Peripheral Behavior Classifier Model Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current solutions for mobile computing devices are inadequate in efficiently identifying and addressing the complex factors that contribute to performance degradation and power utilization issues, as they often require computationally-intensive processes, are limited to detecting known viruses, and do not account for interactions between the device and peripheral devices.
Innovation Solution
A mobile computing device is configured to generate and utilize a classifier model that includes features and behaviors related to connected peripheral devices, allowing for real-time monitoring and correction of undesirable behaviors, such as malicious or performance-degrading activities, by locally generating lean classifier models from full models received from a server or by observing behaviors over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If computationally-intensive processes are used to identify and correct performance degradation, then detection accuracy improves, but device power consumption increases and responsiveness decreases
Solution Approach 1:
The patent segments the classification model into two parts: a full model maintained on a server with high computational power, and a condensed version deployed on the mobile device. This segmentation allows the device to run lightweight inference while the server handles complex model updates and retraining, resolving the contradiction between detection accuracy and power consumption.
Solution Approach 2:
The patent introduces a condensation technique as an intermediary process that transforms the heavy full classification model into a lightweight version suitable for mobile devices. This intermediary transformation enables accurate performance degradation detection on resource-constrained devices without requiring full computational intensity, thus reducing power consumption while maintaining detection capability.
2Reliability
If comprehensive behavior monitoring is implemented to detect all performance-degrading behaviors, then detection coverage improves, but device complexity increases
Solution Approach 1:
The patent extracts and monitors only the specific features and behaviors most relevant to performance degradation detection, rather than implementing comprehensive monitoring of all device activities. The classification model is designed to focus on key indicators such as power consumption patterns, CPU usage trends, and thermal characteristics, simplifying the system while maintaining reliable detection coverage.
Solution Approach 2:
The patent applies local quality by tailoring the monitoring approach to specific device conditions and contexts. The classification model adapts to individual device characteristics and monitors behaviors locally relevant to each device's performance profile, rather than applying a uniform comprehensive monitoring scheme, thus reducing overall system complexity while maintaining detection reliability.
3Productivity
If existing classification models are used without customization, then implementation speed improves, but detection accuracy for peripheral device behaviors decreases
Solution Approach 1:
The patent performs preliminary action by pre-training the full classification model on comprehensive datasets that include various peripheral device behaviors and performance degradation patterns. This pre-trained model is then condensed and deployed to mobile devices, enabling fast implementation while maintaining high detection accuracy for peripheral-specific behaviors through the beforehand preparation of domain-specific knowledge.
Solution Approach 2:
The patent applies parameter changes by adapting the classification model parameters and features to specifically account for peripheral device interactions. The model is customized to recognize patterns unique to peripheral devices such as Bluetooth connections, USB attachments, and wireless accessories, improving detection accuracy for these specific behaviors while maintaining implementation efficiency through the use of pre-established model frameworks.
Data Source
AI summary
The various aspects configure a mobile computing device to efficiently identify, classify, model, prevent, and/or correct the conditions and/or behaviors occurring on the mobile computing device that are related to one or more peripheral devices connected to the mobile computing device and that often degrade the performance and/or power utilization levels of the mobile computing device over time. In the various aspects, the mobile computing device may obtain a classifier model that includes, tests, and/or evaluates various conditions, features, behaviors and corrective actions on the mobile computing device that are related to one or more peripheral devices connected to the mobile computing device. The mobile computing device may utilize the classifier model to quickly identify and correct undesirable behaviors occurring on the mobile computing device that are related to the one or more connected peripheral devices.