I have made a program that calculate the maximum of the correlation spectral density and compare it with a threshold.this algorithm use the approach of the FFT accumulation method (FAM).the threshold is extracted from the formula of the probability of detection. the problem is that I don't know how to calculate the probability of detection after comparing the threshold with the maximum of the CSD.did someone have an idea about it ?
thank you for your response Sir but actually I want to use the algorithm of FAM not the energy detector unless I didn't understood what you exactly meant.
concerning now comparing the CSD to the noise power can you explain to me why should I do this because from what I got from the cyclostationnary spectrum sensor that it is based on the correlation of the main signal and then compare that result of correlation at a certain frequency (alpha) to a certain threshold to know if it contains the information signal at that frequency.
my main problem is that I didn't know how to compute the probability detection with many simulations or to be more clear I have tried to compute it with matlab and I am not sure whether it is correct or not.
Okey , I understood what you meant.so the cyclostationary algorithm is used as 2nd order so why in most articles it is mentioned as another algorithm of spectrum sensing like the energy detector and not mentioned as it depends to the cyclostationary algorithm. Can you take a look please at this article because i am a little bit confused.
thank you for your explanation and help Sir
if it is the case of a normalized lambda,what do you mean exactly by the cumulative density function of the statistics.because in the simulation, every time I compare the threshold to the average of my cyclotationnary spectral density (CSD) and then I increment a counter which supposed to be my POD after being divided by the number of simulations.
Can I sent my code via e-mail to discuss it much better ?
concerning now comparing the CSD to the noise power can you explain to me why should I do this because from what I got from the cyclostationnary spectrum sensor that it is based on the correlation of the main signal and then compare that result of correlation at a certain frequency (alpha) to a certain threshold to know if it contains the information signal at that frequency.
- Badges
- Science topic
- Similar topics
- Engineering
- Telecommunications Engineering