IoT Selective Measurement Reporting via Entropy Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
IoT devices face connectivity issues that lead to data loss and subsequent network overload when reconnecting, as they typically replay all cached measurements upon restoration, overwhelming bandwidth and processing capabilities.
Innovation Solution
IoT devices implement selective reporting based on entropy analysis, prioritizing measurements with higher rates of change or those outside statistical thresholds, reducing the number of reported messages and minimizing network traffic and processing costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If IoT devices replay all cached measurements upon network restoration, then data completeness is improved, but network bandwidth and processing capabilities are overwhelmed
Solution Approach 1:
The patent extracts only the most important measurements (those with highest entropy or rate of change) from the complete cached dataset and transmits only these selected measurements upon network restoration. This extraction approach maintains data completeness for critical information while dramatically reducing network traffic volume compared to replaying all cached measurements.
Solution Approach 2:
The patent applies local quality by differentiating measurements based on their individual characteristics (entropy, rate of change) and treating them differently. High-entropy measurements are prioritized for transmission while low-entropy measurements are filtered out, creating a quality-filtered data stream that optimizes both completeness and bandwidth efficiency.
2Reliability
If IoT devices send all cached measurements upon connectivity restoration, then measurement reporting reliability is improved, but network bandwidth and processing capabilities are taxed excessively
Solution Approach 1:
The system extracts and transmits only the most reliable and significant measurements (those with highest entropy or rate of change) rather than all cached measurements. This selective extraction maintains measurement reporting reliability for critical data while reducing overall network energy consumption by avoiding transmission of redundant or low-value measurements.
Solution Approach 2:
The patent changes the parameter of measurement selection criteria from simple completeness-based replay to entropy-based prioritization. By transforming the selection parameter from binary (send/all or send/none) to a continuous entropy-based ranking system, the system achieves both reliable reporting of important measurements and reduced energy consumption.
3Productivity
If IoT devices implement selective measurement reporting based on entropy analysis, then network efficiency is improved, but device complexity increases
Solution Approach 1:
The patent simplifies device complexity by changing the measurement selection approach to a single dominant parameter (entropy) rather than complex multi-parameter analysis. This parameter transformation enables network efficiency improvements through selective reporting while keeping the implementation relatively simple and computationally feasible for edge devices.
Solution Approach 2:
The system segments the measurement selection process into discrete, manageable steps: calculate entropy for each measurement, rank measurements by entropy value, and transmit top-N measurements. This segmentation of the complex selection process into simple, sequential operations maintains high network efficiency while avoiding excessive computational complexity at the device level.
Data Source
AI summary
In various embodiments, an IoT device may provide selective reporting of collected data measurements. The IoT device may report the data via a network connection to an aggregator device. The IoT device may detect when the network connection has been interrupted, during which messages containing measurements may be cached. Later, when the network connection has been restored, the IoT device may “replay” the cached messages. The IoT device may selectively report cached messages based on an entropy analysis which may detect which measurements exhibit a higher entropy. The entropy analysis may determine which measurements show a higher rate of change or which have a value outside of a set of thresholds. The IoT device may select measurements based on a history of measurements obtained by the IoT device. Other embodiments may be described and/or claimed.