Distributed Luminaire Anomaly Detection via Spatial-Temporal Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The excessive data transmission from luminaires with onboard sensors can overwhelm network bandwidth, waste computational resources, and hinder diagnostic tests, as existing systems lack efficient methods to predict and address hardware and network anomalies without indiscriminate data collection.
Innovation Solution
Implementing a distributed processing system that uses spatial-temporal models generated from sensor data to identify and predict anomalies, allowing individual luminaires to process data locally and compensate for variances, thereby reducing unnecessary data transmission and conserving resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If luminaires indiscriminately collect and transmit sensor data, then comprehensive environmental monitoring is achieved, but network bandwidth is overwhelmed and diagnostic capabilities are hindered
Solution Approach 1:
The patent segments the centralized data collection approach into distributed processing units at each luminaire. Each luminaire independently processes its sensor data using local spatial-temporal models, performing anomaly detection and feature extraction before transmission. This segmentation reduces network traffic by transmitting only processed results rather than raw sensor streams, resolving the bandwidth conflict while maintaining monitoring comprehensiveness.
Solution Approach 2:
The patent implements preliminary action by pre-computing spatial-temporal models and anomaly detection algorithms at each luminaire before data transmission. Each luminaire performs real-time processing of sensor data against pre-loaded models, identifying and flagging anomalies locally. This preliminary processing ensures that only relevant anomaly data, not complete sensor datasets, are transmitted over the network, preserving bandwidth while maintaining monitoring effectiveness.
2Reliability
If all sensor data from all luminaires is transmitted for processing, then accurate anomaly detection is achieved, but computational resources are wasted on processing non-informative data
Solution Approach 1:
The patent applies local quality by enabling each luminaire to independently process and evaluate its own sensor data using locally-stored spatial-temporal models. Each luminaire performs anomaly detection with high accuracy using only its own data and local computational resources, rather than contributing raw data to a centralized processing system. This localized approach maintains detection reliability while eliminating the computational waste of transmitting and processing non-informative data from all luminaires centrally.
3Ease of repair
If network bandwidth is allocated for comprehensive data transmission, then diagnostic tests can be performed, but the volume of data to be transmitted exceeds available bandwidth
Solution Approach 1:
The patent extracts only the essential anomaly-related features from sensor data at each luminaire using local spatial-temporal models. Instead of transmitting complete sensor datasets for diagnostic analysis, each luminaire extracts and transmits only anomaly indicators, timestamps, and relevant metadata. This extraction approach enables effective diagnostic testing while reducing data transmission volume to manageable levels that fit within available network bandwidth.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The described implementations provided herein relate to systems, methods, and apparatuses for using a network of luminaires to perform distributed computation of sensor data to identify hardware and network anomalies. In some implementations, a method is set forth as including operations such as receiving, from a first luminaire (114, 128, 216), first sensor data (122, 124) in response to a stimulus (204) affecting a network of luminaires (110), and receiving, from a second luminaire (208) in the network of luminaires, second sensor data. The method can also include determining a correlation between the first sensor data and the second sensor data, and modifying a luminaire spatial-temporal model (222) based at least partially on the correlation. The method can also include receiving subsequent sensor data from the first luminaire or the second luminaire, and providing a signal (212) to the first luminaire or the second luminaire based on a comparison of the subsequent sensor data to the luminaire spatial-temporal model.