Software-Defined Sensing Network Node Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor networks face challenges in managing large data volumes, particularly with video data, which strains networking infrastructure and poses security risks due to constant data transmission, especially in public infrastructure like the Internet.
Innovation Solution
The implementation of a software-defined sensing network with light-leaf nodes (LLNs) and heavy-leaf nodes (HLNs) that perform processing on sensing events to generate distilled data, minimizing raw data transmission and enhancing security by reducing data sent over the network, while allowing for predictive, adaptive, and collaborative sensing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor networks transmit large volumes of raw data (especially video data) to processing devices, then comprehensive monitoring and detection capabilities are improved, but network bandwidth consumption increases and security risks are amplified
Solution Approach 1:
The patent segments the sensor network into two distinct types of nodes: light-leaf nodes (LLNs) that perform sensing and initial processing, and heavy-leaf nodes (HLNs) that perform sophisticated processing. This segmentation allows processing to be distributed closer to the data source, reducing the need to transmit raw data across the network while maintaining comprehensive detection capabilities.
Solution Approach 2:
The patent implements preliminary action by having LLNs perform initial processing of sensing events to generate distilled data before transmission. This pre-processing step extracts only the most relevant information from raw sensor data, significantly reducing network bandwidth requirements while ensuring that HLN receives data ready for sophisticated analysis.
2Loss of information
If sensor networks transmit large volumes of raw data continuously, then data availability for analysis is improved, but security risks increase due to constant data transmission over public infrastructure
Solution Approach 1:
The patent applies the extraction principle by removing unnecessary data from transmission. LLNs extract only the essential features and distilled information from raw sensing data, transmitting only what is necessary for HLN analysis. This minimizes the data surface exposed to potential security threats while maintaining data availability for sophisticated processing.
Solution Approach 2:
The patent implements local quality by enabling each LLN to perform processing locally based on its capabilities and the specific sensing events it detects. This localized processing ensures that data is available where needed while reducing network transmission, thereby minimizing security exposure to public infrastructure threats.
3Device complexity
If sophisticated processing is performed only at centralized devices, then processing power requirements at sensor nodes are reduced, but data transmission volumes increase
Solution Approach 1:
The patent applies dynamics by making the processing capability distribution flexible and adaptive. LLNs dynamically perform processing based on their capabilities and the nature of sensing events, while HLN provides sophisticated processing when needed. This dynamic approach optimizes the balance between local processing and centralized analysis, reducing data transmission without requiring all sensor nodes to have high processing power.
Data Source
AI summary
Low-level nodes (LLNs) that are communicatively connected to one another each have sensing capability and processing capability. High-level nodes (HLNs) that are communicatively connected to one another and to the LLNs each have processing capability more powerful than the processing capability of each LLN. The LLNs and the HLNs perform processing based on sensing events captured by the LLNs. The processing is performed by the LLNs and the HLNs to minimize data communication among the LLNs and the HLNs, and to provide for software-defined sensing.