Abstract
In order to improve network performance, including reducing computation delay, transmission delay and bandwidth consumption, edge computing and caching technologies are introduced to the fifth-generation wireless network (5G). However, the volume of edge resources is limited, while the number and complexity of tasks in the network are increasing sharply. Therefore, how to provide the most efficient service for network users with limited resources is an urgent problem to be solved. Thus, improving the utilization rate of communication, computing and caching resources in the network is an important issue. The diversification of network resources brings difficulties to network management. The joint resource allocation problem is difficult to be solved by traditional approaches. With the development of Artificial Intelligence (AI) technology, these AI algorithms have been applied to joint resource allocation problems to solve complex decision-making problems. In this paper, we first summarize the AI-based joint resources allocation schemes. Then, an AI-assisted intelligent wireless network architecture is proposed. Finally, based on the proposed architecture, we use deep Q-network (DQN) algorithm to figure out the complex and high-dimensional joint resource allocation problem. Simulation results show that the algorithm has good convergence characteristics, proposed architecture and the joint resource allocation scheme achieve better performance compared to other resource allocation schemes.
-
Author Keywords
- Artificial intelligence ,
- deep Q-network ,
- resource management ,
- edge computing and caching ,
- the fifth-generation wireless network (5G)
-
IEEE Keywords
- Task analysis ,
- Resource management ,
- Wireless networks ,
- Streaming media ,
- Artificial intelligence ,
- Servers ,
- Computer architecture
Introduction
After the frozen of 5G mobile network specification Release15 on June 2018, the first nationwide commercial 5G networklaunched in South Korea on April 2019 [1]. Meanwhile, thestudy of roadmap and enabling technologies beyond the 5Gmobile network are in progress in both the academic andtelecommunication industries.The key challenges for 5G come from the increasingdemands for higher data rate and spectrum utilization, lowerlatency, better and always-on connectivity to support commu-nication among things and devices. Thus, software-definednetworking (SDN) and network functions virtualization (N-FV) are vital in the 5G and beyond 5G mobile networkarchitecture and implementation. For example, SDN con-trollers are deployed to regulate the traffic and organize thevirtualized network devices, which can significantly improvethe efficiency and performance of the network [2] [3].From the aspects of application scenarios, typical novelapplications of Augmented Reality (AR) and Virtual Reality(VR) online game, real-time video processing and intelligenthealth care over the wireless network prompt the deploy-ment of cache and fog/edge computing devices in wirelessnetworks [4] [5]. The deployment and scheduling of storageand computing resources will be a lot more complicated thanwhat came before. The development of AI technology bringsconvenience to all aspects of our life, such as smart medicalcare, smart factories, smart cities and so on [6] [7] [8].In the practical wireless system, many traditional resourceoptimization methods in wireless communication network-s are becoming much more performance-constrained andcomplicated in complex scenarios [9] [10]. Currently, AItechnology which mainly includes machine learning anddeep learning can extract useful information from wirelesssystems [11], learn and make decisions from the dynamicenvironment, are considered as potential solutions for typ-ical complex and previously intractable problems in futurewireless network [12]. In view of those observations, it isnecessary to review how to apply AI technology to solve the complicated decision-making problem and boost networkperformance. Subsequently, an intelligent wireless networkframework with self-adaptation and self-optimization capa-bility is proposed, and an example application is addressedto illustrate the workflow. Computer evaluations are imple-mented to show the gain over traditional schemes. Finally,concluding remarks and implementation challenges for AI-based wireless network resource management are addressed.
Conclusion
In this paper, we proposed an AI-assisted intelligent wire-less network architecture based on 5G, edge computing andcaching technologies. Through the learning of environmentstatus and task attributes by AI algorithms, the AI algo-rithms based orchestrator can reasonably allocate resourcesaccording to different situations. Among them, we chooseDQN algorithm to solve the complex and high-dimensionalresource allocation problem. Simulation results show that theproposed resource allocation scheme has good convergencecharacteristics, and the joint allocation of communication,computing and caching resources bring higher benefits. Fu-ture work should focus on improving the mobility of thesystem.
About KSRA
The Kavian Scientific Research Association (KSRA) is a non-profit research organization to provide research / educational services in December 2013. The members of the community had formed a virtual group on the Viber social network. The core of the Kavian Scientific Association was formed with these members as founders. These individuals, led by Professor Siavosh Kaviani, decided to launch a scientific / research association with an emphasis on education.
KSRA research association, as a non-profit research firm, is committed to providing research services in the field of knowledge. The main beneficiaries of this association are public or private knowledge-based companies, students, researchers, researchers, professors, universities, and industrial and semi-industrial centers around the world.
Our main services Based on Education for all Spectrum people in the world. We want to make an integration between researches and educations. We believe education is the main right of Human beings. So our services should be concentrated on inclusive education.
The KSRA team partners with local under-served communities around the world to improve the access to and quality of knowledge based on education, amplify and augment learning programs where they exist, and create new opportunities for e-learning where traditional education systems are lacking or non-existent.
FULL Paper PDF file:
AI Inspired Intelligent ResourceManagement in Future Wireless NetworkBibliography
author,
S. Fu, F. Yang, and Y. Xiao,
Year
2020
Title
AI Inspired Intelligent Resource Management in Future Wireless Network
Publish in
in IEEE Access, vol. 8, pp. 22425-22433, 2020,
Doi
10.1109/ACCESS.2020.2968554.
PDF reference and original file: Click here
Somayeh Nosrati was born in 1982 in Tehran. She holds a Master's degree in artificial intelligence from Khatam University of Tehran.
Professor Siavosh Kaviani was born in 1961 in Tehran. He had a professorship. He holds a Ph.D. in Software Engineering from the QL University of Software Development Methodology and an honorary Ph.D. from the University of Chelsea.
Nasim Gazerani was born in 1983 in Arak. She holds a Master's degree in Software Engineering from UM University of Malaysia.
