Twin Inference Models for Data Transmission Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data collection systems face high computing resource costs due to the transmission of large data volumes across distributed systems, which can lead to increased energy consumption and reduced availability of resources for other tasks.
Innovation Solution
Implementing a multi-stage data reduction process using twin inference models at data aggregators and collectors, where feature relationship inference models identify relationships in collected data, allowing for selective transmission and reconstruction of data at the aggregator, thereby reducing the amount of data transmitted and conserving computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all collected data is transmitted to the data aggregator, then data accuracy is maintained, but computing resource consumption increases
Solution Approach 1:
The patent extracts only the essential features and relationships from the collected data using inference models, rather than transmitting all raw data. The data collector identifies and transmits only the most relevant data portions that cannot be accurately inferred, while the aggregator reconstructs the complete data set using the transmitted features and inference models.
Solution Approach 2:
The patent creates inference models that copy the essential patterns and relationships of the data at the collector, allowing the aggregator to reconstruct data without receiving all original data. The transmitted data includes copies of only the critical features needed for accurate reconstruction.
2Use of energy by moving object
If data transmission volume is reduced, then computing resources are conserved, but data representation accuracy may deteriorate
Solution Approach 1:
The patent implements feedback mechanisms where the data aggregator evaluates the accuracy of reconstructed data and adjusts the data reduction plan accordingly. Error thresholds are monitored and used to refine which features are transmitted versus inferred, ensuring accuracy is maintained while minimizing transmission.
Solution Approach 2:
The patent transmits slightly more data than the absolute minimum required, including redundant features that provide error margins for reconstruction. This partial excess ensures that even with compression, the reconstructed data remains within acceptable accuracy thresholds.
3Loss of energy
If feature relationship inference models are used to reduce data, then transmission costs decrease, but system complexity increases
Solution Approach 1:
The patent segments the data processing function into two distinct parts: the data collector that extracts and transmits only essential features, and the data aggregator that reconstructs complete data using inference models. This segmentation allows each component to be optimized independently, managing overall system complexity.
Solution Approach 2:
The patent introduces feature relationship inference models as intermediary components that bridge the gap between reduced transmitted data and the original complete data set. These models act as mediators that enable accurate reconstruction without requiring transmission of all original data.
Data Source
AI summary
Methods and systems for managing data collection are disclosed. A data aggregator may aggregate data collected by a data collector. To reduce computing resources used for aggregation, the data aggregator and data collector may implement a multi-stage data reduction processes to reduce the quantity of data transmitted for data aggregation purposes. The multi-stage data reduction process may include implementing twin inference models at the aggregator and collector, identifying relationships in the data collected by the data collector using feature relationship inference models, transmitting a portion of the collected data to the data aggregator and withholding a second portion of the collected data based on acceptable level of error for use of the collected data, and reconstructing the withheld portion of the collected data at the aggregator. The reconstructed portion of the collected data may include the acceptable level of error when compared to a corresponding portion of the collected data.