Q-LEARNING BASED RESOURCE MANAGEMENT IN 5G NETWORKS WITH CO-EXISTENCE OF HUMAN AND MACHINE TYPE COMMUNICATIONS
Amaal Samir Abd El-Hameed;
Abstract
The 4th generation (4G) wireless systems and beyond are used to support both Human to Human (H2H) and Machine to Machine (M2M) communications. Since M2M communication is expected to be massive and the characteristics of M2M traffic is different compared to H2H network traffic in both data packets size and transmission time, the LTE-Advanced system must evolve to face this new type of users.
In LTE-A, M2M communication devices share the same random access channel (RACH) with the H2H communication devices, consequently the RACH has become a new bottleneck in the LTE-A system. In order to solve this problem, we propose a new random access channel scheme based on Q-learning approach to reduce the congestion problem. The scheme adaptively divides the available preambles between both M2M and H2H devices in a way that provides an acceptable service for the H2H devices and maximizes the number of active M2M devices. The results indicate that the proposed approach provides high random access channel success probability for both M2M and H2H devices even with the huge number of M2M devices with 95% and with 50% raise over the other systems not use the Q-learning.
In the second part of the thesis, we study the heterogeneous networks with two layers; Macro-cell and Pico-cell with the co-existence of both H2H and M2M communications. The problem of unbalancing load between the two layers appears as most of the users choose to connect to Macro-cell than Pico-cell due to its larger coverage area, which increase the users' blocking probability. In this work we propose a new scheme uses the Q-learning algorithm to associate the users to different eNB to reduce the users' blocking probability, also it balances the network load between the two layers to increase the network performance. The results indicate that the proposed approach provides low blocking probability by 20% for both M2M and H2H devices even with the huge number of M2M devices and low uplink transmission power for M2M devices by 50%
In LTE-A, M2M communication devices share the same random access channel (RACH) with the H2H communication devices, consequently the RACH has become a new bottleneck in the LTE-A system. In order to solve this problem, we propose a new random access channel scheme based on Q-learning approach to reduce the congestion problem. The scheme adaptively divides the available preambles between both M2M and H2H devices in a way that provides an acceptable service for the H2H devices and maximizes the number of active M2M devices. The results indicate that the proposed approach provides high random access channel success probability for both M2M and H2H devices even with the huge number of M2M devices with 95% and with 50% raise over the other systems not use the Q-learning.
In the second part of the thesis, we study the heterogeneous networks with two layers; Macro-cell and Pico-cell with the co-existence of both H2H and M2M communications. The problem of unbalancing load between the two layers appears as most of the users choose to connect to Macro-cell than Pico-cell due to its larger coverage area, which increase the users' blocking probability. In this work we propose a new scheme uses the Q-learning algorithm to associate the users to different eNB to reduce the users' blocking probability, also it balances the network load between the two layers to increase the network performance. The results indicate that the proposed approach provides low blocking probability by 20% for both M2M and H2H devices even with the huge number of M2M devices and low uplink transmission power for M2M devices by 50%
Other data
| Title | Q-LEARNING BASED RESOURCE MANAGEMENT IN 5G NETWORKS WITH CO-EXISTENCE OF HUMAN AND MACHINE TYPE COMMUNICATIONS | Other Titles | ادارة الموارد في شبكات الجيل الخامس القائمة علي التعلم الذاتي في وجود اتصالات الانسان و الاله | Authors | Amaal Samir Abd El-Hameed | Issue Date | 2021 |
Attached Files
| File | Size | Format | |
|---|---|---|---|
| BB10237.pdf | 879.84 kB | Adobe PDF | View/Open |
Similar Items from Core Recommender Database
Items in Ain Shams Scholar are protected by copyright, with all rights reserved, unless otherwise indicated.