Feature: Find pattern in a channel

[bdutta]

This feature would allow one to look for a signal pattern, within the captured signal for a given channel. The way to specify the signal pattern could be either -

1. defined in a .csv file as: (state, duration) pairs
2. visually copy a section of same capture, or another (saved) capture, to create a compare buffer

It should be possible to click 'Next' to do a forward search for the given pattern. It should be possible to specify a tolerance (as percentage) of duration to accept/ignore minor signal variances.

[jawi]

You mean some kind of serial trigger?

[bdutta]

Not exactly, but then to be honest, I am new to logic-analyzers so this could be a mistake on my part. My understanding was that serial triggers, are conditionals that when fulfilled start the capture. What I am looking for is a tool / aid during the offline analysis of captured data.

[jawi]

True, serial triggers are used to define what is to be captured.

On the other hand, most of the times you need to define such a trigger because the capturing memory is simply limited for capturing long time ranges. Hence, my earlier comment.

Could you provide a (detailed) use case on how this kind of functionality would aid you?

[bdutta]

Consider a case where someone wants to use the Analyzer to determine highest and lowest frequency of signal present in a capture. The way this could help (although may not be with 100% certainty) is by helping identify the lowest frequency segment(s), which could be indicative of high freq noise. Today the way I do it is via visual inspection, which is time-consuming, inaccurate and very inconvenient, because I can only see what a single window-full allows at a zoom level where the smallest marks/spaces are visible. Ditto for large marks/spaces. Also, if there were a way to identify marks/spaces of given duration-range, I could find specific parts of signal that are really interesting. Of course, being able to detect patterns "within" signal that is already captured, without having to do full visual inspection, would be the best thing.

In a nutshell, today the best use case I can think of is separating wheat from chaff. If you know all about the signal, seeing a noise-free line, then this isn't useful, but otherwise is.