Sensor Signal Feature Matching for Local-Remote Event Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor-enabled devices face challenges in efficiently processing heterogeneous and complex sensor signal data, leading to increased costs and difficulties in reliable event detection, especially when limited by hardware and software resources, and remote processing solutions are hindered by power consumption and latency issues.
Innovation Solution
A hybrid processing method that generates a feature profile from sensor data, matching it locally against predetermined profiles, and if a match is not found, the data is processed remotely, utilizing a local evaluation mode for routine operations and a remote evaluation mode for complex cases, optimizing resource use and reducing data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If sensor signal data is processed locally by embedded processing equipment, then response time and operational autonomy are improved, but processing accuracy and reliability deteriorate due to limited hardware and software resources
Solution Approach 1:
The patent divides processing into two segments: local processing for simple, time-critical operations and remote processing for complex, accuracy-critical operations. The processing equipment is segmented into embedded processors in end devices and remote servers, allowing each to handle appropriate task types independently
Solution Approach 2:
The system dynamically selects between local and remote processing based on task complexity and resource availability. The processing architecture transitions from static embedded-only processing to a flexible hybrid model where processing location adapts to operational requirements
2Measurement precision
If heterogeneous sensor data from multiple sensors is processed to improve event detection accuracy, then measurement precision is improved, but device complexity and processing resource requirements worsen
Solution Approach 1:
The patent extracts complex processing tasks from resource-constrained end devices and relocates them to remote servers with充足的 computational resources. Only essential local processing remains in embedded equipment, while sophisticated multi-sensor fusion algorithms are executed remotely
Solution Approach 2:
A communication network acts as an intermediary between end devices and remote servers, enabling the transfer of sensor data and processing results. This intermediary allows complex processing to occur remotely while maintaining system integration and coordination
3Reliability
If all sensor data is transmitted to remote computing systems for processing, then processing accuracy is improved, but power consumption and network resource usage worsen
Solution Approach 1:
Instead of transmitting all sensor data for all operations, the system applies partial processing locally and only transmits data when necessary. Simple events are handled locally without transmission, while complex or uncertain cases are selected for remote processing, reducing overall data transmission volume
Solution Approach 2:
The system changes the parameter of processing location from always-remote to conditionally-local based on task characteristics. This parameter change optimizes the balance between accuracy and energy consumption by selecting the appropriate processing mode for each specific operation
Data Source
AI summary
A method (10) of and a device (50) for performing an operation based on sensor signal data obtained from at least one sensor. From the sensor signal data, by a processor (54) of the device (50), a feature profile is generated that is matched (12) to a set of predetermined feature profiles. The operation is performed (13) by the device (50) based on the sensor signal data, if the generated feature profile matches at least one of the set of predetermined feature profiles, and the operation is performed (14) by the device (50) based on remote processing (40) of the sensor signal data (40) if the generated feature profile does not match at least one of the set of predetermined feature profiles. With the disclosed method, balance in processing power, processing time, and reliable operation by the device (50) is achieved without incurring extra cost for upgrading the device (50).