Use ENBW instead of RBW in plot titles and calculations

[MichaelCotton]

I think we should start using Equivalent Noise Bandwidth (ENBW) instead of Resolution Bandwidth (RBW) throughout the project including in MSOD.

ENBW = 1.128_RBW (for swept frequency measurements w the E4440a) -and- ENBW = fw_delta_f (for DFT measurements), Where fw = window factor (e.g., fw = 2.2153 for GaussTop, 2.0044 for BlackmanHarris, …) delta_f = distance between adjacent bins in a DFT

[ranganathanm]

Please edit description to outline how this information would be used in Data Display (in the spectrogram screen, for example). If we only display ENBW, then how will a user know the size of the frequency bin (e.g. 180 khz) without doing significant (non-obvious) calculations? Also, please note that this will be part of Transfer Spec 14.

---

From: MichaelCotton notifications@github.com Sent: Friday, July 31, 2015 9:44 AM To: usnistgov/SpectrumBrowser Subject: [SpectrumBrowser] Use ENBW instead of RBW in plot titles and calculations (#206)

I think we should start using Equivalent Noise Bandwidth (ENBW) instead of Resolution Bandwidth (RBW) throughout the project including in MSOD.

ENBW = 1.128RBW (for swept frequency measurements w the E4440a) -and- ENBW = fwdelta_f (for DFT measurements), Where fw = window factor (e.g., fw = 2.2153 for GaussTop, 2.0044 for BlackmanHarris, ...) delta_f = distance between adjacent bins in a DFT

Reply to this email directly or view it on GitHubhttps://github.com/usnistgov/SpectrumBrowser/issues/206.

[ranganathanm]

Request to start a new issue for discussion of proposed new Transfer Spec and resultant changes to the data display.

---

From: Ranganathan, Mudumbai Sent: Friday, July 31, 2015 9:50 AM To: usnistgov/SpectrumBrowser; usnistgov/SpectrumBrowser Subject: Re: [SpectrumBrowser] Use ENBW instead of RBW in plot titles and calculations (#206)

Please edit description to outline how this information would be used in Data Display (in the spectrogram screen, for example). If we only display ENBW, then how will a user know the size of the frequency bin (e.g. 180 khz) without doing significant (non-obvious) calculations? Also, please note that this will be part of Transfer Spec 14.

---

From: MichaelCotton notifications@github.com Sent: Friday, July 31, 2015 9:44 AM To: usnistgov/SpectrumBrowser Subject: [SpectrumBrowser] Use ENBW instead of RBW in plot titles and calculations (#206)

I think we should start using Equivalent Noise Bandwidth (ENBW) instead of Resolution Bandwidth (RBW) throughout the project including in MSOD.

ENBW = 1.128RBW (for swept frequency measurements w the E4440a) -and- ENBW = fwdelta_f (for DFT measurements), Where fw = window factor (e.g., fw = 2.2153 for GaussTop, 2.0044 for BlackmanHarris, ...) delta_f = distance between adjacent bins in a DFT

Reply to this email directly or view it on GitHubhttps://github.com/usnistgov/SpectrumBrowser/issues/206.

[MichaelCotton]

Perhaps we need a delta_f =(fStop - fStart)/(nSteps-1) in the title in addition to ENBW. delta_f marks the separation between frequency bins. For LTE, delta_f should align with the resource block width, i.e., delta_f = 180 kHz. The ENBW describes the bin width around each bin center where signal power contributes. Ideally, ENBW would be 180 kHz, but there are constraints in hardware and DSP that make it impossible to hit it exactly. If ENBW is less than delta_f, then there are "little gaps" between the adjacent channels. When ENBW is greater than delta_f then there is "overlap".

[ranganathanm]

Yes, that would make it clear.