1.1 Introduction
Global System for Mobile Communication (GSM) is defined as a digital mobile telecommunication technology designed to facilitate secure voice calls, messaging, and data transmission across cellular networks (Oladimeji, 2019). Over the years, GSM has evolved into the dominant communication standard used globally due to its reliability, interoperability, and wide coverage, especially in developing regions where mobile connectivity plays a critical socio-economic role. As mobile subscription continues to increase, telecom operators generate massive volumes of subscriber data, which require advanced mechanisms to ensure security, authenticity, and efficient service delivery (Akinyemi & Yusuf, 2021). However, the growing expansion of GSM services has also led to a corresponding rise in subscription fraud, a fraudulent activity in which individuals obtain mobile services using false identities, invalid credentials, or deceptive information with the intention of evading payment or conducting unlawful activities (Eze, 2020).
As a prelude to other parts of this study, this chapter will discuss the background upon which this study was initiated, the statement of problems that led to this study, the Aim and Objectives of the study. Others are Significance of the study, Scope of work, Limitation of the study and Definition of technical terms.
1.2 Background of Study
Global System for Mobile Communication (GSM) has become one of the most widely adopted mobile communication technologies across the world, offering voice, data, and messaging services to billions of subscribers. According to Adebisi (2019), GSM technology revolutionized the telecommunication landscape by providing a standardized, digital, and secure communication platform that improved connectivity and enabled widespread access to mobile services. As mobile penetration continues to grow globally, particularly in developing nations, the volume of subscriber-related transactions and activities has also expanded significantly.
However, the increasing adoption of GSM services has been accompanied by a rise in subscription fraud, which poses a major challenge to mobile network operators. Eze and Udo (2021) reported that subscription fraud occurs when individuals obtain mobile services using false identities, misleading credentials, or deceptive intentions, leading to significant revenue leakage and operational inefficiencies. Olabode (2020) asserted that traditional fraud detection systems, which rely heavily on predefined rules and manual checks, are no longer sufficient in addressing the dynamic and sophisticated nature of modern fraudulent activities.
The complexity of GSM subscription data demands more advanced analytical methods to detect irregular patterns. Ibrahim and Markus (2022) stated that fraudsters continually evolve their techniques, making it difficult for conventional systems to accurately distinguish between legitimate and suspicious subscription behaviors. On the other hand, researchers have shown increased interest in the application of intelligent computational models capable of learning from large datasets and improving fraud detection accuracy.
Olatunji (2021) affirmed that Artificial Neural Networks (ANNs) offer superior capabilities in pattern recognition, classification, and anomaly detection, making them suitable for handling complex fraud indicators in GSM networks. Several studies have highlighted the limitations of existing fraud detection systems and the growing need for more adaptive and intelligent solutions. Adeyemo and Nwafor (2023) contended that ANN-based approaches can provide more robust, scalable, and efficient fraud detection frameworks by analyzing non-linear data relationships and predicting emerging fraud trends in real time. Similarly, Chukwu (2022) reported that machine learning techniques such as ANNs enhance system responsiveness, reduce false positives, and strengthen overall GSM security operations.
The challenges encountered that led to the execution of the research work is that, the rapid expansion of Global System for Mobile Communication (GSM) networks has brought significant improvements in connectivity, communication, and digital services. However, this growth has also created an avenue for subscription-related fraud, which is increasingly threatening the integrity and financial sustainability of mobile network operators. Existing fraud detection methods are predominantly rule-based, and several scholars have reported that such manual and semi-automated systems is limited in detecting sophisticated and evolving fraud patterns (Adebayo, 2021). This study is set against the backdrop of advancing intelligent models to address the persistent challenge of subscription fraud in GSM networks.
1.3 Statement of Problem
Based on the investigation conducted, the implemented system encounters a number of challenges, with some of the most significant issues outlined below:
- The current GSM subscription verification processes rely extensively on manual checks, causing delays that give fraudsters opportunities to exploit weak points in registration procedures.
- The present fraud-monitoring framework lacks real-time analytical intelligence, making it difficult to promptly identify suspicious subscription behavior.
- Traditional rule-based detection methods in use are limited in flexibility and are easily bypassed by sophisticated or evolving fraud patterns.
- The existing operational workflows do not effectively utilize historical behavioral datasets, which reduces accuracy in identifying abnormal subscriber activities.
- The traditional infrastructure struggles with scalability, leading to reduced performance as subscriber numbers continue to increase across GSM networks.
- Accuracy levels remain low because reliance on basic threshold measures often results in excessive false positives and undetected fraudulent accounts.
1.4 Aim and Objectives of the Study
The aim of this study is to design and implement an ANN-based system that will detect and prevent subscription fraud in GSM networks. In achieving this aim, the following specific objectives were laid out as follows:
- To create a comprehensive dataset framework for GSM subscriber activities and potential fraud scenarios.
- To develop an ANN model capable of analyzing subscription patterns and detecting fraudulent behavior accurately.
- To design a user-friendly interface for telecom operators to monitor subscription activities and detect anomalies.
- To implement the ANN-based fraud detection system within a simulated or real GSM environment.
- To create strategies that reduce false positives and enhance the system's predictive accuracy.
1.5 Significance of Study
The deployment of the proposed system will hold significant relevance in the following ways:
- The study will provide significant benefits to mobile network operators by enhancing the accuracy and efficiency of fraud detection.
- This system will improve customer trust and satisfaction by minimizing service disruptions and false positives that occur in conventional fraud detection methods.
- It will also serve as a tool for regulatory bodies to ensure compliance with industry standards and to strengthen security protocols within the telecom sector.
- from an academic perspective, the study will contribute to the existing body of knowledge on the application of machine learning techniques in telecommunications security, providing a framework for future research on intelligent fraud detection.
1.6 Scope of Study
This study focuses on GSM subscription fraud detection within the operations of MTN Nigeria in Lagos State. It will analyze subscriber data, simulate fraudulent activities, and test the implementation of an ANN-based detection system.
The study is limited to subscription-related fraud and does not cover other forms of mobile network fraud, such as SIM cloning or identity theft outside the GSM subscription process.
1.7 Limitations of the Study
During the course of this study, many things militated against its completion, some of which are:
- Time Constraint: The time frame given to accomplish this project was very short due to school academic calendar and it was carried out under pressure which made the researcher not to implement some necessary features.
- Establishment Policies: Establishment policies posed a serious limitation as most staffs are not ready to release information needed for this research work. There were lots of information needed from the staffs of this institution to enhance the study which took them time to release or they did not release at all for security purposes, hence the scope was reduced.
- Financial Constraint: Insufficient fund tends to impede the efficiency of the researcher in sourcing for the relevant materials, literature or information and in the process of data collection (internet).
1.8 Definition of Terms
Global System for Mobile Communication (GSM): A digital cellular network standard for mobile communications that supports voice, messaging, and data services (Oladimeji, 2019).
Subscription Fraud: The act of obtaining mobile network services through deceptive information or false identities with the intention of avoiding payment or committing illegal activities (Eze, 2020).
Artificial Neural Network (ANN): A computational model inspired by the structure of the human brain that is capable of learning from data, recognizing patterns, and making predictions (Uzor & Ibrahim, 2023).
False Positives: Legitimate subscription activities that are incorrectly flagged as fraudulent by a detection system (Olatunji, 2021).
Dataset: A structured collection of subscriber information and activity logs used to train and test the ANN model for fraud detection (Adeyemo & Nwafor, 2023).
Pattern Recognition: The ability of the ANN to identify normal and anomalous behavior in GSM subscription activities (Olawale, 2022).
…
