Analog Signal Mixing for Sparse Event Detection With Fewer Channels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting sparse events in time require monitoring individual signals from multiple devices over extended periods, which is inefficient and resource-intensive, especially when events are expected to occur sparsely.
Innovation Solution
The use of analog circuitry to mix down a large set of analog signals using compressive-sensing techniques, reducing the number of measurements needed by applying gain factors to the signals, allowing for event detection with a smaller number of channels, specifically designed for sparse event occurrences in electronic device testing and other applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If individual signals from multiple devices are monitored separately over extended periods, then detection reliability is improved, but resource consumption and system complexity increase significantly
Solution Approach 1:
The patent combines multiple individual signal monitoring channels into a single compressed measurement channel by mixing signals from N devices through a mixing matrix to produce M measurements where M < N. This merging approach maintains detection reliability for sparse events while significantly reducing the number of monitoring channels required, thereby reducing system complexity and resource consumption.
Solution Approach 2:
The patent transforms the monitoring problem from the time domain to a compressed measurement domain by applying a mixing matrix transformation. Instead of monitoring N signals individually over time, the system creates M compressed measurements that encode information from all N signals, enabling sparse event detection with fewer channels through mathematical transformation.
2Productivity
If the number of monitoring channels is reduced, then resource usage is optimized, but the ability to detect sparse events accurately deteriorates
Solution Approach 1:
The patent changes the measurement parameters by using compressive sensing theory to determine the optimal number of measurements M based on the sparsity level k and signal dimension N. The relationship M ≥ C(k log(N/k)) ensures that even with reduced channels, the system can accurately detect sparse events by adjusting the measurement matrix design and utilizing the sparsity property of the signals.
Solution Approach 2:
The patent replaces the traditional mechanical approach of individually monitoring each signal channel with a mathematical transformation approach using compressive sensing. Instead of physically maintaining N separate monitoring channels, the system uses a mixing matrix transformation to compress N signals into M measurements, substituting physical channel resources with mathematical processing capabilities.
3Device complexity
If compressive sensing techniques are applied to mix down signals, then the number of required channels is reduced, but the complexity of signal processing increases
Solution Approach 1:
The patent introduces a mixing matrix as an intermediary transformation that bridges the gap between N input signals and M compressed measurements. This intermediary mathematical transformation enables the compression of signal channels while preserving the information necessary for sparse event detection, managing the trade-off between channel reduction and processing complexity through structured matrix operations.
Data Source
AI summary
Under one aspect, a method is provided for detecting events that are sparse in time. The method can include (a) receiving N analog input signals that are continuous and are independent from one another, wherein each one of the events causes a change in a corresponding one of the analog input signals, and N is 2 or greater. The method also can include (b) by a first analog circuit, for each of the N analog input signals, outputting products of that analog input signal and a plurality of gain factors. The method also can include (c) by a second analog circuit, outputting M sums of the products, wherein M is 2 or greater and is less than or equal to N. The method also can include (d) detecting a first one of the events based on the M sums of the products.