Forum: GNU Radio help for peak detector

Announcement (2017-05-07): www.ruby-forum.com is now read-only since I unfortunately do not have the time to support and maintain the forum any more. Please see rubyonrails.org/community and ruby-lang.org/en/community for other Rails- und Ruby-related community platforms.
85ffd97bae0e78056e1d06312744d7d1?d=identicon&s=25 Markus Feldmann (Guest)
on 2009-03-31 22:24
(Received via mailing list)
Hi All,

i want to build a peak detektor with a Usrp(LFRX),so i hope some of
you can give me some hints about the peak detector modules ?

As i know, there are 3 peak detector:
1. Detect Peak
2. Peak Detector
3. Peak Detector2

Therefore i have some questions:
To 2.) What allowed values for <Look Ahead> ?
        And what does this mean ? Which unit ?
        What is Alpha, and which are allowed values ?
        If the peak detector gets a peak, what does it spit
        out ?
        Which are the allowed values for <TH Factor Rise/Fall>,
        is this in db or a linear factor ?

Is there any example for a peak detector ?
Is there also a peak detector with a complex input ?

Regards Markus
85ffd97bae0e78056e1d06312744d7d1?d=identicon&s=25 feldmaus (Guest)
on 2009-04-07 15:18
(Received via mailing list)
Markus Feldmann <feldmann_markus <at> gmx.de> writes:

>
> Therefore i have some questions:
> To 2.) What allowed values for <Look Ahead> ?
>         And what does this mean ? Which unit ?
>         What is Alpha, and which are allowed values ?
>         If the peak detector gets a peak, what does it spit
>         out ?
>         Which are the allowed values for <TH Factor Rise/Fall>,
>         is this in db or a linear factor ?
>
I studied some of the theory to the EWMA(exponentially weighted moving
average),
so i think the <Look Ahead> can be easily deactivated by setting this to
0.
As i understand, the <Look Ahead> tries to wait until <n> inputs from
the
future are done and take this into the computing of the average ?

Because the description of the Peak Detector talks about <factor> i
think the
<TH Factor Rise/Fall> is a factor meaningfully greater than <1.0> ?
The alpha seems to weight the input values, which are be computed for
the
average, but i think it also results in a average computing of a longer
time ?
The alpha should be meaningfully  0 < alpha <= 1.0 .
Is this correct ?

A other Problem is that the Peak Detektor freezes the graphical
elements,
but i don't get any overrun/underrun.
I have got a decimation of 256 and a sample_rate of 250k.
Are there known Problems with the Peak Detector ?

Regards Markus
This topic is locked and can not be replied to.