Load Balancing in Heterogeneous Network Using Machine Learning Technique
L. Rajesh1, K.Bhoopathy Bagan2,Tamilarasan. K3,Meena. M4
1L.Rajesh,Department of Electronics Engineering, MIT, Anna University, Chennai, (Tamilnadu), India.
2K.Bhoopathy Bagan, Professor, Department of Electronics Engineering, MIT, Anna University, Chennai, (Tamilnadu), India.
3Tamilarasan.K, Department of Electronics Engineering, MIT, Anna University, Chennai, (Tamilnadu), India.
4Meena.M, Department of Electronics Engineering, MIT, Anna University, Chennai, (Tamilnadu), India.
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 1564-1569 | Volume-8 Issue-8, June 2019 | Retrieval Number: F4002048619/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In heterogeneous network load balancing is a major challenge in dense environment. In conventional method of user association, the base station to which the users get service from is determined by the SINR value. In dense environment there would be many pico and femto base stations available with low load the user may receive maximum SINR from the macro base station. This will cause over loading in the macro base station which will in turn reduces the Quality of service for the user. In this paper the load balancing is achieved by user association with the optimal BS that is determined with various factors such as SINR, service fairness, available resources, mobility of user, channel quality to improve the overall service rate and to reduce the variance in the service rate between users. A reinforcement learning algorithm is proposed with variance is service rate as the reward function and by considering the problem as N-arm bandit problem the load is balanced between the various base stations available by providing a good Overall service rate to the users.
Keyword: Reinforcement Learning, Heterogeneous Networks, Load Balancing, N-arm bandit.
Scope of the Article: Artificial Intelligence and Machine Learning.