Deep learning for resource allocation in cellular wireless networks
| dc.contributor.author | Ahmed, Kazi Ishfaq | |
| dc.contributor.examiningcommittee | Ashraf, Ahmed (Electrical and Computer Engineering) Wang, Yang (Computer Science) | en_US |
| dc.contributor.supervisor | Hossain, Ekram (Electrical and Computer Engineering) | en_US |
| dc.date.accessioned | 2019-06-21T18:03:29Z | |
| dc.date.available | 2019-06-21T18:03:29Z | |
| dc.date.issued | 2019-06-10 | en_US |
| dc.date.submitted | 2019-06-11T16:34:04Z | en |
| dc.degree.discipline | Electrical and Computer Engineering | en_US |
| dc.degree.level | Master of Science (M.Sc.) | en_US |
| dc.description.abstract | Optimal resource allocation is a fundamental challenge for dense and heterogeneous wireless networks with massive wireless connections. Traditionally, due to the non-convex nature of the optimization problem, resource allocation is done using some heuristic approaches such as exhaustive search, genetic algorithms, combinatorial and branch and bound techniques. These methods are computationally expensive and therefore not appealing for large-scale heterogeneous cellular networks with ultra-dense base station (BS) deployments, massive connections and diverse QoS requirements for different classes of users. As a result, the next generation of wireless networks will require a paradigm shift from traditional resource allocation mechanisms. Deep learning (DL) is a powerful tool where a multi-layer neural network can be trained to model a resource management algorithm using network data. Therefore, resource allocation decisions can be obtained without intensive online computations which would be required otherwise for the solution of resource allocation problems. In this thesis, I develop a deep learning based resource allocation framework for multi-cell wireless networks with an objective to maximizing the total network throughput. In addition, I explore the deep reinforcement learning (DRL) approach to perform a near-optimal downlink power allocation for multi-cell wireless networks. Specifically, I use a deep Q-learning (DQL) strategy to achieve near-optimal power allocation policy. For benchmarking the proposed approaches, I use a Genetic Algorithm (GA) to obtain near-optimal resource allocation solution. I compare the proposed power allocation scheme with other traditional power allocation schemes by running numerous simulations. | en_US |
| dc.description.note | October 2019 | en_US |
| dc.identifier.citation | Ahmed, Kazi Ishfaq, Hina Tabassum, and Ekram Hossain. "Deep learning for radio resource allocation in multi-cell networks." IEEE Network (2019). | en_US |
| dc.identifier.citation | Ahmed, Kazi Ishfaq, and Ekram Hossain. "A Deep Q-Learning Method for Downlink Power Allocation in Multi-Cell Networks." arXiv preprint arXiv:1904.13032 (2019) | en_US |
| dc.identifier.uri | http://hdl.handle.net/1993/33998 | |
| dc.language.iso | eng | en_US |
| dc.rights | open access | en_US |
| dc.subject | Deep learning, Resource allocation, Wireless network, Deep reinforcement learning, | en_US |
| dc.title | Deep learning for resource allocation in cellular wireless networks | en_US |
| dc.type | master thesis | en_US |