System and method for real-time fraud detection within a telecommunications system

• US 6,327,352 B1
• Filed: 02/24/1997
• Issued: 12/04/2001
• Est. Priority Date: 02/24/1997
• Status: Expired due to Term

• Current Assignee: SBC Properties, L.P.
• Original Assignee: Ameritech Corp. (AT&T, Inc.)
• Inventors: McGuire, Michael Joseph, Hatlak, Michael Joseph, Betts, Judy Lynn
• Primary Examiner(s): Matar, Ahmad
• Assistant Examiner(s): TIEU, BENNY QUOC
• Application Number:
  US08/807,039
• Time in Patent Office: 1,744 Days
• Field of Search: 379/189, 379/111, 379/112-1-1, 379/115-1-1, 379/207.04, 379/207.16, 379/219, 379/220.01, 379/20-1-, 379/145, 379/188, 379/35, 379/133, 379/114-1-1, 379/196
• US Class Current: 379/189
• CPC Class Codes: H04L 63/14 for detecting or protecting against malicious traffic
• CPC Class Codes: H04M 15/00 Arrangements for metering, time-control or time indication ; Metering, charging or billing arrangements for voice wireline or wireless communications, e.g. VoIP
• CPC Class Codes: H04M 15/47 Fraud detection or prevention means
• CPC Class Codes: H04M 2215/0148 Fraud detection or prevention means
• CPC Class Codes: H04M 3/2218 Call detail recording
• CPC Class Codes: H04M 3/36 Statistical metering, e.g. recording occasions when traffic exceeds capacity of trunks

Abstract

Real time detection of the fraudulent use of a telecommunications network is accomplished by analyzing data for each call that is occurring within the network. A signal protocol receiver is used to collect signaling protocol for each call that is occurring within the network. The Signaling protocol data is collected, decoded and formatted into call information records (CIRs). The CIRs contain various operator specified parameters for each call that is occurring within the network. The CIRs are compared to operator defined thresholds. If any of the CIRs exceeds the thresholds, an alert is generated. The alerts are stored in a database where the operator can analyze them and take the appropriate corresponding action to resolve the alert. The alerts and the CIRs are archived in a database so that trends of fraudulent use can be detected and prevented. This method of fraud detection provides for the effective analyzation of every call that is occurring within the network. Accordingly, no call go unanalyzed and ideally no fraud goes undetected. Additionally, the method does not impose an additional load on the network switching equipment and therefore results in a better quality of transmissions.

Claims

1. A method for detecting fraudulent access to a telecommunications network comprising a plurality of switches, the method comprising:

(a) providing a signal protocol receiver independent from the switches;

(b) collecting signaling protocol data for a call from the telecommunications network using the signal protocol receiver;

(c) decoding the signaling protocol data;

(d) correlating the signaling protocol data into call information records during the duration of the call; and

(e) analyzing the call information records during the duration of the call to detect fraudulent use of the telecommunications network.

2. The method of claim 1 further comprising classifying the call information records.

3. The method of claim 1 further comprising storing the call information records in a database.

4. The method of claim 1 wherein (e) comprises comparing the call information records to pre-established thresholds.

5. The method of claim 1 wherein (e) comprises comparing the call information records to one another.

6. The method of claim 4 wherein (e) comprises generating an alert when at least one of the call information records exceeds the thresholds.

7. The method of claim 6 wherein the alert is in the form of audio, visual or remote.

8. The method of claim 6 wherein (e) comprises storing the alerts and call information records in a database and maintaining data for a predetermined number of days, to facilitate an analysis of fraudulent trends.

9. The method of claim 1 wherein the signaling protocol data is in-band and is collected at the plurality of switches within the network.

10. The method of claim 1 wherein the signaling protocol data is out-of-band and is collected from at least one signal transfer point within the network.

11. The method of claim 1 wherein the call information records contain at least one of the following parameters:

originating, terminating, billing type, using duration, aggregate duration, call volume.

12. A system for detecting fraudulent access to a telecommunications network comprising a plurality of switches, the system comprising:

a signal protocol receiver, independent from the switches, for collecting signaling protocol data;

a decoder for decoding the signaling protocol data and formulating call information records during the duration of the call; and

a processor for analyzing the call information records during the duration of the call in order to detect fraudulent use of the telecommunications network.

13. The system of claim 12 wherein the processor comprises a pre-processor for classifying the call information records and a watch point processor for comparing the call information records to operator defined thresholds.

14. The system of claim 12 wherein the processor further comprises a fraud analysis processor for storing the alerts and the call information records in a database.

15. A method for detecting fraudulent access to a telecommunications network comprising a plurality of switches, the method comprising:

(a) providing a signal protocol receiver independent from the switches;

(b) collecting signaling protocol data from the network using the signal protocol receiver;

(c) decoding the signaling protocol data;

(d) correlating the signaling protocol data into call information records containing at least one call parameter during the duration of the call;

(e) classifying the call information records based upon the at least one call parameter;

(f) analyzing the call information records by comparing them to pre-established thresholds or other call information records during the duration of the call;

(g) generating an alert if the call information records exceed the thresholds;

(h) storing the alerts in a database; and

(i) storing the call information recording in a database.