discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Regarding the .grc file.


From: Marcus Müller
Subject: Re: Regarding the .grc file.
Date: Tue, 5 Jul 2022 23:50:02 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.9.0

Hi Sumit,

no need to be so formal! Also, we're definitely not all "Sirs" on this list, there's quite a few ladies and others around here, and we're not just one guy. A simple "Hello!" suffices as greeting :) Welcome here!

Well, spectrum sensing is a big topic, and a lot of it is based on autocorrelation. For example, many parametric methods like ESPRIT and MUSIC work based on autocorrelation. You'll find a few of these estimators in the gr-specest OOT, which I, to my shame, haven't had time to look at lately.

Technically, whenever you use the Fourier transform of any piece of signal as an estimator for the spectrum, that's based on the Wiener-Khinchin theorem which says that for weak-sense stationary processes, the expectation of the Fourier transform and the Fourier transform on the autocorrelation function are equivalent in describing the power spectral density, the "Spectrum".

So, maybe you could tell us what exactly you're looking for. Because the way 
you phrased it,

> Spectrum Sensing Based on the Analysis of the Autocorrelation of Samples

is good, but means we're looking for estimators of the autocorrelation of the signal stochastic process from a limited observation of the process. You can do this in a lot of ways; but just like spectral estimation, correlation estimation will be divided into cases where you have a specific signal model (e.g., your signal is a sum of discrete sinusoids, or your signal is the output of an autoregressive model of some limited order), or cases where you cannot say much about the signal at all (in which case estimators often become very naive – and not rarely directly derived from the FFT of your signal, so you'd be estimating the autocorrelation from a spectral estimate, in order to find a spectral estimate).

So: did you have any specific spectral estimator in mind?

Best regards,
Marcus


On 05.07.22 22:09, Sumit Agrawal (P19EE207) wrote:
Dear Sir,

Could you please provide me the .grc file of development of GNU Radio Blocks for Spectrum Sensing Based on the Analysis of the Autocorrelation of Samples.

I also searched for the different detector blocks like energy detector, cyclostatiinary etc,  but I didn't find any block in gnu radio.

So help me regarding this.

----
Thanks & Regards,
Sumit Kumar Agrawal,
Ph.D. (Electrical Engineering)
Indian Institute of Technology,
Jodhpur, Rajasthan-342037
Mobile No. +91-8410957412



reply via email to

[Prev in Thread] Current Thread [Next in Thread]