IoT Event Data Conversion to Multidimensional Time Series
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing IoT platforms struggle to convert and store time series data from IoT events in a format suitable for data analysis using machine learning, deep learning, and statistics, making big data analysis difficult.
Innovation Solution
A method that converts sensing data from IoT devices into event information, which is then reconverted into multidimensional time series data according to specific conditions, and stored in a database in a format conducive to data analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If sensing data is stored in event format with multiple fields (event identifier, timestamp, device name, resource name, value type, value), then data completeness and information retention are improved, but data analysis efficiency and ease of use deteriorate due to the need for complex filtering and processing
Solution Approach 1:
The patent segments the event data into two distinct formats: event format for complete information storage and time series format for efficient analysis. The conversion process divides data into discrete time series tuples that can be directly utilized by analysis applications without complex filtering, thus resolving the contradiction between information completeness and analysis efficiency.
Solution Approach 2:
The patent introduces an intermediary conversion process that transforms event data into time series data. This intermediary step acts as a bridge between the event storage system and the analysis application, pre-processing the data into a format that is optimized for analysis while preserving all original information through structured conversion.
2Adaptability or versatility
If event data is converted and filtered according to multiple conditions (base device name, time limit, time unit, device name, filter list), then data relevance and analysis focus are improved, but processing complexity and time consumption increase
Solution Approach 1:
The patent applies preliminary filtering and conversion actions during the data ingestion phase. By converting event data to time series format and applying filters upfront, the system prepares data in advance for analysis, eliminating the need for complex filtering operations at query time and thus reducing processing time while maintaining data relevance.
Solution Approach 2:
The patent changes the parameter representation from event-based fields to time series tuples with structured attributes. This parameter transformation enables more efficient filtering and querying by analysis applications, as the converted format naturally supports time-based and device-based filtering without complex conditional logic.
3Quantity of substance
If all sensing data from multiple devices and resources is stored in detail, then data comprehensiveness is improved, but storage requirements and processing overhead increase
Solution Approach 1:
The patent transforms the data from a multi-field event structure to a time series tuple structure, adding a temporal dimension as the primary organizing principle. This dimensional change allows comprehensive storage of all sensing data while simplifying processing, as the time series format naturally groups related data points and enables efficient time-based operations.
Data Source
AI summary
Proposed is a method for converting and storing time series data of Internet of Things (IoT) events. The method may include converting sensing data received from an IoT device into an event according to a preset first data format. The method may also include converting the event into a tuple according to a preset second data format. The method may further include storing the time series data in a database table in a multidimensional time series data format based on the tuple.