2018/03/18 by Roohollah Amiri, Amiri, Roohollah, Hani Mehrpouyan +9
Computer Science · Engineering · #Advanced MIMO Systems Optimization #Advanced Wireless Communication Technologies #Cooperative Communication and Network Coding #FOS: Computer and information sciences #Information Theory (cs.IT)
paper · pdf · doi:10.48550/arxiv.1803.06760
openalex publication_date 2018/03/18 · openalex created_date 2022/10/05 · openalex updated_date 2026/07/28
There is an increase in usage of smaller cells or femtocells to improve\nperformance and coverage of next-generation heterogeneous wireless networks\n(HetNets). However, the interference caused by femtocells to neighboring cells\nis a limiting performance factor in dense HetNets. This interference is being\nmanaged via distributed resource allocation methods. However, as the density of\nthe network increases so does the complexity of such resource allocation\nmethods. Yet, unplanned deployment of femtocells requires an adaptable and\nself-organizing algorithm to make HetNets viable. As such, we propose to use a\nmachine learning approach based on Q-learning to solve the resource allocation\nproblem in such complex networks. By defining each base station as an agent, a\ncellular network is modelled as a multi-agent network. Subsequently,\ncooperative Q-learning can be applied as an efficient approach to manage the\nresources of a multi-agent network. Furthermore, the proposed approach\nconsiders the quality of service (QoS) for each user and fairness in the\nnetwork. In comparison with prior work, the proposed approach can bring more\nthan a four-fold increase in the number of supported femtocells while using\ncooperative Q-learning to reduce resource allocation overhead.\n