IoT Sensor Data Compression via Point Cloud Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current IoT systems face challenges in efficiently handling and storing large volumes of data from geographically distributed sensors, particularly in LPWANs, due to high storage and communication resource needs, despite existing compression methods offering limited efficiency and primarily focusing on low power usage rather than data aggregation.
Innovation Solution
A system and method that employs spatial correlation to compress and aggregate IoT data using a point cloud-based geometric data model, enabling efficient lossy and lossless compression and archiving, which can be processed at IoT gateways or edge servers, reducing storage costs and supporting increased connectivity in next-generation networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data from thousands to millions of sensors are transmitted and stored, then complete data availability is achieved, but storage and communication resource needs become huge
Solution Approach 1:
The patent extracts only the essential and relevant features from the massive sensor data using feature extraction techniques. Instead of storing and transmitting complete raw data from all sensors, the system identifies and extracts key characteristics that maintain data reliability while significantly reducing the quantity of data that needs to be stored and communicated.
Solution Approach 2:
The patent transforms the sensor data by changing its parameters through compression algorithms and aggregation operations. The raw high-volume sensor data is converted into compressed representations with modified parameters that preserve the essential information needed for reliability while reducing the overall data volume and resource requirements.
2Loss of information
If all sensor data is transmitted for processing, then complete information is available for analytics, but communication link bandwidth becomes burdened
Solution Approach 1:
The patent segments the sensor network into hierarchical groups and clusters, processing and aggregating data at multiple levels before transmission. This segmentation allows local preprocessing and filtering of data, ensuring that only essential information is transmitted over the communication links, thus maintaining information completeness while reducing bandwidth consumption.
Solution Approach 2:
The patent performs preliminary data processing, aggregation, and filtering actions at the sensor nodes and edge devices before data transmission. By conducting these preliminary actions locally, the system ensures that the most relevant information is prepared and condensed in advance, reducing the communication burden while maintaining the completeness of essential information for analytics.
3Quantity of substance
If compression is applied to reduce data size, then storage and transmission efficiency improves, but data loss may occur
Solution Approach 1:
The patent implements feedback mechanisms where the compression and aggregation processes continuously monitor and adjust their operations based on data characteristics and quality requirements. This feedback ensures that compression is applied in a way that minimizes information loss, maintaining data accuracy by adapting the compression level and method based on the specific sensor data being processed.
Solution Approach 2:
The patent carefully manages parameter changes during compression by selecting and controlling which parameters are modified and to what extent. The system transforms data parameters in a controlled manner that achieves size reduction while preserving the critical parameters necessary for data accuracy and analytical value, balancing compression efficiency with information retention.
Data Source
AI summary
A system and method to compress, aggregate and archive Internet of Things data originating from sensors can use a point cloud based geometric data model and a distribution and aggregation method based on compressed point cloud representations. The compression can occur at a gateway or a point of presence near the access point, or alternatively in an IoT server located anywhere else in a network, such as in a core data center. In one embodiment a cloudlet in an access point performs the data modeling and compression at the access point. In some embodiments, a system or method stores and distributes data to relevant entities formatted in a point cloud based model that combines (x,y,z) geometry attributes with values attached in the attributes (a1, a2, a3) and uses point cloud compression. A receiver can then decode the point cloud and recompose the aggregated sensor data, enabling inspection of the data.