Nwokolo, Innocent Ozulonye and Covenant University, Theses (2023) DEVELOPMENT OF A MACHINE LEARNING BASED FAULT DETECTION MODEL FOR RECEIVED SIGNAL LEVEL IN TELECOMMUNICATION ENTERPRISE INFRASTRUCTURE. Masters thesis, Covenant University Ota.
![[img]](http://eprints.covenantuniversity.edu.ng/style/images/fileicons/application_pdf.png) |
PDF
Download (149kB) |
Abstract
Wireless telecommunication infrastructure can fail without notice for maintenance action. However, these failures may not always result in a complete downtime, but rather degraded performance that can be difficult to pinpoint without specialized tools. Received signal level (RSL) is an important metrics of the quality of a wireless connection and can be used to determine the strength of the signal from the transmitting to the receiving device. It is important to regularly monitor the RSL to ensure that the network is operating at peak performance and to proactively address any issues that may arise. This research aims to develop a machine learning based fault detection model for received signal level in telecommunication enterprise infrastructure. The research methodology involves the modeling of an enterprise point to multipoint wireless communication network using pathloss 5.0 software. Data was extracted from the vector images of the simulated wireless network containing the free space pathloss, transmit power output, transmit antenna gain, transmitter loss, miscellaneous loss and receiver loss. The dataset obtained was then used to train the gradient boosting regression (GBR), random forest regression (RFR) and K-Nearest Neighbor (KNN) regression model. The algorithm compares a threshold value with the received signal levels (RSL) of new and unseen dataset and then trigger a “Fault” or “No-fault” condition. A Fault condition signifies a deviation in the received signal level which then prompt the field support team to perform maintenance on the wireless link. Whereas a No-Fault condition signifies that the RSL is within the accepted range, hence no maintenance is required on the wireless link. In order to choose an optimal machine learning model to achieve the objectives of this project, the performance evaluation metrics of mean absolute error (MAE), mean square error (MSE), R-squared and Root mean square error (RMSE) were compared and experimental results shows that the RFR model is better than the GBR and KNN with 'MAE': 0.007101, 'MSE': 0.000610, 'R-squared': 0.999992, and 'RMSE': 0.024697. By leveraging on the developed machine learning-based fault detection models, service providers can quickly optimize network performance, reduce downtime and increase customer satisfaction.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Machine learning, Enterprise Wireless, Telecommunication, Received Signal Levels (RSL |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering |
| Divisions: | Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science Faculty of Engineering, Science and Mathematics > School of Engineering Sciences |
| Depositing User: | nwokealisi |
| Date Deposited: | 08 May 2023 21:44 |
| Last Modified: | 08 May 2023 21:44 |
| URI: | http://eprints.covenantuniversity.edu.ng/id/eprint/16870 |
Actions (login required)
 |
View Item |
