Institutional-Repository, University of Moratuwa
Welcome to the University of Moratuwa Digital Repository, which houses postgraduate theses and dissertations, research articles presented at conferences by faculties and departments, university-published journal articles and research publications authored by academic staff. This online repository stores, preserves and distributes the University's scholarly work. This service allows University members to share their research with a larger audience.
 Research Publications |  Thesis & Dissertation |  E- Books |
 UoM Journal Publications |  UoM Conference Proceedings |  Articles published in Scimago's Q1 journals |  UoM Research Reports |  Other Articles authored by UoM staff |
Recent Submissions
item: Thesis-Abstract
Valuation approach for 5G spectrum in Sri Lanka
(2023) Jayasinghe, SL; Sumanasena , A; Hemachandra, K
The radio spectrum is a scarce and vital resource. Every country in the world has sovereign control over its spectrum. Demand for the spectrum in terms of mobile communication-related applications is increasing continuously. As a result, demand for additional spectrum acquisition is also increased. 5G technology, as the latest generation of mobile telephony, having many advanced features have opened many doors to other industries such as virtual reality, the Internet of Things(IoT), smart cities, smart health care, etc., and demand for them is trending. As a result of its technological advancement of 5G technology, network operators are keen enough to migrate to 5G considering its economic benefit. Due to the excessive demand for the spectrum, the true value of the spectrum has gone up drastically. Therefore, the understanding of the true value of the spectrum has become a deem requirement to acquire them at the right price. The main objective of this research is to identify a suitable method to estimate the value of the 5G spectrum of Sri Lanka. A literature survey compares techniques used for estimating the value of the spectrum. More specifically, this research tests the suitability of multivariate regression analysis to estimate the value of 5G spectrum. This study estimates the value of 5G spectrum in Sri Lanka using multivariate regression analysis and compares its accuracy with the historical auction prices.
item: Thesis-Full-text
Compressed sensing-based receivers for spatial modulation-MIMO NOMA systems
(2023) Thuvarakan, M; Wavegedara, C
Spatial modulation (SM) offers a promising solution to the demands of up-and-coming wireless networks. In this thesis, the Generalized Orthogonal Matching Pursuit (gOMP) algorithm is introduced into the receiver of NOMA aided SM-MIMO system to improve processing speed and reduce computational complexity. Furthermore, we present comprehensive analysis of the latest research contributions for NOMA based SM-MIMO system which has considerable attention recently. In addition, we focused our attention on the CS based receivers for NOMA based SM and compared the complexity which contributes performance degradation, with typical ML detectors. Detection of SM can be thought of as a process of sparse reconstruction, due to its an inherent property of sparsity. The sparse signal can be detected by Compressed sensing (CS) algorithm which became competitive alternatives. Some modified CS detection algorithms which showed improved performance compared to the conventional CS based once, have not been adopted in NOMA based SM-MIMO system. We formulate the recovery problem by exploiting the SM-MIMO communication system as downlink power domain NOMA where the base station supports more users in downlink scenario. Also, we deployed a low complexity pairing in this system to increase sum throughput. In addition, we analyzed the behavior of gOMP algorithm at this system. Simulations show that the gOMP detector outperforms conventional CS detection schemes
item: Thesis-Abstract
Estimation of intra-site inter-sector coverage overlapping using user data for LTE network
(2023) Sondarangalla, MB; Hemachandra, K
The increase in user demand has created challenges for network capacity implementation. Due to the limited available LTE spectrum, capacity upgrades can take different directions such as adding new sites, new sectors, or using multi-beam antennas for cell splitting instead of adding new frequencies. However, the introduction of higher order sectorization has led to coverage overlaps and increased interference towards adjacent sectors, which limits the performance of Long term Evolution (LTE) networks and degrades spectral efficiency. Although resource scheduling methods have been introduced to cancel inter-cell interference, they limit resource utilization and impact network capacity. Therefore, network operators need to minimize coverage overlap to an acceptable level to mitigate the negative impact on spectral efficiency. Physical optimization has played a key role in achieving this, but traditional methods of identifying inter-cell overlapping have been inefficient due to the lack of geographical view and the time taken to locate the actual issue. The introduction of minimization of drive test (MDT) has moved optimization activities to a data-driven perspective. This thesis proposes a method of estimating intra-site inter-sector overlapping using MDT data and introduces three metrics to identify the granularity of the overlapping and its impact on Signal to interference and noise ratio (SINR) . These metrics can be used to prioritize the worst-affected sites and required azimuth changes to avoid overlapping. The proposed approach can overcome limitations such as false detection and time taken for identification in current methodologies. It accurately identifies issues with over 95% confidence, where spectrum efficiency can be improved, resulting in savings on network operating expenses.
item: Thesis-Full-text
Sim Box detection using machine learning techniques
(2022) Dissanayake, WPTD; Rodrigo, R
Subscriber Identity Module (SIM) box fraud causes severe International Direct Dialing (IDD) revenue losses to network operators and also causes harm to their brand names due to its low Quality of Service (QoS). Therefore, it is very important to prevent this fraud by disconnecting SIM numbers used for SIM box fraud with smaller call attempts to hinder fraudsters from generating revenue. However, detection of SIM box fraud numbers for disconnection consumes lots of work hours as it is necessary to minimize the impact on genuine subscribers. Call Detail Record (CDR) analysis is one of the methods to detect SIM numbers used in the SIM box. Machine learning techniques for fast data analysis are suitable for detecting SIM box numbers using CDR. This research was carried out to accurately identify suitable machine learning methods to detect SIM box fraud numbers. Subscriber profiles are created using CDR to train and evaluate machine learning models. Support vector classifier, decision tree classifier, and K-nearest neighbors classifier algorithms show better results from different types of machine learning models that were trained and evaluated to identify suitable algorithms to detect accurately. A weighted linear combination of these algorithms was used to get the best result with a macro F1-score value of 0.89 using the ensemble voting classifier algorithm. The result confirms that machine learning models can be used to detect SIM box fraud numbers accurately
item: Thesis-Full-text
Policy framework to streamline illegal international voice call termination
(2021) Weerasinghe, WHMS; Samarasinghe, ATLK
Telecommunication service providers around the world lose billions of dollars annually due to various fraudulent activities. Illegal call termination, which is also know and SIMBox fraud and extreme IDD usage related to International Revenue Share Fraud (ISRF) are identified as two of most challenging telecommunication frauds prevalent in the world, which causes a considerable percentage of revenue loss. These fraudulent activities not only deteriorate the operator revenue but also it degrades the customer experience. This is considered as a national issue, due to which the country loses revenue as tax revenue loss and faces security concerns as these calls bypass lawful intercept systems. Having no common platform to address this issue is a problem prevailing in many countries. Therefore, as a solution for this, a common framework is proposed, where all the operators in a country can inter-work, with the proper intervention of regulatory body of the country. The relevant work was reviewed with a literature survey and the gaps were identified. The drawbacks in the current process were identified via a census carried out among the mobile operators in the country and with a focus group discussion with the regulatory body. The different aspects for the framework were outlined by analyzing the data gathered and a centralized system for IDD fraudulent number tracking was designed where a proper coordination within the operators and the regulatory body can be achieved. ISRF related suspected number sharing mechanism was introduced using the common platform. A common framework was established as the output, where identified issues can be minimized. The common framework consists of three major aspects. Establishing a centralized system for IFDD fraudulent number tracking, adding reforms to the existing regulations, and increasing user awareness are the three aspects covered with the proposed common framework. With the proposed centralized system and it is possible to achieve higher revenue and lower fraudulent activities within a country with an increased IDD call quality.