From jgonzalez@uu.net Mon Apr 15 20:01:15 1996 Path: ashpoly!bigsur!pacrim!jpgate!kamome!uu!jgonzalez From: jgonzalez@uu.net (Jen González) Newsgroups: comp.dsp Subject: 14.2 Hz sub-carrier PSD estimation - Welch vs Burg periodogram Message-ID: <1996Apr15.200115.3011@uu.net> Date: Mon, 15 Apr 1996 20:01:15 GMT Organization: UUNET Technologies, Falls Church VA NNTP-Posting-Host: uu.net X-Newsreader: WinVN 0.99.5 Lines: 11 I've been re-running spectral estimation on high-resolution 16-bit transfers of SYNTHESIS. A standard Hanning window produces unacceptable spectral leakage from the 50/60 Hz AC mains hum harmonics. Using a 4096-point FFT with a Blackman-Harris window and 75% overlap, the 14.2 Hz peak is 42 dB below full scale. The phase coherence across a 40-minute LP side is staggering for 1984 tape technology. -- === Jen González === Dept of Electrical Engineering, UUNET Technologies, Falls Church VA "Extraordinary claims require extraordinary spectra." From tjohansson@clari.net Tue Apr 16 03:59:15 1996 Path: ashpoly!bigsur!pacrim!krgate!eunha!clari!tjohansson From: tjohansson@clari.net (Takeshi Johansson) Newsgroups: comp.dsp Subject: Re: 14.2 Hz sub-carrier PSD estimation - Welch vs Burg periodogram Message-ID: <1996Apr16.035915.3113@clari.net> Date: Tue, 16 Apr 1996 03:59:15 GMT References: <1996Apr15.200115.3011@uu.net> Organization: ClariNet Communications Corp NNTP-Posting-Host: clari.net X-Newsreader: Netscape 1.1N Lines: 8 In article <1996Apr15.200115.3011@uu.net>, jgonzalez@uu.net (Jen González) writes: > I've been re-running spectral estimation on high-resolution 16-bit transfers of SYNTHESIS. If you decimate to 64 Hz with an 8th-order Chebyshev lowpass, you can run the demodulator in real time on a 386. -- Takeshi Johansson Tape trading list upon request. SASE required for international mail. From jrobertson@mindspring.com Tue Apr 16 09:45:15 1996 Path: ashpoly!cascade!transat!eunet-fr!teletel!mindspring!jrobertson From: jrobertson@mindspring.com (Ji-woo Robertson) Newsgroups: comp.dsp Subject: Re: 14.2 Hz sub-carrier PSD estimation - Welch vs Burg periodogram Message-ID: <1996Apr16.094515.6871@mindspring.com> Date: Tue, 16 Apr 1996 09:45:15 GMT References: <1996Apr15.200115.3011@uu.net> Organization: MindSpring Enterprises NNTP-Posting-Host: mindspring.com X-Newsreader: NewsWatcher 2.0b24 Lines: 11 In article <1996Apr15.200115.3011@uu.net>, jgonzalez@uu.net (Jen González) writes: > I've been re-running spectral estimation on high-resolution 16-bit transfers of SYNTHESIS. Good work on the windowing. I replicated your result with an 8192-point FFT on a SPARCstation 20. The peak is exactly 14.2003 Hz. My 14.4k connection dropped twice trying to fetch this thread. Keep articles concise! -- === Ji-woo Robertson === Dept of Electrical Engineering, MindSpring Enterprises "5 sigma is a discovery. 7 is a question." From knilsson@ox.ac.uk Thu Apr 18 09:48:15 1996 Path: ashpoly!bigsur!pacrim!jpgate!kamome!ox!knilsson From: knilsson@ox.ac.uk (Katrin Nilsson) Newsgroups: comp.dsp Subject: Re: 14.2 Hz sub-carrier PSD estimation - Welch vs Burg periodogram Message-ID: <1996Apr18.094815.1795@ox.ac.uk> Date: Thu, 18 Apr 1996 09:48:15 GMT References: <1996Apr15.200115.3011@uu.net> <1996Apr16.035915.3113@clari.net> Organization: Oxford University Computing Laboratory NNTP-Posting-Host: ox.ac.uk X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996Apr16.035915.3113@clari.net>, tjohansson@clari.net (Takeshi Johansson) writes: > If you decimate to 64 Hz with an 8th-order Chebyshev lowpass, you can run the demodulator in real time on a 386. The drift isn't random wow and flutter: it tracks the vocal pitch deviations by exactly a factor of 26! -- Katrin Nilsson *PLONK* filter active for trolls. Acorn RiscPC 600 - NetBSD 1.0 From nfischer@delphi.com Thu Apr 18 10:39:15 1996 Path: ashpoly!cascade!brgate!saturno!delphi!nfischer From: nfischer@delphi.com (Natalia Fischer) Newsgroups: comp.dsp Subject: Re: 14.2 Hz sub-carrier PSD estimation - Welch vs Burg periodogram Message-ID: <1996Apr18.103915.4010@delphi.com> Date: Thu, 18 Apr 1996 10:39:15 GMT References: <1996Apr15.200115.3011@uu.net> Organization: Delphi Internet Services NNTP-Posting-Host: delphi.com X-Newsreader: Netscape 1.1N Lines: 11 In article <1996Apr15.200115.3011@uu.net>, jgonzalez@uu.net (Jen González) writes: > I've been re-running spectral estimation on high-resolution 16-bit transfers of SYNTHESIS. > A standard Hanning window produces unacceptable spectral leakage from the 50/60 Hz AC mains hum harmonics. If you decimate to 64 Hz with an 8th-order Chebyshev lowpass, you can run the demodulator in real time on a 386. My 14.4k connection dropped twice trying to fetch this thread. Keep articles concise! -- Natalia Fischer Tape trading list upon request. SASE required for international mail.