IS3CQL : Intelligent sleep scheduling of small cells in high density 5G MIMO networks via incremental Deep Q learning optimizations

Abstract

Effective management of small cell networks is crucial as the demand for fast, lowlatency communication grows as a result of the development of 5G and Massive Multiple Input Multiple Output (M-MIMO) technology. Small cell sleep cycle scheduling in highdensity networks frequently encounter issues as it tries to strike a compromise between energy efficiency, quality of service (QoS), and maintaining a high throughput. Additionally, these traditional methods ignore the complex spatial and temporal communication characteristics, resulting in subpar network performance. In order to overcome these difficulties, we suggest a brand-new intelligent sleep scheduling approach that makes use of both fundamental green computing ideas and incremental deep learning. In order to determine the best sleep schedules for tiny cells, our model takes into account both temporal and spatial communication factors, such as end-to-end delay, packet delivery ratio, throughput, and energy consumption of the nodes. The model adjusts the small cells’ sleep scheduling based on the ongoing spatial-temporal fluctuations in the network, providing the best possible use of resources. Additionally, its focus on green computing helps to lower energy consumption during extensive connections, making it an environmentally beneficial alternative for different scenarios. Under enormous MIMO situations, the results from our model show a significant improvement over the current approaches. These improvements include a notable improvement in communication consistency, a notable increase in throughput, a noticeable increase in packet delivery ratio, and a noticeable decrease in energy consumption and delay. In summary, our concept is a major step forward in the continuous development of intelligent, resilient, and effective 5G MIMO networks.