Edge Device Sensor Configuration Model Update
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Smart-city systems face limitations in data processing and sensing quality due to the centralized nature of cloud computing architectures, which restricts the free flow of information and requires improved edge device capabilities for efficient data collection and analysis.
Innovation Solution
A network system comprising edge devices with sensors and processing means that can update configuration parameters using a model derived from input data, allowing for continuous improvement in sensing quality and energy efficiency, with intelligence distributed across remote devices, fog devices, and central control systems to manage computational resources effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If cloud computing architecture is used for data processing in smart cities, then centralized data management is achieved, but information flow is restricted and edge device capabilities are insufficient
Solution Approach 1:
The system segments data processing tasks between edge devices and remote devices. Edge devices perform local sensing and initial processing, while remote devices handle complex model updates. This segmentation allows edge devices to operate independently with limited computational resources while still benefiting from centralized intelligence.
Solution Approach 2:
The patent introduces a temporal dimension to the processing architecture by continuously updating models over time. Instead of requiring all computational power at the edge, the system evolves models progressively and transfers updates to edge devices, enabling complex processing capabilities to be achieved over time rather than requiring high instantaneous computational power at edge devices.
2Measurement precision
If sensor configuration parameters are fixed, then device simplicity is maintained, but sensing quality and energy efficiency cannot be improved
Solution Approach 1:
The system implements self-service through automatic model updates. Remote devices continuously improve sensing models based on collected data and automatically transfer updated models to edge devices. This eliminates the need for manual configuration while maintaining simple edge devices that automatically benefit from improved sensing quality through updated models.
Solution Approach 2:
The system incorporates feedback loops where sensing data from edge devices is used to continuously improve models at remote devices. The updated models are then transferred back to edge devices, creating a closed-loop system that automatically improves sensing quality over time without increasing edge device complexity.
3Productivity
If more computational resources are allocated to edge devices, then local data processing capability is improved, but system scalability and resource efficiency deteriorate
Solution Approach 1:
The system segments computational tasks by function and complexity. Edge devices handle time-critical local processing with minimal resources, while remote devices handle computationally intensive model training and updates. This segmentation enables scalability because edge devices remain simple and can be deployed widely, while centralized remote devices provide advanced processing capabilities.
Solution Approach 2:
The patent creates a universal architecture where remote devices serve multiple functions: collecting data from multiple edge devices, training models, and distributing updates. This multi-functionality at the remote device level allows the system to scale by adding more remote devices without increasing edge device complexity, maintaining versatility while enabling scalability.
Data Source
AI summary
Example embodiments relate to network systems with sensor configuration model updates. One example network system includes a plurality of edge devices. The plurality of edge devices is arranged at a plurality of locations. The plurality of edge devices includes at least a sensor. The sensor is configured for obtaining environmental data related to an event in the vicinity of the edge device. The sensor is set up according to at least one configuration parameter. The plurality of edge devices also includes a processing means configured to process input data in accordance with a model to derive the at least one configuration parameter of the sensor. The network system is configured to determine an updated model over time and to reset the processing means so as to process input data in accordance with the updated model.