Estimating Control Attributes from Encrypted Communication Traffic

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing communication control systems face challenges in accurately estimating control attributes from communication traffic due to limitations in feature analysis, particularly with diversified traffic patterns and encryption, which hinders precise communication control and user satisfaction.

Innovation Solution

An estimating apparatus and system that calculates and estimates multiple types of control attributes from communication traffic using feature quantity calculating units and estimating models, improving accuracy by learning parameters from correct answers and applying them to communication control.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If communication traffic is multiplexed and encrypted to improve communication speed and security, then communication efficiency and security are enhanced, but the ability to accurately estimate control attributes from observable traffic deteriorates

Engineering Contradiction:
Improvecommunication securityVSAvoidcontrol attribute estimation accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent introduces feature quantity calculating units as intermediaries that extract meaningful characteristics from encrypted and multiplexed communication traffic. These units calculate various feature quantities (packet size, interval, frequency, etc.) that serve as mediators between the encrypted traffic and the control attribute estimation, enabling accurate estimation without decrypting the traffic.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent transforms the estimation problem by changing parameters from direct traffic analysis to feature quantity-based analysis. Multiple types of feature quantities are calculated and combined to form composite features, which are then used for estimation. This parameter transformation enables accurate control attribute estimation while preserving traffic encryption and multiplexing benefits.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If multiple types of control attributes are estimated simultaneously to improve communication control precision, then control accuracy is enhanced, but system complexity increases

Engineering Contradiction:
Improvecontrol attribute estimation accuracyVSAvoidestimation system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the estimation system into multiple specialized feature quantity calculating units, each dedicated to calculating specific types of feature quantities (packet size, interval, frequency, etc.). This segmentation allows the system to handle multiple control attributes simultaneously while maintaining modularity and managing complexity through functional decomposition.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a universal estimation framework where multiple feature quantity calculating units work together with a single estimation unit. The system is designed to estimate any combination of control attributes by selectively activating the appropriate feature quantity calculators, providing multi-functionality without proportionally increasing overall system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If feature quantities are calculated for multiple control attribute types to improve estimation accuracy, then estimation precision is enhanced, but calculation complexity and processing time increase

Engineering Contradiction:
Improvecontrol attribute estimation accuracyVSAvoidfeature quantity calculation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent implements partial action by allowing the system to calculate only the feature quantities necessary for the specific control attributes being estimated. The feature quantity calculating units can be selectively activated based on the estimation requirements, avoiding unnecessary calculations and reducing processing time while maintaining accuracy for the needed attributes.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent performs preliminary calculation of feature quantities from communication traffic before the actual control attribute estimation. By pre-calculating and storing feature quantities (packet sizes, intervals, frequencies, etc.), the system prepares the data in advance, reducing the time required during the actual estimation process and enabling faster response to changing traffic conditions.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11558769B2Estimating apparatus, system, method, and computer-readable medium, and learning apparatus, method, and computer-readable medium
Publication Date: 2023.01.17 NEC CORP
  • US11558769B2 patent drawing
  • US11558769B2 patent drawing
  • US11558769B2 patent drawing

AI summary

An estimating apparatus (1) according to the present invention includes a plurality of feature quantity calculating units (111 to 11n) configured to implement processing of calculating, from a communication traffic, a differing feature quantity group for each of a plurality of types of control attributes relating to communication control, the plurality of feature quantity calculating units being each associated with each of the plurality of types of control attributes, a calculating unit (12) configured to calculate, with regard to each of two or more of the plurality of types of control attributes, each of the feature quantity groups from the communication traffic by use of the feature quantity calculating unit being associated with each of the control attributes, and an estimating unit (13) configured to estimate an attribute value of each of the two or more control attributes from each of the calculated feature quantity groups.