Tools
Tools
Jamil, M, Hassan, B, Ahmed, SS, Maheshwari, M and Sahu, BJR (2022) Deep Learning-Based Cell Outage Detection in Next Generation Networks. In: Intelligent and Cloud Computing Proceedings of ICICC 2021 . pp. 491-500. (International Conference in Innovative Computing and Communication 2021, 20th Feb - 21st Feb 2021, Virtual).
|
Text
Deep Learning-Based Cell Outage Detection in Next Generation Networks.pdf - Accepted Version Download (386kB) | Preview |
Abstract
5G and beyond wireless networks will support high data rate, seem-less connectivity and a massive number of users as compared to 4G network. It is also expected that the end-to-end latency in transferring data will also reduce significantly, i.e., 5G will support ultra-low latency services. To provide the users with all these advantages, 5G utilizes the Ultra-Dense Networks (UDN) technique. UDN helps manage the explosive traffic data of users as multiple small cells are deployed in both indoor and outdoor areas, for seamless coverage. However, outage is difficult to detect in these small cells as these small cells have high density of users. To overcome this hindrance, Cell Outage Detection (COD) technique is utilized which aims to detect outage autonomously. This reduces maintenance cost and outages can be detected beforehand. In this paper, Long Short Term Memory (LSTM) is used for outage detection. The LSTM network is trained and tested on subscriber activities values which include SMS, Call and Internet activity. Our proposed LSTM model has classification accuracy of 85% and a FPR of 15.7303%.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | Computer Science & Mathematics |
| Publisher: | Springer Nature |
| Related URLs: | |
| SWORD Depositor: | A Symplectic |
| Date Deposited: | 24 Sep 2024 14:14 |
| Last Modified: | 24 Sep 2024 14:14 |
| DOI or ID number: | 10.1007/978-981-16-9873-6_45 |
| URI: | https://researchonline.ljmu.ac.uk/id/eprint/24203 |
 |
View Item |