Metadata-Driven Analytics Selection for Premises Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing premises monitoring systems face challenges in efficiently analyzing and responding to various events due to the sheer volume and diversity of data generated by multiple devices, leading to potential delays or misinterpretations in critical situations.
Innovation Solution
The implementation of a premises monitoring system that incorporates edge analytics units, system analytics units, and remote analytics units, which selectively apply analytics models such as machine learning and artificial intelligence to metadata and data collected from premises devices, enabling real-time analysis and prediction of events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple devices of various types are deployed for monitoring premises, then the quantity and diversity of data collected increases, but the complexity of analyzing and responding to events increases
Solution Approach 1:
The system segments the analytics processing into multiple distributed units: edge analytics units at individual premises devices, system analytics units at the control device, and remote analytics units at external servers. This segmentation allows each unit to handle specific analytics tasks independently, reducing the complexity burden on any single component while collectively processing diverse data from multiple devices.
Solution Approach 2:
The patent introduces metadata as an intermediary layer between raw data and analytics processing. Metadata elements describe characteristics of the data and conditions, enabling the system to selectively apply appropriate analytics models without manually configuring each device. This intermediary layer simplifies the complexity of handling diverse data by providing structured information about data characteristics.
2Measurement precision
If analytics models are applied to all data from multiple devices, then the accuracy of event detection improves, but the processing time and computational resources increase
Solution Approach 1:
The system applies analytics models selectively rather than universally. The edge analytics unit evaluates metadata and applies only those analytics models that are relevant to the current data and conditions. This partial action approach maintains detection accuracy for critical events while avoiding the computational overhead of applying all possible analytics models to every data stream.
Solution Approach 2:
Different analytics models are applied to different data streams based on local conditions and data characteristics described in metadata. Each edge device can select and apply analytics models appropriate to its specific sensor types and monitoring requirements, rather than using a uniform analytics approach across all devices. This local quality approach optimizes processing time by applying only necessary analytics.
3Productivity
If selective analytics application is implemented using metadata elements, then the processing efficiency improves, but the system complexity increases
Solution Approach 1:
The patent creates a universal metadata structure that can describe various types of data from different sensor devices using a common framework. This metadata schema enables the same selective analytics application logic to work across diverse data types and devices, improving processing efficiency without proportionally increasing system complexity. The universal metadata approach allows one system to handle many different device types.
Data Source
AI summary
A method in a premises device is provided, where the premises device is configured with a plurality of analytics models and located in a premises environment. The method includes generating a dataset comprising surveillance data associated with a premises environment in which the premises device operates, determining metadata associated with the dataset, determining, for each analytics model of the plurality of analytics models, a corresponding predicted performance score based at least in part on the metadata, selecting a highest-scoring analytics model of the plurality of analytics models based at least in part on the corresponding predicted performance score, and analyzing the dataset using the highest-scoring analytics model to generate a prediction.