Flash Call Risk Detection and Blocking for OTP Network Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Flash calls, used for one-time password delivery, increase costs and network burden for mobile network operators due to high volumes of missed calls and increased network activity, while illegitimate service providers avoid detection mechanisms.
Innovation Solution
A method using machine learning to analyze call data patterns and implement call intervention actions to block flash calls, including delaying setup, diverting to voicemail or announcement services, and monitoring for subsequent SMS messages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If flash calls are allowed for OTP delivery, then authentication service is provided, but network costs and burden increase for MNO
Solution Approach 1:
The system performs preliminary risk assessment and classification of incoming calls before they are fully processed. By evaluating call patterns, device behavior, and historical data in advance, the system identifies potential flash calls early in the call setup process, allowing intervention before significant network resources are consumed.
Solution Approach 2:
The patent introduces an intermediary risk assessment system between the calling party and the target device. This intermediary layer analyzes call characteristics and makes decisions about call handling, acting as a mediator that protects the network from harmful flash calls while allowing legitimate calls to proceed normally.
2Loss of energy
If flash calls are blocked, then network costs are reduced, but legitimate OTP delivery may be affected
Solution Approach 1:
The system applies different handling strategies to different categories of calls based on their risk profiles. Low-risk calls are permitted to proceed normally, medium-risk calls undergo additional verification or are diverted to alternative channels, and high-risk calls are blocked. This localized quality approach ensures that legitimate calls are not adversely affected while still blocking harmful flash calls.
Solution Approach 2:
The system dynamically changes handling parameters based on risk assessment results. Instead of a static block-or-allow decision, the system adjusts call treatment based on multiple parameters including risk category, call pattern matching results, and real-time network conditions, allowing flexible response to different call scenarios.
3Measurement precision
If risk assessment is performed on all calls, then flash calls are identified, but processing time and complexity increase
Solution Approach 1:
The system performs partial risk assessment by focusing on key indicators and call characteristics that are most predictive of flash calls. Rather than analyzing every possible parameter in depth, the system evaluates a targeted set of features including call duration patterns, device response behavior, and historical data, achieving effective detection with reduced processing complexity.
Solution Approach 2:
The risk assessment process is segmented into multiple stages: initial filtering based on basic call characteristics, intermediate risk evaluation for borderline cases, and detailed analysis only for high-risk scenarios. This segmentation allows the system to maintain high detection accuracy while avoiding unnecessary complex processing for low-risk calls.
4Object-affected harmful factors
If call intervention actions are taken, then flash calls are blocked, but call setup time is delayed
Solution Approach 1:
Risk assessment and classification are performed preliminarily during call setup phases before full connection is established. By identifying flash calls early in the signaling process, the system can terminate problematic calls before they consume significant network resources or cause user-perceptible delays, while legitimate calls experience minimal or no delay.
Solution Approach 2:
For low-risk calls that pass initial assessment, the system skips detailed verification steps and allows rapid call establishment. This rushing through of the assessment process for trusted callers minimizes call setup time, while more thorough evaluation is applied only to calls that trigger risk indicators.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is provided a method for blocking a flash call at a target device, the method comprising: determining a risk that a call is a flash call; determining that the risk is in a first risk category; and in response to determining that the risk is in the first risk category: performing one or more call intervention actions, the one or more call intervention actions operable to prevent the flash call from being received by the device.