Call Center DoS Detection via Confidence Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Call centers face challenges in processing incoming calls efficiently due to denial-of-service attacks, which overwhelm their systems and prevent them from handling legitimate calls effectively.
Innovation Solution
A system that collects call statistics to determine the confidence level of incoming calls being denial-of-service attacks, using feedback from clients to augment statistics, and either routes, tests, or drops calls based on this confidence level, employing passive tests and queries to differentiate between human and machine callers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the call center processes all incoming calls without filtering, then all legitimate calls can be handled, but the system becomes overwhelmed by denial-of-service attacks and cannot process calls effectively
Solution Approach 1:
The system performs preliminary analysis of incoming calls by collecting statistics about call patterns, sources, and characteristics before the calls are routed to agents. This advance preparation allows the system to identify potential denial-of-service attacks early and take preventive actions such as filtering or routing adjustments, ensuring reliable processing of legitimate calls while maintaining efficiency.
2Reliability
If the system implements comprehensive call filtering and analysis, then denial-of-service attacks can be effectively blocked, but the system complexity increases
Solution Approach 1:
The call filtering system is divided into separate functional modules: a statistics collection module that gathers call data, a confidence metric calculation module that analyzes the statistics, and a call routing module that acts on the analysis results. This segmentation allows each module to perform its specific function independently, making the overall system more manageable and less complex while maintaining effective attack resistance.
Solution Approach 2:
The system introduces an intermediary confidence metric calculation layer between the incoming calls and the call routing decision. This intermediary module calculates confidence scores based on collected statistics and uses these scores to determine appropriate routing actions, simplifying the decision-making process and reducing system complexity while maintaining reliable attack filtering.
3Measurement precision
If the system collects and analyzes extensive call statistics, then accurate identification of attacks is achieved, but the time required for analysis increases
Solution Approach 1:
The system collects more call statistics data than strictly necessary for basic filtering, gathering extensive information about call patterns, sources, frequencies, and characteristics. This excessive data collection provides a richer basis for accurate attack detection, allowing the confidence metric calculation to make more informed decisions with higher precision, while the modular architecture ensures this extensive analysis does not excessively delay call processing.
Data Source
AI summary
A system may receive data collected by a network for a plurality of calls that are to be routed by the network to a call center, detect an event based on the received data, determine a confidence level, for the event, that the event is a denial-of-service attack, determine a call service to be provided to the call center in response to detecting the event, and provide the call service based on the confidence level.