Wearable Computing Power Scheduling for Task Offloading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wearable devices face limitations in computing power due to hardware constraints, restricting their functionality and user experience.
Innovation Solution
A computing power scheduling method that allocates tasks across multiple cores and an electronic device, determining a target computing power provider based on task requirements and available capabilities to ensure efficient processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If wearable devices use more hardware resources to increase computing capability, then functionality and user experience are improved, but power consumption increases and hardware constraints are exacerbated
Solution Approach 1:
The patent segments computing tasks into different categories based on their computational requirements. Simple tasks are handled by lightweight models on the wearable device, while complex tasks are offloaded to cloud-based powerful models. This segmentation allows the system to achieve high computing capability for complex tasks without continuously running high-power components on the wearable device, thus reducing overall power consumption.
Solution Approach 2:
The patent implements dynamic task allocation that adjusts computing resource usage based on real-time conditions. The system dynamically determines whether to process tasks locally or in the cloud based on task complexity, available power, and network conditions. This dynamic approach allows the computing capability to adapt to power availability, ensuring high performance when needed while conserving energy during power-constrained periods.
2Productivity
If wearable devices allocate more computing resources to complex tasks, then task processing capability is improved, but the limited hardware resources are exhausted faster
Solution Approach 1:
The patent introduces a cloud computing service as an intermediary between the wearable device and complex computational tasks. Instead of requiring the wearable device's limited hardware resources to handle all computational demands, the system uses the cloud as a mediator to provide additional computing power. This allows complex tasks to be processed with high capability while the wearable device's hardware resources are preserved for essential functions.
3Speed
If wearable devices process all tasks locally, then response speed is improved, but computing power requirements exceed device capabilities
Solution Approach 1:
The patent segments tasks into two categories: simple tasks processed locally on the wearable device for fast response, and complex tasks offloaded to the cloud where powerful models can handle them. This segmentation ensures that time-critical simple tasks maintain fast local response speed, while complex tasks that require more computing power are handled by the cloud infrastructure, preventing the wearable device from being overwhelmed by computational demands.
Data Source
AI summary
A computing power scheduling method, a wearable device, and a storage medium are provided. The computing power scheduling method includes: when a first core obtains a first task, determining whether a computing power required by the first task exceeds a computing power range of the first core; when the computing power required by the first task exceeds the computing power range of the first core, determining a target computing power provider according to the computing power required by the first task, the computing power ranges of the N cores, and a computing power range of an electronic device, wherein the computing power required by the first task falls within a computing power range of the target computing power provider, and the target computing power provider is at least one of the electronic device and the N cores; and processing the first task by using the target computing power provider.