AI Wake-Up Chip for Low-Power Sensing Function Activation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices, such as smartphones, consume excessive power due to unused hardware components remaining active, and existing solutions fail to precisely recognize sensing data, leading to inefficient power management and inaccurate function performance.
Innovation Solution
A trained AI recognition model using an artificial neural network is created and embedded in a dedicated chip, which receives sensing data to determine whether to perform specific functions, allowing the device to only activate when necessary and reducing power consumption by remaining off until specific data is received.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If power is turned on for all hardware components to ensure function availability, then reliability is improved, but energy consumption increases
Solution Approach 1:
The system performs preliminary sensing and AI-based determination before activating main hardware components. The sensing unit continuously monitors for wake-up conditions, and the AI recognition model pre-evaluates whether full device activation is necessary, allowing the device to remain in a low-power state when no interaction is detected
Solution Approach 2:
The system dynamically adjusts its operational state based on real-time sensing data and AI determination. Hardware components transition between active and inactive states according to the determined function performance requirements, optimizing the balance between reliability and energy consumption
2Measurement precision
If sensing data is continuously monitored to improve function recognition accuracy, then measurement precision is improved, but energy consumption increases
Solution Approach 1:
The system performs preliminary AI-based evaluation of sensing data before activating full processing functions. The AI recognition model quickly assesses whether the sensed data requires full device activation, filtering out unnecessary processing and reducing energy consumption while maintaining accurate function recognition
Solution Approach 2:
The system applies partial processing to sensing data by using the AI recognition model to determine only the necessary function performance level. Not all sensing data triggers full device activation - only data that meets specific determination criteria activates main hardware, reducing overall energy consumption while maintaining recognition accuracy for relevant functions
3Measurement precision
If AI recognition model is trained in real-time to improve accuracy, then measurement precision is improved, but productivity decreases due to processing time
Solution Approach 1:
The AI recognition model is trained in advance using big data before deployment. This preliminary training phase allows the model to learn optimal function determination patterns offline, so that during actual operation, the model can quickly make accurate determinations without requiring time-consuming real-time training
Solution Approach 2:
The system uses a pre-trained AI recognition model that has copied learned patterns from extensive training data. Instead of training from scratch during operation, the model replicates previously learned knowledge to make rapid, accurate function determinations in real-time scenarios
4Speed
If hardware components are kept active to reduce response time, then speed is improved, but energy consumption increases
Solution Approach 1:
The system maintains only essential sensing capabilities in an active state, while keeping main hardware components in a low-power state. The AI recognition model quickly evaluates sensing data to determine if full hardware activation is needed, enabling rapid response only when necessary rather than maintaining constant high-speed readiness
Solution Approach 2:
The system dynamically transitions hardware components between active and inactive states based on AI-determined function performance needs. This dynamic state management allows the device to optimize the balance between response speed and energy consumption by activating full hardware only when the AI model determines it is necessary
Data Source
AI summary
A learning model creation method for performing a specific function for an electronic device, according to an embodiment of the present invention, can include the steps of: preparing big data for training an artificial neural network including, in pairs, sensing data received from a random sensing data generation unit for sensing human behaviors and specific function performance determination data for determining whether to perform a specific function of an electronic device with respect to the sensing data; preparing an artificial neural network model, which includes nodes of an input layer through which the sensing data is inputted, nodes of an output layer through which the specific function performance determination data of the electronic device is outputted, and association parameters between the nodes of the input layer and the nodes of the output layer, and calculates inputs of the sensing data for the nodes of the input layer in order to output the specific function performance determination data from the nodes of the output layer; and repeatedly performing a process of inputting the sensing data included in the prepared big data into the nodes of the input layer and outputting the specific function performance determination data that pairs with the sensing data included in the big data from the nodes of the output layer so as to update the association parameters, thereby mechanically training the artificial neural network model.