Oscilloscope triggering upon recognising a portion of a 'sampled' signal

[anfractuosity]

Hi,

I'm wondering if anyone has seen an oscilloscope feature, where you capture a signal, choose and record part
of that signal and then the oscilloscope will run a sliding window over newly captured data, and will trigger
upon recognising the same recorded portion of signal (e.g. comparing with a simple mean squared error and threshold)
and produce a 'high' output also, when triggered.

This would be for use for side channel analysis and fault injection.

I've seen specialised devices such as the Chipwhisperer Husky can accomplish this, I'm just wondering
if this can also be achieved with a traditional oscilloscope.

I've had a little look at features such as zone triggering, which wouldn't really allow complex waveforms to be detected

Many thanks

[RoGeorge]

Rigol DS1054z has a mask pass/fail test.



Can only tell if the signal on the screen "fits" inside a predefined area/mask, and the pass/fail result of the comparison can be read remotely by SCPI, for automated testing.  It does not slide/seek/detect along the signal, only compares full screen.

Does not trigger on match, it is your job to trigger and set the oscilloscope such that the mask and the active signal look/fit the same on the display.

[tautech]

I'm wondering if anyone has seen an oscilloscope feature, where you capture a signal, choose and record part
of that signal and then the oscilloscope will run a sliding window over newly captured data, and will trigger
upon recognising the same recorded portion of signal (e.g. comparing with a simple mean squared error and threshold) and produce a 'high' output also, when triggered.

A Mask could be suitable but only for repetitive waveforms.

I've had a little look at features such as zone triggering, which wouldn't really allow complex waveforms to be detected

A dual Zone trigger and Trig Out might better suit you needs along with deep memory and Search capability, all of which are configurable but you won't get all this functionality in a bargain basement priced DSO.

What's your budget ?

[anfractuosity]

Thanks both for the replies.

I thought with a zone trigger you define areas where a signal intersects or doesn't, which I didn't think would allow detecting of a more complex waveform, e.g. say looking for 4 peaks
at potentially different amplitudes.

When you mention search capability though, is that something else, more akin to the sliding window and mean square error comparison?

I'm looking at spending £1k max.

[tautech]

I thought with a zone trigger you define areas where a signal intersects or doesn't, which I didn't think would allow detecting of a more complex waveform, e.g. say looking for 4 peaks at potentially different amplitudes.

Yes zone triggers work like that and when you have a couple of them triggering on complex waveforms becomes much easier but difficult when:

I'm looking at spending £1k max.

Cheaper DSO's do offer a complex trigger suite far beyond just Edge triggers but without an example of what you are working with it's difficult to give advice.

Screenshot if it's not too much trouble please.

When you mention search capability though, is that something else, more akin to the sliding window and mean square error comparison?

A Search example from the $499 SDS1104X-E:
https://www.eevblog.com/forum/testgear/siglent-sds1204x-e-released-for-domestic-markets-in-china/msg1370717/#msg1370717

[anfractuosity]

Sorry for the delay, I've attached an example from my scope.

The section of the waveform marked in green, is what I'd like to search for and to immediately after the green segment, for the scope to generate a 'high' pulse.

[RAPo]

Just an idea:
Do you have the green signal always present on its own channel?
For a two-channel scope with math options and pulse width trigger possibility where the trigger source can be math channel:
Put your signal on channel 1, the green signal on channel 2, calculate channel 1 - channel 2 and trigger on this channel for a pulse width according to the length of the green signal with a suitable holdoff
Not a fail safe option (there can be other regions that cancel out).

If the question is about extracting the green part, a mso5000 series scope can transfer a part of the signal to a waveform editor, select the wanted range and use that as a new awg shape.

[tautech]

Sorry for the delay, I've attached an example from my scope.

The section of the waveform marked in green, is what I'd like to search for and to immediately after the green segment, for the scope to generate a 'high' pulse.

Triggering on part of that should be no problem as I see 2 characteristics you can specify with either an Interval or Runt trigger.
Most modern DSO's can do this.

[anfractuosity]

RAPo - unfortunately the green signal isn't always present on its own channel.  Will have a look at the waveform editor you mention, that sounds interesting.

tautech - Thanks, will look at both of those trigger modes.