IoT Data Inconsistency Prediction With ML and Adaptive Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
IoT systems experience data inconsistency due to heterogeneous devices, varying sampling rates, and different communication conditions, leading to inefficiencies and performance issues in edge cloud applications.
Innovation Solution
An automated system using machine learning (ML) models, such as LSTM and CNN, to predict IoT data inconsistency by identifying contributing factors and generating inconsistency rules to prevent data inconsistency before it occurs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual monitoring and adjustment of sampling rates for IoT devices is implemented, then data consistency can be maintained, but the system complexity and operational effort increase significantly
Solution Approach 1:
The system enables self-service by automatically detecting data inconsistency issues and adjusting sampling rates without human intervention. The monitoring system autonomously identifies inconsistent data patterns across IoT devices and dynamically modifies their sampling rates to resolve inconsistencies, eliminating the need for manual configuration and reducing operational complexity.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring data quality metrics and using this information to dynamically adjust sampling rates. The monitoring system analyzes incoming data streams, detects inconsistency patterns, and feeds this information back to automatically modify device sampling behavior, creating a closed-loop control system that maintains data consistency adaptively.
2Reliability
If sampling rate is increased for all IoT devices to improve data consistency, then data quality improves, but energy consumption and network load increase
Solution Approach 1:
The system applies local quality by customizing sampling rates for individual IoT devices based on their specific needs and observed data inconsistency patterns. Rather than uniformly increasing sampling rates across all devices, the monitoring system identifies which specific devices exhibit inconsistency and adjusts only their sampling rates, thereby maintaining data quality while minimizing energy consumption and network load on devices that do not require higher sampling rates.
3Measurement precision
If automated monitoring system is deployed to detect data inconsistency, then detection accuracy improves, but system complexity and computational resources increase
Solution Approach 1:
The system performs preliminary action by pre-defining inconsistency detection rules and thresholds before deployment. The monitoring system is pre-configured with knowledge of what constitutes data inconsistency for different device types and protocols, allowing it to accurately detect issues without requiring complex real-time analysis algorithms. This pre-processing of detection logic reduces computational complexity while maintaining high detection accuracy.
Data Source
AI summary
The disclosure relates to a method and apparatus for predicting Internet of Things (IoT) data inconsistency. The method comprises obtaining labelled IoT data, the IoT data being collected from a plurality of IoT devices by a monitoring system. The method comprises analyzing characteristics of the labelled IoT data and identifying features of IoT data inconsistency. The method comprises training, using the labelled IoT devices data and the features of IoT data inconsistency, a ML model to predict the IoT data inconsistency. The method comprises generating, using the labelled IoT data and the features of IoT data inconsistency, a set of inconsistency rules to be applied to live IoT data predicted as inconsistent by the ML model.