A question about the calculation of peak flux density
Opened this issue · 4 comments
Hello,I have a question about the caculation of peak flux density,in fluence_flux_energy.py .some paper I read give the parameter use radiometer equation with a factor width** (-1/2) not the (tsamp)** (-1/2).so I want to konw ,Is my understanding wrong or is the code wrong?
flux = np.max(profile_burst * radiometer(tsamp, bw, 2, SEFD)) # peak flux density)
Hi, since we convert each time bin to flux density units in order to determine the peak flux density and fluence we use the sampling time or bin size as the 'width'. If you were to assume your burst is a boxcar with some height S/N then you could use the width of your boxcar/burst in the radiometer equation to estimate the fluence/average flux, however this is inaccurate since FRBs are not boxcar shaped and exhibit a more complex burst profile structure.
Hi, since we convert each time bin to flux density units in order to determine the peak flux density and fluence we use the sampling time or bin size as the 'width'. If you were to assume your burst is a boxcar with some height S/N then you could use the width of your boxcar/burst in the radiometer equation to estimate the fluence/average flux, however this is inaccurate since FRBs are not boxcar shaped and exhibit a more complex burst profile structure.
Thanks for your answer.my understanding is ,that other people use the SNR after match filter ,it will increase SNR by a factor （n）1/2 ,assume burst shape is a boxcar with some height S/N and width is ntsamp 。so the paper usually give the radiometer equation with a factor like （ntsamp） (-1/2) or (width)**(-1/2) and SNR after match filter to calculate a not very accurate S(peak) ，to remove this effect.
Hi, I think for a Gaussian shaped burst the flux at the peak should be close to the flux in a boxcar with FWHM width. I agree that the maximum is a bit crude as it will overestimate the flux for weak bursts. I don't think the peak flux is a very useful quantity for complex bursts as it also depends on the dispersion measure that is used. The fluence should be calculated via summation as it is done in the fluence_flux_energy_SEFDversion.py, but I don't know how others are doing this.
Finally are you (planning on) using these scripts? They are not very well documented because I didn't even know they are public, they are also quite specific to the AO and our data format, but I'd be happy to change that at some point.
Hi, I think for a Gaussian shaped burst the flux at the peak should be close to the flux in a boxcar with FWHM width. I agree that the maximum is a bit crude as it will overestimate the flux for weak bursts. I don't think the peak flux is a very useful quantity for complex bursts as it also depends on the dispersion measure that is used. The fluence should be calculated via summation as it is done in the fluence_flux_energy_SEFDversion.py, but I don't know how others are doing this.
Finally are you (planning on) using these scripts? They are not very well documented because I didn't even know they are public, they are also quite specific to the AO and our data format, but I'd be happy to change that at some point.
I plan to analyze some singlepulse data about magnetar ，but I don't how to calculate the S （peak），so I search the code on github and finally I find this project.