Home Appliance Malware Detection Using Stable Power Signatures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing malware detection methods for home appliances are inaccurate due to varying power consumption patterns influenced by external factors and internal states, such as operating conditions and environmental settings, without sufficient consideration of device-specific information.
Innovation Solution
A malware detection method that selects a detection model based on device type and operating state settings, using stable power or current consumption within a predetermined range to accurately identify malware in home appliances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If power consumption is used for malware detection in home appliances, then detection capability is provided, but detection accuracy deteriorates due to varying power consumption patterns from external factors and internal states
Solution Approach 1:
The patent applies local quality by selecting detection models specific to each home appliance type (e.g., air conditioner, refrigerator, washing machine) rather than using a universal detection model. Each appliance type has its own detection model that is trained on appliance-specific power consumption patterns, allowing the system to account for local variations in power consumption caused by appliance-specific operations and states.
Solution Approach 2:
The patent changes the parameter of power consumption analysis by focusing on stable power consumption values within a predetermined range rather than using raw power consumption data. The system identifies periods of stable power consumption and uses these stabilized values for malware detection, transforming the detection approach from analyzing absolute power values to analyzing stability patterns, which reduces the impact of varying operational conditions.
2Measurement precision
If detection models are selected based on device type and operating state, then detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-classifying home appliances into different types (air conditioner, refrigerator, washing machine, etc.) and pre-preparing detection models for each type before actual malware detection occurs. The system stores these pre-classified appliance types and corresponding detection models in advance, so that during operation, the appropriate model can be quickly selected without complex real-time analysis, reducing the computational burden during detection.
Solution Approach 2:
The patent implements dynamics by making the detection model selection adaptive to the current operating state of the home appliance. The system dynamically determines which detection model to use based on the appliance's current state (e.g., whether it is in heating mode, cooling mode, or standby state), allowing the detection accuracy to adapt to varying operational conditions without requiring a completely complex reconfiguration of the detection system.
3Reliability
If stable power consumption within predetermined range is used for detection, then false positive rate is reduced, but detection coverage is limited to specific operational states
Solution Approach 1:
The patent applies segmentation by dividing the operational states of home appliances into different segments (e.g., heating operation, cooling operation, power saving mode, standby state) and preparing separate detection approaches for each segment. The system identifies periods of stable power consumption within each operational segment and uses these segmented stable values for malware detection, allowing reliable detection across multiple operational states while maintaining low false positive rates for each specific state.
Data Source
AI summary
A malware detection method for a home network system including one or more home appliances that are connected to a home network includes: obtaining a plurality of setting values including at least information indicating a device type and an operating state of a target device subject to malware detection; selecting one detection model out of a plurality of detection models according to the plurality of setting values obtained; obtaining power consumption or current consumption of the target device; and detecting whether the target device is infected with malware, based on stable power or stable current obtained in the obtaining of the power consumption or the current consumption using the one detection model selected in the selecting, when the power consumption indicates stable power that varies within a predetermined range or the current consumption indicates stable current that varies within a predetermined range.