Edge Device Power Management via Dynamic Neural Switching

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video surveillance systems face challenges in managing power consumption efficiently, particularly in networks with large numbers of cameras, leading to significant energy usage and potential battery life issues in battery-powered devices.

Innovation Solution

Implementing intelligent power management strategies in edge devices, including situational neural network switching, motion detection management, reduced frames-per-second and resolution, smart video streaming, power-saving peripherals, and a power management dashboard, to dynamically adjust operations based on environmental conditions and user input for optimized power usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If video analytics are continuously performed at high processing levels, then detection accuracy and response time are improved, but power consumption increases significantly

Engineering Contradiction:

Improvedetection accuracyVSAvoidpower consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system dynamically switches between multiple neural networks with different processing capabilities based on real-time conditions. A first neural network with higher processing power is used when detection accuracy is critical, while a second neural network with lower processing power is used when power consumption needs to be reduced, allowing the system to adapt its detection precision to current operational requirements

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes operational parameters by switching between different neural network configurations. The first neural network is configured for high-precision detection with more computational resources, while the second neural network is configured for lower power consumption with reduced computational complexity, enabling parameter adjustment to balance accuracy and energy usage

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If multiple neural networks are deployed for different power saving criteria, then power consumption flexibility is improved, but device complexity increases

Engineering Contradiction:

Improvepower consumption flexibilityVSAvoiddevice complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system segments the neural network functionality into multiple specialized networks, each optimized for specific power consumption levels. This segmentation allows the device to select appropriate processing capabilities based on power availability, providing flexibility without requiring a single complex network to handle all scenarios

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Multiple neural networks are deployed to provide universal adaptability across different operating conditions. Each network serves a specific power consumption criterion, and collectively they provide multi-functionality that allows the device to operate effectively whether power is abundant or constrained, making the system universally adaptable to various deployment scenarios

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Use of energy by moving object

If video analytics are placed in sleep mode to save power, then power consumption is reduced, but detection response time increases

Engineering Contradiction:

Improvepower consumptionVSAvoiddetection response time

Core Design Contradiction:

Use of energy by moving objectVSLoss of time

Solution Approach 1:

The system implements periodic action by placing video analytics in sleep mode during low-activity periods and activating them when motion is detected. This periodic activation pattern reduces power consumption during idle times while ensuring that detection capabilities are restored quickly when events occur, balancing energy savings with response time requirements

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system uses feedback from motion detection sensors to control the state of video analytics. When motion is detected, the system activates the video analytics from sleep mode, creating a feedback loop that ensures detection response occurs only when necessary, thereby reducing overall power consumption without significantly impacting response time for actual events

Inventive Principle:

Principle #23Feedback

4Measurement precision

If high resolution and frame rate are used for video analytics, then surveillance quality is improved, but power consumption increases

Engineering Contradiction:

Improvesurveillance qualityVSAvoidpower consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system applies local quality by processing video at different resolutions based on the region of interest. When a specific area is detected as containing potential events, the system increases processing resolution and frame rate for that local region while maintaining lower settings for other areas, thereby improving surveillance quality where needed while reducing overall power consumption

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20250301214A1Intelligent edge power management

Publication Date: 2025.09.25 JOHNSON CONTROLS TYCO IP HLDG LLP

AI summary

Example implementations include a method, apparatus, and computer-readable medium for power management of an edge device using one or more of: “Situational Switching Neural Networks,”“Motion Detection Management of Analytics,”“Reduced Frames-Per-Second (FPS) and Resolution Power Saving,”“Smart Video Streaming,”“Power Saving Peripherals Management,”“Smart Power Saving Camera Lens Defog,” and “Power Management Dashboard.”