AI-Assisted Energy-Efficient and Intelligent Routing for Reconfigurable Wireless Networks
ABSTARCT :
Intelligent network management for reconfigurable wireless networks such as 5G and beyond applications is crucial for many industrial applications, and has been the subject of ongoing research. This paper proposes an Artificial Intelligence(AI)-Assisted energy-efficient and intelligent routing, based on both energy efficiency prioritization and AI theory, in order to meet the exacting demands particularly in a real-world scenario. Specifically, to achieve network intelligence and quality of service (QoS), we use the AI theory to enhance routing adaptivity for intelligent network management in reconfigurable wireless networks. The software-defined networking idea is used to achieve this goal from a network-level perspective. To facilitate self-awareness, self-study, self-decision making, and self-configuration, we construct a mathematical model to convert the energy-efficient and intelligent routing problem into a multi-constraint optimal problem. Then an AI-assisted intelligent routing algorithm is designed to dynamically and adaptively change link weighs, which allows us to achieve optimal energy efficiency. Findings from our simulation suggest the potential of our proposed approach.
EXISTING SYSTEM :
? In contrast with the existing works, reference [79] exploit the channel correlation to more reliably configure the RIS reflection interaction.
? A tradeoff exists between imaging reconstruction speed and image quality.
? The intelligent manipulation of RISs is expected to achieve high-resolution sensing and high accuracy localization coexisting with the basic wireless communication functions, which can constantly share the abundant data and information in the time, frequency, and space domains.
? Supervised learning and unsupervised learning are two learning schemes that can be distinguished by the existense or absense of the output labels.
DISADVANTAGE :
? To address this problem and improve the benefits of RISs, artificial intelligence (AI)-based methods can be applied to design MAC protocols for RIS-aided wireless networks.
? In the distributed MAC protocol, an RL-based computational model can be employed by each user to solve the resource allocation and RIS configuration problems because no RIS channel-information exchange is required.
? The RL-based computational model is more suitable for small RISs to avoid potential dimensionality problems.
? Due to the typical hidden and exposed terminal problems, conventional carrier sensing techniques suffer from spectrum sensing deficiencies.
PROPOSED SYSTEM :
• In, low-overhead channel estimation algorithms have been proposed by taking advantage of the channel sparsity.
• In, the neural network that combines the data acquisition process with imaging process is proposed with information metamaterial coding enrolled.
• The proposed network connects the entire imaging system into a whole, and the author can find optimal metamaterial radiation patterns (codings) with less number of radiation pattern image requirement through the joint training while ensuring the imaging quality.
• To improve the estimation accuracy, the proposed framework take fully advantages of CNN in feature extraction and DReL in denoising.
ADVANTAGE :
? The intelligently controlled features of RISs lead to potential benefits for future wireless networks, such as their coverage enhancement, EE/SE performance improvement, leading to improved throughput and security.
? With the ever-increasing demands on wireless networks, research in wireless communications continues to focus on meeting the challenges of improving the energy efficiency (EE) versus spectral efficiency (SE) trade-offs.
? In practice, when designing an efficient AI-assisted MAC protocol for RIS-aided wireless networks, especially a distributed version, accurate spectrum sensing is quite critical for reducing collisions and interferences.
|