Steve Underwood wrote:> Clay wrote: > >> >> I remember when the PLL technique started to become popular back during >> the 1970s. Up until then people were using Ratio detectors and Foster >> Seeley Discriminators. But the PLL technique became inexpensive and it >> worked quite well. I figure it is the standard thing today in plain >> vanilla analog FM receivers. > > > More specifically, Signetics made some very usable early PLL chips. > First, all the stereo decoders started using them. All the FM decoders > soon followed. After that, every other consumer IC maker had to follow.I am afraid you are mistaken. The old good discriminator is used for audio demodulation. Afrer it, the stereo subcarrier is recovered by the PLL. I don't know of any consumer TV or AM/FM radio that is using the coherent demodulation. Vladimir Vassilevsky DSP and Mixed Signal Design Consultant http://www.abvolt.com
FM Demodulation
Started by ●January 17, 2007
Reply by ●January 19, 20072007-01-19
Reply by ●January 19, 20072007-01-19
Randy Yates wrote: [talking about PLL]> Also, if this helps improve performance for analog FM, wouldn't it do > the same for m-ary FSKs and variants? Would GMSK be improved with this > type of receiver?The bandwidth of the analog FM is usually much broader then the bandwidth of the audio signal. That's why the PLL can provide for some gain. For the digital signals, bandwidth ~ modulation rate, so nothing can be gained that way. The coherent demodulation, however, is the different story. Although for the simple signals like MSK or BPSK, the gain of the coherent demodulation is only about ~1dB at BER ~1e-2 and below. Vladimir Vassilevsky DSP and Mixed Signal Design Consultant http://www.abvolt.com
Reply by ●January 20, 20072007-01-20
Vladimir Vassilevsky <antispam_bogus@hotmail.com> wrote:>Yes, of course. Pretty much any good algorithm of the frequency >estimation can be used. It will make the FM demodulator which performs >better then the PLL. Also, the statistics of the audio signal can be >exploited... However the radical solution will be encoding the audio >into MP3 and transmitting it digitally. No analog solution can beat that.Only with strong signals, when you are getting to the edge, digital systems usually fail. The best FM receiver I ever have seen uses a DSP for demodulating the last IF; but in this radio FM is only a side effect, it is designed for completely digital voice transmission. In static environment digital and analog mode are comparable in maximum range, with better audi in digital mode, but in a dynamic environment digital just sucks, and only analog FM gives audio one can understand :)>Vladimir Vassilevsky > >DSP and Mixed Signal Design Consultant > >http://www.abvolt.comRalph.
Reply by ●January 20, 20072007-01-20
Vladimir Vassilevsky <antispam_bogus@hotmail.com> wrote:>Yes, of course. Pretty much any good algorithm of the frequency >estimation can be used. It will make the FM demodulator which performs >better then the PLL. Also, the statistics of the audio signal can be >exploited... However the radical solution will be encoding the audio >into MP3 and transmitting it digitally. No analog solution can beat that.Only with strong signals, when you are getting to the edge, digital systems usually fail. The best FM receiver I ever have seen uses a DSP for demodulating the last IF; but in this radio FM is only a side effect, it is designed for completely digital voice transmission. In static environment digital and analog mode are comparable in maximum range, with better audi in digital mode, but in a dynamic environment digital just sucks, and only analog FM gives audio one can understand :)>Vladimir Vassilevsky > >DSP and Mixed Signal Design Consultant > >http://www.abvolt.comRalph.
Reply by ●January 20, 20072007-01-20
"Ralph A. Schmid, DK5RAS" <dk5ras@dk5ras.de> writes:> Vladimir Vassilevsky <antispam_bogus@hotmail.com> wrote: > >>Yes, of course. Pretty much any good algorithm of the frequency >>estimation can be used. It will make the FM demodulator which performs >>better then the PLL. Also, the statistics of the audio signal can be >>exploited... However the radical solution will be encoding the audio >>into MP3 and transmitting it digitally. No analog solution can beat that. > > Only with strong signals, when you are getting to the edge, digital > systems usually fail. > > The best FM receiver I ever have seen uses a DSP for demodulating the > last IF; but in this radio FM is only a side effect, it is designed > for completely digital voice transmission. In static environment > digital and analog mode are comparable in maximum range, with better > audi in digital mode, but in a dynamic environment digital just sucks, > and only analog FM gives audio one can understand :)Them's fightin' words around here, Ralph! It's not nice to tease the natives. Seriously, what possesses you to make such claims? Is it just a joke? It doesn't seem like you're joking. Usually people make these types of claims when they don't understand digital. When an analog signal is properly dithered before converting to digital, the result is EXACTLY the same as an analog signal in noise - e.g., resolution increases with narrowing bandwidth. -- % Randy Yates % "Bird, on the wing, %% Fuquay-Varina, NC % goes floating by %%% 919-577-9882 % but there's a teardrop in his eye..." %%%% <yates@ieee.org> % 'One Summer Dream', *Face The Music*, ELO http://home.earthlink.net/~yatescr
Reply by ●January 20, 20072007-01-20
Ralph A. Schmid, DK5RAS wrote:>>However the radical solution will be encoding the audio >>into MP3 and transmitting it digitally. No analog solution can beat that. > > > Only with strong signals, when you are getting to the edge, digital > systems usually fail.This is a popular misconception. A fairly simple voiceband digital systems perform well at the SNR of 6dB, whereas the analog narowband FM requires at least 12dB. Both occupy the same bandwidth. BTW, that's why all cellular systems are digital.> The best FM receiver I ever have seen uses a DSP for demodulating the > last IF; but in this radio FM is only a side effect, it is designed > for completely digital voice transmission. In static environment > digital and analog mode are comparable in maximum range, with better > audi in digital mode, but in a dynamic environment digital just sucks, > and only analog FM gives audio one can understand :)Poor design and/or lousy DSP programming. Vladimir Vassilevsky DSP and Mixed Signal Design Consultant http://www.abvolt.com
Reply by ●January 20, 20072007-01-20
Randy Yates <yates@ieee.org> wrote:>Seriously, what possesses you to make such claims? Is it just a joke? >It doesn't seem like you're joking.Maybe I just dod not make the point clear.>Usually people make these types of claims when they don't understand >digital. When an analog signal is properly dithered before converting >to digital, the result is EXACTLY the same as an analog signal in >noise - e.g., resolution increases with narrowing bandwidth.I never said that was not true. The problem with digital radio communications is, in case of fading and at the edge of the coverage short dropouts render the signal almost unreadable in digital system, because of the voice codec and its compression algo. Some bits are missing, and large portions of the message can not be understood any more. An anloge transmission fades out, the noise increases, and short interruptions are not a problem, due to the redundancy of human speech. This redundancy is what the bandwidth efficient voice codecs use to reduce the bandwidth, and so this redundancy gets lost. It is a fact that for example a narrowband voice channel of 12.5KHz gives better coverage in analogue than in digital voice. Try it out!
Reply by ●January 20, 20072007-01-20
Ralph A. Schmid, DK5RAS wrote:> Jerry Avins <jya@ieee.org> wrote:>>The FM detector produces -- before deemphasis -- baseband audio (the sum >>channel), a difference signal modulated onto a nearly-40-MHz subcarrier,> 38 KHz, not MHz :)>>and a half-carrier-frequency pilot tone. Those three components are> Yes, 19 KHz.Or 31.46853 kHz and 15.73426 kHz, for stereo TV sound. -- glen
Reply by ●January 20, 20072007-01-20
Ralph A. Schmid, DK5RAS wrote:> The problem with digital radio > communications is, in case of fading and at the edge of the coverage > short dropouts render the signal almost unreadable in digital system, > because of the voice codec and its compression algo. Some bits are > missing, and large portions of the message can not be understood any > more. An anloge transmission fades out, the noise increases, and short > interruptions are not a problem, due to the redundancy of human > speech. This redundancy is what the bandwidth efficient voice codecs > use to reduce the bandwidth, and so this redundancy gets lost.This is the case when the SNR is much lower then it is intended to be by design. The digital will have a brickwall degradation, the analog will degrade more gracefully. Anyway the communication system is not supposed to operate in such conditions.> It is a fact that for example a narrowband voice channel of 12.5KHz > gives better coverage in analogue than in digital voice. Try it out.It is a fact that all modern systems are digital. Because they can be optimized to perform much better then the analog for any given quality. MELP at 1200 bps has somewhat similar quality to the NBFM with the bandwidth of 12.5kHz. For MELP, the SNR required will be about 7dB in the equvalent bandwidth of 1.2KHz. For analog, it will take at least 10dB in ~3KHz. So the gain of the digital is over 6dB. Vladimir Vassilevsky DSP and Mixed Signal Design Consultant http://www.abvolt.com
Reply by ●January 20, 20072007-01-20
glen herrmannsfeldt <gah@ugcs.caltech.edu> writes:> Ralph A. Schmid, DK5RAS wrote: > >> Jerry Avins <jya@ieee.org> wrote: > >>> The FM detector produces -- before deemphasis -- baseband audio >>> (the sum channel), a difference signal modulated onto a >>> nearly-40-MHz subcarrier, > >> 38 KHz, not MHz :) > >>> and a half-carrier-frequency pilot tone. Those three components are > >> Yes, 19 KHz. > > Or 31.46853 kHz and 15.73426 kHz, for stereo TV sound.According to the BTSC specification, the pilot tone is nominally 15734 Hz (+/- 2 Hz), not 15734.26 Hz. -- % Randy Yates % "I met someone who looks alot like you, %% Fuquay-Varina, NC % she does the things you do, %%% 919-577-9882 % but she is an IBM." %%%% <yates@ieee.org> % 'Yours Truly, 2095', *Time*, ELO http://home.earthlink.net/~yatescr
