Processor-Based Mini-Sensor Assembly for Edge Event Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current environmental sensors used in IoT devices face challenges such as high power consumption, large size, and latency issues due to complex processors, and require significant bandwidth and storage resources, which limits their efficiency and effectiveness in monitoring and controlling systems.
Innovation Solution
The development of a processor-based mini-sensor assembly (PMA) that integrates a sensor with a low-power local processing module in a compact package, allowing for edge processing and communication through various networks, using simple or neural network architectures for event detection and data analytics, with pre-trained values for efficient event recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If complex processors are used in environmental sensors, then event detection capability is improved, but power consumption increases
Solution Approach 1:
The system segments processing tasks between edge devices (simple processors) and cloud servers (complex processors). Edge devices perform only basic event detection using simplified algorithms, while complex neural network processing is offloaded to the cloud, reducing local power consumption while maintaining detection capability.
Solution Approach 2:
A simplified edge processor acts as an intermediary between the sensor and cloud server. It performs preliminary processing and filtering of sensor data, transmitting only relevant information to the cloud, thereby reducing the computational burden and power consumption at the edge while maintaining overall system detection capability.
2Difficulty of detecting and measuring
If complex processors are used in environmental sensors, then event detection capability is improved, but device size increases
Solution Approach 1:
The system divides the processing architecture into two segments: a compact edge device with minimal processing capabilities and a remote cloud server with full computational power. This segmentation allows the edge sensor to remain small while still achieving complex event detection through cloud-based processing.
Solution Approach 2:
Complex processing functions are extracted from the edge sensor and relocated to the cloud server. The edge device retains only essential sensing and basic processing components, significantly reducing its size while maintaining event detection capability through remote processing.
3Productivity
If complex processors are used in environmental sensors, then data processing capability is improved, but latency increases
Solution Approach 1:
The edge device performs preliminary processing of sensor data locally, including filtering, feature extraction, and basic event detection. This preliminary action prepares data for cloud processing, reducing the time required for cloud-based analysis and overall system latency.
Solution Approach 2:
The system maintains continuous local processing at the edge device, ensuring that data is always being prepared and pre-processed even before cloud communication occurs. This continuous action eliminates idle time and reduces overall latency in the data processing pipeline.
4Productivity
If complex processors are used in environmental sensors, then analytics capability is improved, but bandwidth requirements increase
Solution Approach 1:
Complex analytics and heavy computational tasks are extracted from the edge device and performed remotely in the cloud. The edge device transmits only essential raw data or pre-processed features, significantly reducing bandwidth requirements while maintaining advanced analytics capability through cloud-based processing.
Solution Approach 2:
Instead of transmitting and processing all raw sensor data locally, the system transmits selected features or processed copies of the data to the cloud for advanced analytics. This approach reduces bandwidth consumption while preserving the ability to perform complex analysis.
Data Source
AI summary
Systems and methods having a local processor at the extreme edge of an environmental monitoring/control system are disclosed. A sensor is provided in close proximity to the processing module. The processing module is a relatively simple processor capable of detecting the occurrence of one particular event, or a limited number of particular events. The processing is done with a relatively simple neural network. Processing module outputs indicate that a particular event has occurred. The processing module includes a memory into which values can be loaded that are determined by “pre-training” for use with the processing module. The downloading and use of such pre-trained values allows the processing module to efficiently detect particular events for which the values were pre-trained.