FFT SW improvements

Forum for discussing PicoScope version 6 (non-automotive version)
Post Reply
kubik
Newbie
Posts: 0
Joined: Wed Nov 25, 2015 5:45 pm

FFT SW improvements

Post by kubik »

We have recently acquired a PicoScope 4262 to measure the RF signal Phase Noise.
To do so, we demodulate the RF signal using a clean LO signal and we use the 4262 to measure the noise power density in the baseband signal.
The systems work well but some calibrations and corrections are needed and we must export the data and process it to calculate the real value.
The PicoScope 6 could easily do the job with some minors additions to the software, which are:
-Ta add a new log unit in the FTT mode, showing the power density converted to nominal 1Hz Bandwidth. To do so you just need to divide the current result by the resolution bandwidth (which should be the total bandwidth divided by the number of spectrum bins).
-To include the possibility of adding an offset to the Y axis in dB. It can be already done in the current SW revision changing the reference in the "Arbitrary dB" mode but you need do calculate it.
-To be able to change the Log unit shown in the y axis, when selecting the "Arbitrary dB" mode (I would like to see dBc/Hz).
Another feature which will be also interesting would be to calculate and represent the addition of a FTT log signal with another one previously stored in the memory. However I understand this is more complex to program.

This features will be usefull for other applications like sound and vibration analisys.

Hitesh

Re: FFT SW improvements

Post by Hitesh »

Hi kubik,

Thank you for the suggestions which have been passed through to the Development Team.

Regards,

springburo
Newbie
Posts: 0
Joined: Mon Jun 08, 2015 5:24 pm

spectrum resolution bandwidth

Post by springburo »

Is there any way to estimate the effective resolution bandwidth in the spectrum analyser mode to make it semi compatible to standard spectrum anaylsers?

Gerry
PICO STAFF
PICO STAFF
Posts: 1145
Joined: Mon Aug 11, 2014 11:14 am

Re: FFT SW improvements

Post by Gerry »

Hi Springburo,

A standard spectrum analyzer, that uses a sweeping bandpass filter as the 'window' for detecting signals, enables the bandwidth of the filter to be changed via the front panel of the analyzer. It's bandwidth is also known as the 'Resolution Bandwidth' (RBW) of the analyzer, and although it affects the Signal-to-Noise Ratio (SNR) the RBW specification does not incorporate the SNR value. As you would like to know the parameter in a Spectrum plot of PicoScope 6 that is compatible with the RBW of standard spectrum analyzers, I suspect that you mean the 'compatible resolution bandwidth', rather than the 'effective resolution bandwidth' (which is a different parameter, that DOES incorporate the SNR - see below).
In PicoScope 6 you can display all of the key parameters relating to a mode of display by just turning on the 'View Properties' window (under the 'Views' menu). In Spectrum mode this will display the 'Bin width', which is the frequency resolution (or compatible resolution bandwidth) of the FFT being plotted, irrespective of noise and distortion, and I believe is the value you are looking for. The frequency resolution will change depending upon the number of spectrum bins that you select in the FFT options window.

Strictly speaking, Effective Resolution Bandwidth (ERBW) is the signal frequency range over which the spectrum analyzer can resolve signals of equal amplitude, before noise & distortion corrupts the comparison, i.e. where the difference between 2 signal frequencies being analyzed is greater in frequency than the point on an SNR spectrum plot where the noise & distortion corrupts the comparison (-6dB point for a voltage amplitude spectrum).
For a spectrum analyzer that incorporates an Analog-to-Digital Converter (ADC), the ERBW is the signal frequency range over which the Effective Number Of Bits (ENOB) is still true, i.e. the frequency range up to the point where ENOB drops by 1/2 a bit (ENOB is the voltage resolution over which the ADC can resolve 2 signals of different amplitude, before noise & distortion corrupts the comparison).

Regards,

Gerry
Gerry
Technical Specialist

Post Reply