A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
IRS‐Aided Mobile Edge Computing: From Optimization to Learning
In this chapter, we explore the optimization‐based and deep learning approaches for resource allocation of intelligent reflecting surface (IRS)‐aided multiuser mobile edge computing (MEC) systems. The total completed task‐input bits (TCTB) of all user equipment (UEs) with limited energy budgets is maximized during a given time slot through joint IRS reflecting coefficients design, receive beamforming design, and energy partition optimization. A three‐step block coordinate descending (BCD) algorithm is first introduced to effectively solve the non‐convex TCTB maximization problem with guaranteed convergence. In order to reduce the computational complexity and facilitate lightweight online implementation of the optimization algorithm, two deep learning architectures are further overviewed. The first one takes channel state information (CSI) as input, while the second one exploits the UEs' locations only for online inference. The two data‐driven approaches are trained using data samples generated by the BCD algorithm via supervised learning. The simulation results reveal a close match between the performance of the optimization‐based BCD algorithm and the low‐complexity learning‐based architectures, all with superior performance to existing schemes in both cases with perfect and imperfect input features. Importantly, the location‐only deep learning method is shown to offer a particularly practical and robust solution, alleviating the need for CSI estimation and feedback when line‐of‐sight (LoS) direct links exist between UEs and the access point (AP).
IRS‐Aided Mobile Edge Computing: From Optimization to Learning
In this chapter, we explore the optimization‐based and deep learning approaches for resource allocation of intelligent reflecting surface (IRS)‐aided multiuser mobile edge computing (MEC) systems. The total completed task‐input bits (TCTB) of all user equipment (UEs) with limited energy budgets is maximized during a given time slot through joint IRS reflecting coefficients design, receive beamforming design, and energy partition optimization. A three‐step block coordinate descending (BCD) algorithm is first introduced to effectively solve the non‐convex TCTB maximization problem with guaranteed convergence. In order to reduce the computational complexity and facilitate lightweight online implementation of the optimization algorithm, two deep learning architectures are further overviewed. The first one takes channel state information (CSI) as input, while the second one exploits the UEs' locations only for online inference. The two data‐driven approaches are trained using data samples generated by the BCD algorithm via supervised learning. The simulation results reveal a close match between the performance of the optimization‐based BCD algorithm and the low‐complexity learning‐based architectures, all with superior performance to existing schemes in both cases with perfect and imperfect input features. Importantly, the location‐only deep learning method is shown to offer a particularly practical and robust solution, alleviating the need for CSI estimation and feedback when line‐of‐sight (LoS) direct links exist between UEs and the access point (AP).
IRS‐Aided Mobile Edge Computing: From Optimization to Learning
Wu, Qingqing (editor) / Duong, Trung Q. (editor) / Ng, Derrick Wing Kwan (editor) / Schober, Robert (editor) / Zhang, Rui (editor) / Hu, Xiaoyan (author) / Wong, Kai‐Kit (author) / Masouros, Christos (author) / Jin, Shi (author)
2023-12-27
22 pages
Article/Chapter (Book)
Electronic Resource
English
Federated Learning for Energy-balanced Client Selection in Mobile Edge Computing
| BASE | 2021
A deep learning based approach for image retrieval extraction in mobile edge computing
| Springer Verlag | 2024
Towards Architect-Aided Computing Design
| SAGE Publications | 2006
Towards architect-aided computing design
| Online Contents | 2006
IEEE | 2022