkinson moment matching method can offer closed-form solutions for underlying parameters of the approximating log-normal distributions. For further solving the nonconvex precoding problem, the successive convex approximation (SCA) approach is adopted, which is particularly applicable for resource-limited underwater environments. The effectiveness of the proposed CFW approach and precoding strategy is evaluated by extensive simulation results for various user deployments and different network loads.
- 23 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (14) , 26782-26798
Precoding Optimization for Rate Splitting Enabled Internet of Underwater Things Over Optical Wireless Underwater Turbulent Channels
With the booming development of the underwater Internet of Things, underwater devices and data volume are explosively increasing. Thanks to the advantages of high efficiency and low complexity, the emerging rate splitting multiple access (RSMA) can be integrated with underwater optical wireless communication (UOWC), which can help improve resource utilization and system performance. However, the oceanic characteristics are distinctive and pose great challenges for RSMA-enabled UOWC networks. Precoding design is crucial for benefitting from RSMA-enabled systems, and thus, we explore the precoding optimization for RSMA-enabled UOWC networks over underwater turbulent channels. Specifically, the RSMA-enabled UOWC network is modeled, where a combined turbulent channel involving the effects of absorption, scattering, and underwater turbulence is considered, and a message transmission scheme with rate splitting design is analyzed. A precoding problem is formulated to maximize the ergodic sum rate (ESR) under the constraint of total transmitted power. Since the accurate expression of ESR is a nonclosed form and intractable, we first propose a consecutive Fenton-Wilkinson (CFW)-based ergodic rate approximation approach to solve this formulation problem. Fenton-Wilkinson moment matching is consecutively employed to achieve a closed-form and low-complexity expression of the ergodic rate, as only Fenton-Wilkinson moment matching method can offer closed-form solutions for underlying parameters of the approximating log-normal distributions. For further solving the nonconvex precoding problem, the successive convex approximation (SCA) approach is adopted, which is particularly applicable for resource-limited underwater environments. The effectiveness of the proposed CFW approach and precoding strategy is evaluated by extensive simulation results for various user deployments and different network loads.
- 12 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 39 (4) , 254-261
Joint Sensing, Computation, and DAI Resource Management for Integrated Non-Terrestrial and Terrestrial Networks
Integrated Non-Terrestrial and Terrestrial Networks (NTN-TN) have been emerging as a promising architecture to facilitate state sensing, computation migration, and privacy protection for terrestrial users, where both low, medium and geostationary orbit satellites collaboratively support various tasks and allocate resources effectively. Nevertheless, an integration of sensing, computation, and artificial intelligence (AI) in dynamic NTN-TNs has certain challenges in achieving high-efficient task offloading and guaranteeing robust privacy protection. To address these issues, we introduce an intelligent NTN-TN framework, focusing on sensing, computation migration, and privacy protection. Furthermore, we develop a versatile and scalable experimental platform for NTN-TNs to validate related network functions. Simulation results show that the proposed framework reduces network latency compared with advanced benchmarks. Finally, we present potential research directions and innovative solutions in future NTN-TNs.
- 12 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 39 (4) , 254-261
Joint Sensing, Computation, and DAI Resource Management for Integrated Non-Terrestrial and Terrestrial Networks
Integrated Non-Terrestrial and Terrestrial Networks (NTN-TN) have been emerging as a promising architecture to facilitate state sensing, computation migration, and privacy protection for terrestrial users, where both low, medium and geostationary orbit satellites collaboratively support various tasks and allocate resources effectively. Nevertheless, an integration of sensing, computation, and artificial intelligence (AI) in dynamic NTN-TNs has certain challenges in achieving high-efficient task offloading and guaranteeing robust privacy protection. To address these issues, we introduce an intelligent NTN-TN framework, focusing on sensing, computation migration, and privacy protection. Furthermore, we develop a versatile and scalable experimental platform for NTN-TNs to validate related network functions. Simulation results show that the proposed framework reduces network latency compared with advanced benchmarks. Finally, we present potential research directions and innovative solutions in future NTN-TNs.
- 24 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 74 (2) , 3545-3549
In-Orbit Computation and Security Authentication for Satellite-Ground Twin Networks
Satellite-ground twin networks are considered a promising network structure, which can relieve network congestion and provide pervasive intelligent services. Since massive terrestrial users are not willing to share mobile data, it is necessary to design a novel security authentication method. In this paper, we consider a two-layer Stackelberg game model and propose a deep federated meta reinforcement learning (LSTDFMRL) framework based on Lyapunov stability theory to orchestrate the cycle frequency, channel assignment and block size. Simulation results confirm that the proposed LST-DFMRL framework outperforms existing baseline methods in terms of server profits, network throughput and security authentication overhead.
- 24 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 74 (2) , 3545-3549
In-Orbit Computation and Security Authentication for Satellite-Ground Twin Networks
Satellite-ground twin networks are considered a promising network structure, which can relieve network congestion and provide pervasive intelligent services. Since massive terrestrial users are not willing to share mobile data, it is necessary to design a novel security authentication method. In this paper, we consider a two-layer Stackelberg game model and propose a deep federated meta reinforcement learning (LSTDFMRL) framework based on Lyapunov stability theory to orchestrate the cycle frequency, channel assignment and block size. Simulation results confirm that the proposed LST-DFMRL framework outperforms existing baseline methods in terms of server profits, network throughput and security authentication overhead.
- 11 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 72 (1053587X) , 4395-4411
Adaptive Model Pruning and Personalization for Federated Learning Over Wireless Networks
Federated learning (FL) enables distributed learning across edge devices while protecting data privacy. However, the learning accuracy decreases due to the heterogeneity of devices’ data, and the computation and communication latency increase when updating large-scale learning models on devices with limited computational capability and wireless resources. We consider a FL framework with partial model pruning and personalization to overcome these challenges. This framework splits the learning model into a global part with model pruning shared with all devices to learn data representations and a personalized part to be fine-tuned for a specific device, which adapts the model size during FL. Our approach reduces both computation and communication latency and increases the learning accuracy for devices with non-independent and identically distributed data. The computation and communication latency and convergence of the proposed FL framework are mathematically analyzed. To maximize the convergence rate and guarantee learning accuracy, Karush–Kuhn–Tucker (KKT) conditions are deployed to jointly optimize the pruning ratio and bandwidth allocation. Finally, experimental results demonstrate that the proposed FL framework achieves similar learning accuracy and a remarkable reduction of approximately 50% computation and communication latency compared with FL with partial model personalization.
- 11 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 72 (1053587X) , 4395-4411
Adaptive Model Pruning and Personalization for Federated Learning Over Wireless Networks
Federated learning (FL) enables distributed learning across edge devices while protecting data privacy. However, the learning accuracy decreases due to the heterogeneity of devices’ data, and the computation and communication latency increase when updating large-scale learning models on devices with limited computational capability and wireless resources. We consider a FL framework with partial model pruning and personalization to overcome these challenges. This framework splits the learning model into a global part with model pruning shared with all devices to learn data representations and a personalized part to be fine-tuned for a specific device, which adapts the model size during FL. Our approach reduces both computation and communication latency and increases the learning accuracy for devices with non-independent and identically distributed data. The computation and communication latency and convergence of the proposed FL framework are mathematically analyzed. To maximize the convergence rate and guarantee learning accuracy, Karush–Kuhn–Tucker (KKT) conditions are deployed to jointly optimize the pruning ratio and bandwidth allocation. Finally, experimental results demonstrate that the proposed FL framework achieves similar learning accuracy and a remarkable reduction of approximately 50% computation and communication latency compared with FL with partial model personalization.
- Top Cited Papers
- 23 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 72 (9) , 5532-5546
Computation and Privacy Protection for Satellite-Ground Digital Twin Networks
Satellite-ground integrated heterogeneous networks can relieve network congestion, release network resources and provide ubiquitous intelligence services for terrestrial users. Furthermore, digital twin technology can enable nearly-instant data mapping from the physical world to digital systems. The integration between satellite-ground integrated heterogeneous networks and digital twin alleviates the gap between data analyses and physical unities. However, the current challenges, such as the pricing policy, the stochastic task arrivals, the time-varying satellite locations, mutual channel interference, and resource scheduling mechanisms between the users and cloud servers, severely affect the improvement of quality of service. Hence, we establish a blockchain-aided Stackelberg game model for maximizing the pricing profits and network throughput in terms of minimizing privacy overhead, which is able to perform computation offloading, decrease channel interference, and improve privacy protection. Due to the long-term task queue in Stackelberg model, we propose a Lyapunov stability theory-based model-agnostic meta-learning aided multi-agent deep federated reinforcement learning framework to transfer the long-term task queue into the single time slot, and then optimize the central processing unit frequency, channel selection, task-offloading decision, block size, and cloud server price, which facilitate the integration of communication, computation, and block resources. Subsequently, several performance analyses show that the proposed learning framework can strengthen the privacy protection, approach the optimal time average function, and fulfill the long-term average queue size via lower computational complexity. Finally, our simulation results indicate that the proposed learning framework is superior to the existing baseline methods in terms of network throughput, channel interference, cloud server profits, and privacy overhead.
- 23 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 72 (9) , 5532-5546
Top Cited PapersComputation and Privacy Protection for Satellite-Ground Digital Twin Networks
Satellite-ground integrated heterogeneous networks can relieve network congestion, release network resources and provide ubiquitous intelligence services for terrestrial users. Furthermore, digital twin technology can enable nearly-instant data mapping from the physical world to digital systems. The integration between satellite-ground integrated heterogeneous networks and digital twin alleviates the gap between data analyses and physical unities. However, the current challenges, such as the pricing policy, the stochastic task arrivals, the time-varying satellite locations, mutual channel interference, and resource scheduling mechanisms between the users and cloud servers, severely affect the improvement of quality of service. Hence, we establish a blockchain-aided Stackelberg game model for maximizing the pricing profits and network throughput in terms of minimizing privacy overhead, which is able to perform computation offloading, decrease channel interference, and improve privacy protection. Due to the long-term task queue in Stackelberg model, we propose a Lyapunov stability theory-based model-agnostic meta-learning aided multi-agent deep federated reinforcement learning framework to transfer the long-term task queue into the single time slot, and then optimize the central processing unit frequency, channel selection, task-offloading decision, block size, and cloud server price, which facilitate the integration of communication, computation, and block resources. Subsequently, several performance analyses show that the proposed learning framework can strengthen the privacy protection, approach the optimal time average function, and fulfill the long-term average queue size via lower computational complexity. Finally, our simulation results indicate that the proposed learning framework is superior to the existing baseline methods in terms of network throughput, channel interference, cloud server profits, and privacy overhead.
- 8 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (6) , 5985-5999
Adaptive Federated Pruning in Hierarchical Wireless Networks
Federated Learning (FL) is a promising privacy-preserving distributed learning framework where a server aggregates models updated by multiple devices without accessing their private datasets. Hierarchical FL (HFL), as a device-edge-cloud aggregation hierarchy, can enjoy both the cloud server’s access to more datasets and the edge servers’ efficient communications with devices. However, the learning latency increases with the HFL network scale due to the increasing number of edge servers and devices with limited local computation capability and communication bandwidth. To address this issue, in this paper, we introduce model pruning for HFL in wireless networks to reduce the neural network scale. We present the convergence analysis of an upper on the l2-norm of gradients for HFL with model pruning, analyze the computation and communication latency of the proposed model pruning scheme, and formulate an optimization problem to maximize the convergence rate under a given latency threshold by jointly optimizing the pruning ratio and wireless resource allocation. By decoupling the optimization problem and using Karush–Kuhn–Tucker (KKT) conditions, closed-form solutions of pruning ratio and wireless resource allocation are derived. Simulation results show that our proposed HFL with model pruning achieves similar learning accuracy compared with the HFL without model pruning and reduces about 50% communication cost.
- 8 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (6) , 5985-5999
Adaptive Federated Pruning in Hierarchical Wireless Networks
Federated Learning (FL) is a promising privacy-preserving distributed learning framework where a server aggregates models updated by multiple devices without accessing their private datasets. Hierarchical FL (HFL), as a device-edge-cloud aggregation hierarchy, can enjoy both the cloud server’s access to more datasets and the edge servers’ efficient communications with devices. However, the learning latency increases with the HFL network scale due to the increasing number of edge servers and devices with limited local computation capability and communication bandwidth. To address this issue, in this paper, we introduce model pruning for HFL in wireless networks to reduce the neural network scale. We present the convergence analysis of an upper on the l2-norm of gradients for HFL with model pruning, analyze the computation and communication latency of the proposed model pruning scheme, and formulate an optimization problem to maximize the convergence rate under a given latency threshold by jointly optimizing the pruning ratio and wireless resource allocation. By decoupling the optimization problem and using Karush–Kuhn–Tucker (KKT) conditions, closed-form solutions of pruning ratio and wireless resource allocation are derived. Simulation results show that our proposed HFL with model pruning achieves similar learning accuracy compared with the HFL without model pruning and reduces about 50% communication cost.
- Top Cited Papers
- 7 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Surveys & Tutorials
- Vol. 26 (1) , 571-618
Federated Learning and Meta Learning: Approaches, Applications, and Directions
Over the past few years, significant advancements have been made in the field of machine learning (ML) to address resource management, interference management, autonomy, and decision-making in wireless networks. Traditional ML approaches rely on centralized methods, where data is collected at a central server for training. However, this approach poses a challenge in terms of preserving the data privacy of devices. To address this issue, federated learning (FL) has emerged as an effective solution that allows edge devices to collaboratively train ML models without compromising data privacy. In FL, local datasets are not shared, and the focus is on learning a global model for a specific task involving all devices. However, FL has limitations when it comes to adapting the model to devices with different data distributions. In such cases, meta learning is considered, as it enables the adaptation of learning models to different data distributions using only a few data samples. In this tutorial, we present a comprehensive review of FL, meta learning, and federated meta learning (FedMeta). Unlike other tutorial papers, our objective is to explore how FL, meta learning, and FedMeta methodologies can be designed, optimized, and evolved, and their applications over wireless networks. We also analyze the relationships among these learning algorithms and examine their advantages and disadvantages in real-world applications.
- 7 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Surveys & Tutorials
- Vol. 26 (1) , 571-618
Top Cited PapersFederated Learning and Meta Learning: Approaches, Applications, and Directions
Over the past few years, significant advancements have been made in the field of machine learning (ML) to address resource management, interference management, autonomy, and decision-making in wireless networks. Traditional ML approaches rely on centralized methods, where data is collected at a central server for training. However, this approach poses a challenge in terms of preserving the data privacy of devices. To address this issue, federated learning (FL) has emerged as an effective solution that allows edge devices to collaboratively train ML models without compromising data privacy. In FL, local datasets are not shared, and the focus is on learning a global model for a specific task involving all devices. However, FL has limitations when it comes to adapting the model to devices with different data distributions. In such cases, meta learning is considered, as it enables the adaptation of learning models to different data distributions using only a few data samples. In this tutorial, we present a comprehensive review of FL, meta learning, and federated meta learning (FedMeta). Unlike other tutorial papers, our objective is to explore how FL, meta learning, and FedMeta methodologies can be designed, optimized, and evolved, and their applications over wireless networks. We also analyze the relationships among these learning algorithms and examine their advantages and disadvantages in real-world applications.
- 23 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 9 (6) , 1596-1609
Inter-Cell Interference Mitigation for Cellular-Connected UAVs Using MOSDS-DQN
In 5G and beyond, UAVs are integrated into cellular networks as new aerial mobile users to support many applications and provide higher probability of line-of-sight (LoS) transmission to base stations (BSs). Nevertheless, due to limited frequency bandwidth and spectrum resource reuse when BSs serving terrestrial users (TUEs) and UAVs, it causes severe downlink interference to TUEs, especially when the network has a heavy load. Thus, in this paper, we study the performance of radio connectivity of UAVs and TUEs in an urban area and introduce a downlink inter-cell interference coordination mechanism. Then, we propose adaptive cell muting interference and resource allocation scheduling schemes. A value function approximation solution (VFA), Tabular-Q, and Deep-Q Network (DQN) are proposed to maximize the long-term network throughput of TUEs while guaranteeing the data rate requirements of UAVs. With increasing number of UAVs and TUEs and dynamic wireless environment, we further propose a Muting Optimization Scheme and Dynamic time-frequency Scheduling (MOSDS) algorithm to increase throughput and satisfactory level for both UAVs and TUEs. Simulation results show that the proposed algorithms achieve 80% performance improvement of throughput of UAV and TUE networks and mitigate the interference among them. Also, the proposed MOSDS-DQN shows 18% improvement compared to the DQN algorithm.
- 23 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 9 (6) , 1596-1609
Inter-Cell Interference Mitigation for Cellular-Connected UAVs Using MOSDS-DQN
In 5G and beyond, UAVs are integrated into cellular networks as new aerial mobile users to support many applications and provide higher probability of line-of-sight (LoS) transmission to base stations (BSs). Nevertheless, due to limited frequency bandwidth and spectrum resource reuse when BSs serving terrestrial users (TUEs) and UAVs, it causes severe downlink interference to TUEs, especially when the network has a heavy load. Thus, in this paper, we study the performance of radio connectivity of UAVs and TUEs in an urban area and introduce a downlink inter-cell interference coordination mechanism. Then, we propose adaptive cell muting interference and resource allocation scheduling schemes. A value function approximation solution (VFA), Tabular-Q, and Deep-Q Network (DQN) are proposed to maximize the long-term network throughput of TUEs while guaranteeing the data rate requirements of UAVs. With increasing number of UAVs and TUEs and dynamic wireless environment, we further propose a Muting Optimization Scheme and Dynamic time-frequency Scheduling (MOSDS) algorithm to increase throughput and satisfactory level for both UAVs and TUEs. Simulation results show that the proposed algorithms achieve 80% performance improvement of throughput of UAV and TUE networks and mitigate the interference among them. Also, the proposed MOSDS-DQN shows 18% improvement compared to the DQN algorithm.