TR201517613A2 - A SYSTEM FOR DETERMINATION OF FRAUD IN TELECOMMUNICATION SYSTEMS - Google Patents

Abstract

The present invention relates to a fraud detection system (1) that uses network performance indicators (KPI_Key Performance Indicators) as the primary input for mobile communications operator firms.

Description

DESCRIPTION FRAUD IN TELECOMMUNICATION SYSTEMS A SYSTEM TO DETERMINE Technical Area This invention has been used as a basic input for mobile communication operator companies. a spoof that uses performance tokens (KPI_Key Performance [ndicators)] (fraud) is related to the determination system. Prior Art Fraud in telecommunications systems It goes back to the date of its first creation. One of the first fraud methods used one of the subscriber lines with parallel connections made physically to the subscriber lines. is to search for expensive destinations. Today, the same process is used in GSM infrastructures. It is done by cloning the SIM card.

Today, telecommunications fraud is one of the biggest problems in the telecommunications industry. is one of its biggest problems. The emergence of new generation communication technologies With the emergence of new types of fraudulent earning methods, a great increase has occurred. has arrived. Communication Fraud Control AssociationýCFCO) telecommunications fraud that occurred in 2009 Between $70 and $78 billion in damage caused by their ventures predicts it will happen. Competitive pressures in the telecommunications market telecommunications companies of revenue losses due to fraud It directs them to do more effective work on reducing it.

Telecommunication fraud only harms telecommunication operators. cause damage to the end users as well as is happening. Firms that are telecommunications operators protect their reputations in order to prevent unauthorized use of the services offered to its subscribers. are taking precautions.

Today, telecommunications operator companies Various fraud control systems are used to prevent fraudulent systems. They use systems (Fraud Management System). The systems in question Detecting fraud attempts by analyzing subscribers' usage reports they ensure. Fraud control systems are typical data mining applications. Fraud control systems retrospective a kind of finger that determines the usage trends of each customer from usage data They enable the creation of a trace database. However, nowadays any fraud detection system fraud detection in telecommunications system does not use anomalies in network performance indicators. system from a computer-assisted fraud detection method. is mentioned. Within the system in question, the user's use of the network The fraudulent use case is detected based on some of the relevant information.

This test system measures the user's contact times, how often they communicate, and location information that he communicated with etc. by comparison using information Allows identifying fraudulent use.

Brief Description of the Invention The aim of this invention is to use mobile communication operators as a basic input for companies. a fraud detection system using network performance tokens is to perform.

Detailed Description of the Invention To achieve the aim of this invention, the “Telecommunication A System for Detecting Fraud in Their Systems” is shown as appendix. and this figure; Figure 1 - is the schematic view of the subject system of the invention.

The pieces seen in the figures are numbered one by one, and the corresponding numbers are given below. 1. System 2 database 3. Anomaly detection module . central processing module Network performance as a basic input for mobile communication operator companies. mobile communication using tokens (KPI_Key Performance Indicators) subject of the invention enabling the detection of frauds in the network (MCN) fraud detection system (1); customer-based network performance in the mobile communication network (MCN) at least one database (2) in which tokens are recorded, communication with the database (2) and recorded in the database (2). the network performance in question by processing network performance tokens. at least one anomaly detection module (3) that detects abnormalities in markers communication with the database (2) and recorded in the database (2). predetermined rules on network performance indicators. possible frauds in said network performance tokens by running at least one fraud detection module (4), and Communication with the anomaly detection module (3) and fraud detection module (4) network detected by the anomaly detection module (3) anomaly detection information about abnormalities in performance markers fraudulent arising from the said anomalies by obtaining from the module (3) network detected by the fraud detection module (4) information about possible fraud in performance tokens. from the identification module (4), resulting from such possible frauds to the detected fraud, which enables the determination of fraudulent predetermined necessary actions to prevent fraud It contains at least one central processing module (5) that enables

The database (2), anomaly detection module (3), and communicates with the fraud detection module (4). In database (2) customer-based network performance in the mobile communication network (MCN) tokens are kept under record.

Anomaly detection module (3), database (2) and it is in communication with the central processing module (5). Detect anomaly module (3) data by processing the network performance indicators recorded in the base (2) It detects anomalies in network performance indicators.

Fraud detection module (4) database (2) in the inventive system (1) and in communication with the central processing module (5). Fraud detection module (4) on network performance indicators recorded in the database (2) the performance of the network in question by running the predetermined rules. detects possible frauds in tokens› The central processing module (5) in the system (1), which is the subject of the invention, determines the anomaly. module (3) and fraud detection inodule (4). Central network detected by the processing module (5) anomaly detection module (3) anomaly detection information about abnormalities in performance markers It is fraudulent arising from the aforementioned anomalies by taking it from the module (3). network detected by the fraud detection module (4) information about possible fraud in performance tokens. from the identification module (4), resulting from such possible frauds It allows to determine whether there is fraud or not. central processing module (5) furthermore, to prevent fraud related to detected fraud, more ensures that predetermined necessary actions are taken. find one central processing module (5) later in similar situations anomaly so that fraudulent frauds can be easily detected received by the detection module (3) and considered fraudulent. records the information about the cases in the fraud detection module (4). takes.

In an exemplary application of the subject system (1), it is recorded in the database (2). network performance indicators received, such as smart phone, mobile phone, electronic lMSl, which is the number of connections made by the devices in the mobile communication network. (International Mobile Subscriber Identity_U1international Mobile Subscriber Identity) the number of connections and the total number of mobile data users have downloaded and installed. The value of 2G/3G data volume can be given. Anomaly in the sample application in question the number of IMSI connections registered in the detection module (3) database (2) higher than the predetermined values or again 2G/3G data in case the volume is considerably higher than the predetermined values The network in question detects anomaly in performance indicators and sends the data regarding the anomaly to the central processing module (5).

Service from a mobile communication network (MCN) within the database (2) of the invention daily voice call made by an electronic device such as a smart phone, mobile phone, etc. as a network performance indicator and the number of voice calls normally An example of a daily recording between 10-50 value ranges. application in the mobile communication network by the anomaly detection module (3). (MNC) is an electronic device that makes between 3000-5000 voice calls per day. is detected, this situation is detected as an anomaly. Identify the anomaly module (3) to the central processing module (5) about the detected anomaly. is sending. The data coming to the central processing module (5) is preferably fraudulent. evaluated manually by experts and this evaluation If fraud is detected as a result of the situation in question, the In the case fraud module (4), the same fraud can be easily detected later. be recorded so that it can be recorded.

The subject of the invention is “A Method for Detection of Fraud in Telecommunication Systems. It is possible to develop a wide variety of applications for the system (1). not limited to the examples described herein, but essentially as specified in the claims.

Claims (7)

A SYSTEM FOR DETECTION OF FRAUD IN REQUESTS 5 This invention relates to a fraud detection system (1) that uses network performance indicators (KPI_Key Performance Indicators) as basic inputs for mobile communication operator companies. REQUESTS 1. It enables the detection of frauds in the mobile communication network (MCN) by using network performance indicators as basic input for mobile communication operator companies; At least one database (2) in which customer-based network performance tokens in the mobile communication network (MCN) are recorded, by processing the network performance indicators in communication with the database (2) and recorded in the database (2), the said network performance indicators At least one anomaly detection module (3), which detects anomalies in tokens, runs pre-determined rules on network performance tokens that are in communication with the database (2) and recorded in the database (2), and detects possible frauds in said network performance tokens. By obtaining information from the anomaly detection module (3), the anomaly detection module (3) containing at least one fraud detection module (4) and the anomalies in the network performance indicators detected by the anomaly detection module (3) in communication with the fraud detection module (4) fraud resulting from said anomalies o It enables to determine whether there is fraud caused by fraud detection module (4) by obtaining information about possible frauds in network performance tokens detected by fraud detection module (4), and taking pre-determined actions to prevent fraud regarding detected fraud. a fraud detection characterized by at least one central processing module (5) that provides . A system (1 ). Claim 1 or 2, characterized by the anomaly detection module (3), which communicates with the database (2) and the central processing module (5) and detects the anomalies in the said network performance indicators by processing the network performance indicators recorded in the database (2). A system as in (1). With the fraud detection module (4), which is in communication with the database (2) and the central processing module (5) and detects possible frauds in the said network performance tokens by running pre-determined rules on the network performance tokens recorded in the database (2). a system (1) as in any one of the above claims, characterized. It receives information about the anomalies in the network performance indicators detected by the anomaly detection module (3), which is in communication with the anomaly detection module (3) and the fraud detection module (4), from the anomaly detection module (3), and determines whether there is fraud due to the said anomalies and whether there is fraudulent detection. A system (1) as in any of the above claims, characterized by the central processing module (5), which provides the determination of whether there is fraud due to such possible frauds by obtaining information about possible frauds in the network performance tokens detected by the fraud detection module (4) from the fraud detection module (4). . A system (1) as in Claim 57, characterized by a central processing module (5), which ensures that the necessary predetermined actions are taken in order to prevent fraud related to detected fraud. 7. Any of the above claims, characterized by the central processing module (5), which records the information about the cases considered as fraudulent and received by the anomaly detection module (3) in the fraud detection module (4), in order to easily detect frauds that may be carried out in similar situations later on. a system as in one (1).