Fraud Detection Device Analyzing Call Records for Network Stability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Mobile communication networks face challenges in detecting and preventing fraudulent user activities, such as group calling and mass short message sending, which occupy resources, reduce call completion rates, and cause system instability, leading to negative impacts on network security and service income.

Innovation Solution

A method and system for detecting fraudulent users by analyzing call attempt records and short message records using set fraud identification rules, extracting relevant parameters, and implementing service restriction policies to prevent further usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If fraudulent users send mass short messages and make group calls, then the number of messages and calls increases, but network resources are occupied and system stability deteriorates

Engineering Contradiction:
Improvemessage sending volumeVSAvoidsystem stability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements preliminary detection and identification of fraudulent users by analyzing call attempt records and short message records against fraud identification rules before the fraudulent activities can cause significant system instability. This allows the system to proactively restrict service usage of identified fraudulent users, preventing resource occupation and maintaining system reliability.

Inventive Principle:
Principle #10Preliminary action

2Quantity of substance

If fraudulent users make group calls and send mass messages, then traffic volume increases, but call completion rate decreases

Engineering Contradiction:
Improvetraffic volumeVSAvoidcall completion rate
Core Design Contradiction:
Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent establishes a feedback mechanism where the fraud detection device continuously monitors call attempt records and short message records, compares them against fraud identification rules, and dynamically adjusts service restrictions for fraudulent users. This feedback loop ensures that as fraudulent activities increase traffic volume, the system responds by restricting those users' services, thereby maintaining call completion rates for legitimate users.

Inventive Principle:
Principle #23Feedback

3Loss of time

If the detection period is shortened to improve timeliness, then fraudulent users are identified faster, but the complexity of real-time analysis increases

Engineering Contradiction:
Improvedetection timeVSAvoidanalysis complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The patent segments the fraud detection process into distinct functional modules: a fraud detection device that collects and analyzes records, a fraud identification rule engine that evaluates patterns, and a service restriction mechanism that enforces decisions. This segmentation allows the system to handle real-time analysis complexity through modular processing while maintaining short detection periods for timely identification of fraudulent users.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3214861B1Method, device and system for detecting fraudulent user
Publication Date: 2019.07.31 ZTE CORP
  • EP3214861B1 patent drawingFigure 1~3
  • EP3214861B1 patent drawingFigure 4~5
  • EP3214861B1 patent drawingFigure 6~7

AI summary

A method, device and system for detecting a fraudulent user. The method comprises: analyzing data extracted from call attempt records or short message caller bills of users in a detection period according to a set fraud identification rule, so as to identify a fraudulent user.