Home |
Search |
Today's Posts |
#1
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
My satellite TV service is now supplying XM sound. Sounds "pretty"
good for most pop music and some classical, except for opera. My intent is to see whether it makes sense to subscribe to the service directly. Here's the scenario. I record the XM sound...opera... on my PC, but when I replay it , it sounds like hell. I am pretty picky....I can sometimes pick up on differences between ripping or sampling at 256K vs. 320Kb. A local FM station does an excellent job with classical...comparison of ripped tracks vs recorded tracks just doesn't show any deviations to my ear. But, these XM broadcast for inherently high dynamic range music like opera are a true litmus test. When I use a small utility from Nero...WaveEditor...I can see the frequency and amplitude of the recorded sound. Pretty funky compared to CD sound....clipped signals and frequency truncations and artificial boost at higher frequencies. Plays havoc with opera. Other stuff seems to "sound" reasonable...who can tell with the Stones??? Went to AOL/XM streams and recorded some opera at my dialup speed of 44kb...damned slow but these new compression algorithms are amazing. The sound was still not good enough to capture classical, especially opera, but looking at the signals showed that the signals were not as garbaged up as those from my satellite TV source. Stands to reason that some extra truncations, compressions, and filtering are occurring on the satellite beam. Now, here's the issue: if the XM or Sirius signals are that manipulated when directly transmitted, then, again, the opera is likely to be the most damaged. I don't want to spend the time or money on the service and hardware unless I am truly getting CD quality sound. There is a wrinkle. My satellite TV has a small amount of objectionable ghosting...three or more very small ghosts... that they cannot figure out...I think its back or forward refelction/deflections from a nearby tree that is ... for all practical purposes...interferring with the clear line of sight to the satellite. If it is in the video, it may be in the audio, too, but I don't have a way of sampling the video. Anybody have some experience they could share?? On any aspect of this series of experiments?? Thanks. Henry |
#2
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
On Mon, 28 Nov 2005 18:32:46 GMT, NewGuy wrote:
My satellite TV service is now supplying XM sound. Sounds "pretty" good for most pop music and some classical, except for opera. My intent is to see whether it makes sense to subscribe to the service directly. Here's the scenario. I record the XM sound...opera... on my PC, but when I replay it , it sounds like hell. I am pretty picky....I can sometimes pick up on differences between ripping or sampling at 256K vs. 320Kb. A local FM station does an excellent job with classical...comparison of ripped tracks vs recorded tracks just doesn't show any deviations to my ear. But, these XM broadcast for inherently high dynamic range music like opera are a true litmus test. When I use a small utility from Nero...WaveEditor...I can see the frequency and amplitude of the recorded sound. Pretty funky compared to CD sound....clipped signals and frequency truncations and artificial boost at higher frequencies. Plays havoc with opera. Other stuff seems to "sound" reasonable...who can tell with the Stones??? Went to AOL/XM streams and recorded some opera at my dialup speed of 44kb...damned slow but these new compression algorithms are amazing. The sound was still not good enough to capture classical, especially opera, but looking at the signals showed that the signals were not as garbaged up as those from my satellite TV source. Stands to reason that some extra truncations, compressions, and filtering are occurring on the satellite beam. Now, here's the issue: if the XM or Sirius signals are that manipulated when directly transmitted, then, again, the opera is likely to be the most damaged. I don't want to spend the time or money on the service and hardware unless I am truly getting CD quality sound. There is a wrinkle. My satellite TV has a small amount of objectionable ghosting...three or more very small ghosts... that they cannot figure out...I think its back or forward refelction/deflections from a nearby tree that is ... for all practical purposes...interferring with the clear line of sight to the satellite. If it is in the video, it may be in the audio, too, but I don't have a way of sampling the video. Anybody have some experience they could share?? On any aspect of this series of experiments?? Thanks. Henry Is this a digital satellite service? Ghosting is not possible with digital, so I presume it is analogue, although I thought all new users were digital - oh well. The source of the ghost image can be estimated by measuring how far across the screen from the original image it is. I have to say, though, that ghosting with satellite is unusual because the dish beam is so narrow. If you are on a communal distribution system, it is more likely to be reflections from unterminated cable runs. Talk to the installer about this. d Pearce Consulting http://www.pearce.uk.com |
#4
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
On Mon, 28 Nov 2005 19:16:07 GMT, (GregS) wrote:
Is this a digital satellite service? Ghosting is not possible with digital, so I presume it is analogue, although I thought all new users were digital - oh well. I think digital processing can produce ghosts. Actually, a pre and a post ghost. I see in on analog transmissions of digitally processed video. I see many digital artifacts on many TV transmissions, while either viewing digital TV or digitally processed TV. Having both is a bigger distraction. I not talking HDTV, allthough I see different things there also. greg No, it can't produce ghosts - there simply isn't a mechanism. What does happen, though, particularly on channels that are starved of bandwidth is that artificial edge-sharpening techniques are sued that produce a sort of nasty white outline to picture features. I suppose these might be construed as ghosts. HDTV is producing some huge problems of its own. Having to compress a pretty huge original stream into a normal channel makes for some pretty horrid MPEG artifacts at times. I'm not saying that the picture would necessarily look better at standard definitions, but that is certainly the case on occasions. d Pearce Consulting http://www.pearce.uk.com |
#5
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
In article , (Don Pearce) wrote:
On Mon, 28 Nov 2005 19:16:07 GMT, (GregS) wrote: Is this a digital satellite service? Ghosting is not possible with digital, so I presume it is analogue, although I thought all new users were digital - oh well. I think digital processing can produce ghosts. Actually, a pre and a post ghost. I see in on analog transmissions of digitally processed video. I see many digital artifacts on many TV transmissions, while either viewing digital TV or digitally processed TV. Having both is a bigger distraction. I not talking HDTV, allthough I see different things there also. greg No, it can't produce ghosts - there simply isn't a mechanism. What does happen, though, particularly on channels that are starved of bandwidth is that artificial edge-sharpening techniques are sued that produce a sort of nasty white outline to picture features. I suppose these might be construed as ghosts. I've had a chance to look at some newer formats and some new TV's lately. I have seen the white outlines, but on my old Toshiba 36 inch direct view, I have seen the worst problems viewed, many on sports broadcasts. The grass looks like vertical lines, and so does the background crowd., and I swear on this TV, I have seen pre and post outlines. It may be the Toshiba is intensifying those problems, because I have not seen it anywhere near as bad. I was sitting in front of large HDTV LCD and plasma screens on Thanksgiving. I find much of my time looking at all the problems, rather than just enjoying the broadcast. greg HDTV is producing some huge problems of its own. Having to compress a pretty huge original stream into a normal channel makes for some pretty horrid MPEG artifacts at times. I'm not saying that the picture would necessarily look better at standard definitions, but that is certainly the case on occasions. d Pearce Consulting http://www.pearce.uk.com |
#6
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
On Mon, 28 Nov 2005 19:48:38 GMT, (GregS) wrote:
I've had a chance to look at some newer formats and some new TV's lately. I have seen the white outlines, but on my old Toshiba 36 inch direct view, I have seen the worst problems viewed, many on sports broadcasts. The grass looks like vertical lines, and so does the background crowd., and I swear on this TV, I have seen pre and post outlines. It may be the Toshiba is intensifying those problems, because I have not seen it anywhere near as bad. I was sitting in front of large HDTV LCD and plasma screens on Thanksgiving. I find much of my time looking at all the problems, rather than just enjoying the broadcast. greg Yup - MPEG artifacts, I'm afraid. Get used to them :-( d Pearce Consulting http://www.pearce.uk.com |
#7
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
Wow! You folks jumped on the ghost issue...and I only threw it in as
an afterthought. Frankly, the issue of digital vs analog is confusing. Any digital signal can be "synthisized" from an infinite series of analog sine waves etc etc. In a way, even digital signals are inherently analog...it is how the information is picked off the signal. I had hoped never and try and re-acquaint myself with this stuff. My former cable company began sending digial signals that actually improved the picture on my analog receiver...but it was not HDTV...which is more, denser, info but not all that different. Like an MS and PhD... More of the Same and Piled Higher and Deeper. Old timers know ghosting as a multipath phenomenon. The Satellite guy actually put in a fresh antenna detector unit and directly hooked a cable to it and dragged it into my house with no connections other than at my TV. Since I have two other TVs using in house wiring that show exactly the same ghosting effect, and if the satellite signal is "ghost" free, then the problem is in what my receiver is detecting. These are faint ghosts that smear out small objects like on-screen lettering or football player in a wide-angle field shot...totally blurred out. When I called to complain the first time, nobody knew what I was talking about when I said a ghost image....it is apparently that rare. Now, to beat the horse to death, anything that is limiting the bandwidth adds artifacts called ringing, and scanning causes aliasing....ringing is typically at higher harmonics that are filered out. Aliasing, a scanning artifact, shows up in TVs when a fine-patterned image...coat patterns or tie patterns on clothing seem to "run" and not be stable. Digital signals should get rid of this but analog sets with their analog signal scanning just jitter back and forth giving the impressing of flowing patterns. Its the edges and the widths in the scan direction. Egads, what are you making me remember??? What I want is to know if the XM signals that I should be able to recieve are inherently CD quality...which is how they are billed. Henry On Mon, 28 Nov 2005 19:27:39 GMT, (Don Pearce) wrote: On Mon, 28 Nov 2005 19:16:07 GMT, (GregS) wrote: Is this a digital satellite service? Ghosting is not possible with digital, so I presume it is analogue, although I thought all new users were digital - oh well. I think digital processing can produce ghosts. Actually, a pre and a post ghost. I see in on analog transmissions of digitally processed video. I see many digital artifacts on many TV transmissions, while either viewing digital TV or digitally processed TV. Having both is a bigger distraction. I not talking HDTV, allthough I see different things there also. greg No, it can't produce ghosts - there simply isn't a mechanism. What does happen, though, particularly on channels that are starved of bandwidth is that artificial edge-sharpening techniques are sued that produce a sort of nasty white outline to picture features. I suppose these might be construed as ghosts. HDTV is producing some huge problems of its own. Having to compress a pretty huge original stream into a normal channel makes for some pretty horrid MPEG artifacts at times. I'm not saying that the picture would necessarily look better at standard definitions, but that is certainly the case on occasions. d Pearce Consulting http://www.pearce.uk.com |
#8
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
On Mon, 28 Nov 2005 19:54:52 GMT, NewGuy wrote:
What I want is to know if the XM signals that I should be able to recieve are inherently CD quality...which is how they are billed. Sorry - went for the interesting bit. Very unlikely. They will be transmitting in some sort of compressed format, which will be sub-CD standard. Happily, it is also pretty certain that they will be keeping the bit rate high enough that you shouldn't hear the difference. d Pearce Consulting http://www.pearce.uk.com |
#9
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
"NewGuy" wrote in message
news My satellite TV service is now supplying XM sound. Sounds "pretty" good for most pop music and some classical, except for opera. My intent is to see whether it makes sense to subscribe to the service directly. Here's the scenario. I record the XM sound...opera... on my PC, but when I replay it , it sounds like hell. I am pretty picky....I can sometimes pick up on differences between ripping or sampling at 256K vs. 320Kb. XM and Sirius have been getting careful scrutiny from a group of experienced audio engineers I'll refer to as "The Detroit Audio Mafia" ever since they first came out. These guys have been listening to XM and Sirius since the only receivers around were prototypes for the big 3 auto makers to try out. After all, this *is* Detroit and the reason XM and Sirius exist primarily for listening in cars. "The word" on the street in Detroit from "The Detroit Audio Mafia" is that the sound quality of XM and Sirius has been flushed down the porcelain convenience, as more and more concurrent services were added. This makes perfect sense, of course. As the bitrate per service goes down, the sound quality *must* go down. At this time XM and Sirius are generally agreed-upon by "The Detroit Audio Mafia" to be generally *unacceptable* for serious listening, even in a moving car. XM and Sirius weren't always this way, but that was then and this is now. |
#10
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
Wow! Well, I asked for it. I use to work in the spread spectrum world
and my hope was that technology would eventually render all this bandwidth angst to the junk yard. True spread spectrum, like the miliary use with very expensive...and in the past...GaAs based processors, renders bandwidth an obsolete concern....in so far as specrual allocations and bandwidth has the same narrow definition as in the past. With SS, you actually spread the signal over such a large bandwidth that any signal looks like noise...non-detectible noise against all other signals cluttering any given bandwidth allocation....with enough spectrum, signal reception, multipath, power, and signal fidelity and noise in the traditional S/N become of historical interest. Sadly, the technology in cheap silicon still looks deficient and too expensive. However, it won't be long. WiMax is suppose to be this technology, but I am not sure. As I look at the way WiMax is being implemented, I suspect they are not using true spread specturm. I also appreciate the comments on the deficiencies of modern TVs to the various signal formats. My sets range from high end to crap and they each see the same signal from the satellite...so it is the satellite signal modulated by my live oak tree foliage. Looks like XM will be useful for non-classical recording...for which it seems to excel. I guess until further notice, I won't bother with XM or Sirius for anything other than convenience. Thanks for the info. Henry On Mon, 28 Nov 2005 16:57:24 -0500, "Arny Krueger" wrote: "NewGuy" wrote in message news My satellite TV service is now supplying XM sound. Sounds "pretty" good for most pop music and some classical, except for opera. My intent is to see whether it makes sense to subscribe to the service directly. Here's the scenario. I record the XM sound...opera... on my PC, but when I replay it , it sounds like hell. I am pretty picky....I can sometimes pick up on differences between ripping or sampling at 256K vs. 320Kb. XM and Sirius have been getting careful scrutiny from a group of experienced audio engineers I'll refer to as "The Detroit Audio Mafia" ever since they first came out. These guys have been listening to XM and Sirius since the only receivers around were prototypes for the big 3 auto makers to try out. After all, this *is* Detroit and the reason XM and Sirius exist primarily for listening in cars. "The word" on the street in Detroit from "The Detroit Audio Mafia" is that the sound quality of XM and Sirius has been flushed down the porcelain convenience, as more and more concurrent services were added. This makes perfect sense, of course. As the bitrate per service goes down, the sound quality *must* go down. At this time XM and Sirius are generally agreed-upon by "The Detroit Audio Mafia" to be generally *unacceptable* for serious listening, even in a moving car. XM and Sirius weren't always this way, but that was then and this is now. |
#11
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
In article ,
NewGuy wrote: Wow! Well, I asked for it. I use to work in the spread spectrum world and my hope was that technology would eventually render all this bandwidth angst to the junk yard. True spread spectrum, like the miliary use with very expensive...and in the past...GaAs based processors, renders bandwidth an obsolete concern....in so far as specrual allocations and bandwidth has the same narrow definition as in the past. With SS, you actually spread the signal over such a large bandwidth that any signal looks like noise...non-detectible noise against all other signals cluttering any given bandwidth allocation....with enough spectrum, signal reception, multipath, power, and signal fidelity and noise in the traditional S/N become of historical interest. Sadly, the technology in cheap silicon still looks deficient and too expensive. However, it won't be long. T'ain't quite so simple, I'm afraid. You can't get something for nothing. Spread-spectrum comes at a price... it does raise the noise level (as seen by other users of that part of the spectrum) throughout whatever transmission bandwidth it uses. Although the increase in noise level created by one SS transmitter may be small and difficult to detect, put enough of 'em together in one area and you've got a large amount of added noise energy to deal with. This added noise can degrade both SS and non-SS uses of the frequencies in question. The amateur radio community has been dealing with the results of this for some time. I've been told that in many urban areas, it's no longer possible to do weak-signal work in the 2.4 GHz ham bands, due to the presence of many thousands of low-powered 802.11b/802.11g direct-sequence spread spectrum radios. WiMax is suppose to be this technology, but I am not sure. As I look at the way WiMax is being implemented, I suspect they are not using true spread specturm. My understanding is that WiMax uses OFDM (orthogonal frequency division multiplexing), a system with large numbers of closely-spaced carriers. It's similar to 802.11g in that respect. There's some new UWB (ultra-wide-band) work being done, which shows promise for very high bandwidths over very short distances (ideal for in-home video and that sort of thing). One UWB group is proposing or using OFDM, while another is using high-speed-impulse modulations which I suspect are closer to the "true spread spectrum" stuff you worked on in the military. -- Dave Platt AE6EO Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior I do _not_ wish to receive unsolicited commercial email, and I will boycott any company which has the gall to send me such ads! |
#12
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
Your points are well taken.....you are right about wimax...I had
forgotten that technology in my zeal to promote SS ;). On the other hand, true SS is literally noise....if it is cluttering the frequency bands to other more commercial users, then it is a not the real thing. Consider noise: FM...lightning strikes wipe out AM but FM barely hears a click. The noise immunity in SS is even greater. The beauty of SS is that a well-designed receiver picks the signal out of literally the noise...very low power using optimal filtering and signal tracking. My guess is that WiFi is a cheap imitation and therein lies the problem. Commercial silicon is still not up to the task...but it will be. As for WiMax, it will not be worth the efforts until a true SS implementation at low prices can be fielded. As for something for nothing, you are paying for the capability through use of very wide bandwidth and very expensive transmitters and signal processing....though with todays production costs for semiconductor signal processors, I can't image the cost would remain an issue...it is still likely a performance hit due to marginal capabilities at affordable commercial prices. Be that as it may...this discussion is making my head hurt...XM is not in my future. Thanks for the dialog. Henry On Tue, 29 Nov 2005 01:47:47 -0000, (Dave Platt) wrote: In article , NewGuy wrote: Wow! Well, I asked for it. I use to work in the spread spectrum world and my hope was that technology would eventually render all this bandwidth angst to the junk yard. True spread spectrum, like the miliary use with very expensive...and in the past...GaAs based processors, renders bandwidth an obsolete concern....in so far as specrual allocations and bandwidth has the same narrow definition as in the past. With SS, you actually spread the signal over such a large bandwidth that any signal looks like noise...non-detectible noise against all other signals cluttering any given bandwidth allocation....with enough spectrum, signal reception, multipath, power, and signal fidelity and noise in the traditional S/N become of historical interest. Sadly, the technology in cheap silicon still looks deficient and too expensive. However, it won't be long. T'ain't quite so simple, I'm afraid. You can't get something for nothing. Spread-spectrum comes at a price... it does raise the noise level (as seen by other users of that part of the spectrum) throughout whatever transmission bandwidth it uses. Although the increase in noise level created by one SS transmitter may be small and difficult to detect, put enough of 'em together in one area and you've got a large amount of added noise energy to deal with. This added noise can degrade both SS and non-SS uses of the frequencies in question. The amateur radio community has been dealing with the results of this for some time. I've been told that in many urban areas, it's no longer possible to do weak-signal work in the 2.4 GHz ham bands, due to the presence of many thousands of low-powered 802.11b/802.11g direct-sequence spread spectrum radios. WiMax is suppose to be this technology, but I am not sure. As I look at the way WiMax is being implemented, I suspect they are not using true spread specturm. My understanding is that WiMax uses OFDM (orthogonal frequency division multiplexing), a system with large numbers of closely-spaced carriers. It's similar to 802.11g in that respect. There's some new UWB (ultra-wide-band) work being done, which shows promise for very high bandwidths over very short distances (ideal for in-home video and that sort of thing). One UWB group is proposing or using OFDM, while another is using high-speed-impulse modulations which I suspect are closer to the "true spread spectrum" stuff you worked on in the military. |
#13
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
In article , NewGuy wrote:
Your points are well taken.....you are right about wimax...I had forgotten that technology in my zeal to promote SS ;). On the other hand, true SS is literally noise....if it is cluttering the frequency bands to other more commercial users, then it is a not the real thing. Consider noise: FM...lightning strikes wipe out AM but FM barely hears a click. The noise immunity in SS is even greater. The beauty of SS is that a well-designed receiver picks the signal out of literally the noise...very low power using optimal filtering and signal tracking. My guess is that WiFi is a cheap imitation and therein lies the problem. Commercial silicon is still not up to the task...but it will be. I don't know all these formats. I have a spread spectrum portable phone which is spread sprectrum, but I found it fairy simplified just kind of switching freqs, at a prescribed interval. Then, I thought thats what spread sprectrum was. Glad to hear something else exists. I used to run some equipment that was modulated by a digital psedo random code, which appeared as noise around the baseband signal. That was the manned spacecraft 2275 freqs, and the code was used to aquire range data. They made the code long enough to do the moon and back. greg |
#14
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
In article ,
GregS wrote: I don't know all these formats. I have a spread spectrum portable phone which is spread sprectrum, but I found it fairy simplified just kind of switching freqs, at a prescribed interval. Then, I thought thats what spread sprectrum was. Glad to hear something else exists. I'm aware of at least four types of modulation which fall under the general category of "spread spectrum". One of them is FH (frequency-hopping), which is the sort that your cordless phone uses. It uses a single, narrow-band carrier, which is switched or "hopped" between multiple frequencies on a periodic and pre-programmed basis. The idea with this form of SS is to keep the carrier from sitting for any great length of time on a single frequency where it might be interfered with, or interfere with other users of the channels. Dropouts in transmission can and do occur, when the hopper lands on a frequency that's in use by another set of stations, but it's usually limited to one or two "hop times" (e.g. a fraction of a second of voice outage on a portable phone). Several different FH users can be operating at once, as long as they're using different frequency-hopping patterns (or the same pattern but offset in time). Bad things happen if two or more users happen to pick the same hopping pattern and fall into synchronization - at that point they interfere with one another 100% of the time! This form of SS was, I believe, the first one deliberately used for security and robustness against interception. It was used during World War II, and was first patented by an actress (Hedy Lamarr) and a composer (George Antheil)! Reportedly, their idea was inspired by consideration of a player-piano scroll. A second form of spread spectrum is "direct sequence". In this system, a single carrier frequency is chosen, and is then modulated (multiplied, or just turned on and off) at very high speeds by a binary "spreading sequence". This modulation has the effect of spreading the carrier's energy across a wide bandwidth, in a way which causes the energy in each portion of the bandwidth to closely resemble random noise. A receiving station tunes to the center frequency of the range, multiplies the RF signal by the same spreading sequence, and gets out a clean sequence of pulses which reproduce the original data. Narrowband interference in the bandwidth (e.g. a non-spread- spectrum signal) tends to be "smeared out" by the receiver's spreading sequence multiplier and "looks" to the receiver like white noise... degrading the SS receiver's signal-to-noise ratio somewhat but not interfering very strongly. Similarly, SS receivers which use different spreading sequences, or narrow-band receivers, simply "see" the white-noise-like signal (which degrades their signal-to-noise ratio somewhat but is not a strong interferer). 802.11b wireless uses direct-sequence spread spectrum. A third form is OFDM (orthogonal frequency division multiplexing), which transmits a whole bunch of carrier signals at closely-spaced frequencies, each one of which is modulated at a relatively low symbol rate and carries some portion of the payload information. OFDM-like multicarrier modulations have been in use for quite a few years... I believe they were developed at Bell Labs back in the 1960s, were used in the classic Telebit Trailblazer modems (the first to achieve reliable, inexpensive 9600-bit/second-and-faster transmission over voice-grade phone lines), and are the basis of most ADSL network hookups today (DMT, or "discrete multitone" modems are widely used for this). Individual carriers in an OFDM system can be degraded or wiped out by narrow-band interference. OFDM systems will often react to this in an adaptive fashion (e.g. the sender and receiver figure out which carrier frequencies work best at any given point in time), and/or include forward error correction coding of one sort or another to recover those bits which were transmitted on carriers that were interfered with. WiMax uses OFDM, as does one of the upcoming UWB (ultra-wide-band) semi-standards. OFDM has an advantage over direct-sequence spread spectrum, in that it's more resistant to multipath distortion (signal cancellation due to the signal arriving via two or more routes which differ in length). The fourth method is impulse-based ultra-wide-band, which transmits signals as a series of very closely spaced pulses with very rapid rise-times (and thus an extremely wide range of frequencies). The information can be conveyed by the timing, width, spacing, and/or amplitude of the pulses. All of these systems can be used to reduce the degree to which the SS system interferes with other users of the frequency band, and to make the SS transmission itself less sensitive to narrowband noise. None of them is, or can be, idea. The signal transmitted by a SS transmitter will, at best, look like purely random "white noise" to other users of that frequency band, and will necessarily degrade the signal-to-noise ratio seen by those other users. Similarly, other users of a frequency (whether narrow-, wide-, or spread-spectrum) will at best look like white noise to the receiver in a spread spectrum system, and will degrade that receiver's SNR to some extent. The amount of data any system can transmit depends on the signal bandwidth, and the signal-to-noise ratio: it's set by the Shannon limit. Spread-spectrum systems can pump a lot of data through because their bandwidth is quite wide, and because they can tend to make interfering signals look like random noise... but they aren't immune to the Shannon limit themselves (their error rate increases as the noise level rises) nor can they avoid degrading to some extent the useful bandwidth of other users of their frequency band. Modern error-correction coding techniques (specifically, turbo codes) seem to be able to deliver a useful bandwidth which is very close indeed to the Shannon limit. I used to run some equipment that was modulated by a digital psedo random code, which appeared as noise around the baseband signal. That was the manned spacecraft 2275 freqs, and the code was used to aquire range data. They made the code long enough to do the moon and back. Sounds like a direct-sequence system, presumably using a Barker code or some similar self-synchronizing pseudorandom code. -- Dave Platt AE6EO Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior I do _not_ wish to receive unsolicited commercial email, and I will boycott any company which has the gall to send me such ads! |
#15
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
None of them is, or can be, idea. The signal transmitted by a SS transmitter will, at best, look like purely random "white noise" to other users of that frequency band, and will necessarily degrade the signal-to-noise ratio seen by those other users. Similarly, other users of a frequency (whether narrow-, wide-, or spread-spectrum) will at best look like white noise to the receiver in a spread spectrum system, and will degrade that receiver's SNR to some extent. Dave, great post.. I have one little nit to pick..... It is my understanding that 2 or more SS users that are synchronized to each other and are using __orthogonal__ spreading codes will not create and interference to each other, not even noise.... To the OP If you have 3 TVs hooked up, your ghosts are probably from all the wiring..... try it with one TV directly connected ot the sat rcv. And also to the OP... I assumme you are talking about Direct TV which now has XM radio....how did you like the previous Music Choice Service... Mark |
#16
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
On 30 Nov 2005 18:12:53 -0800, "Mark" wrote:
None of them is, or can be, idea. The signal transmitted by a SS transmitter will, at best, look like purely random "white noise" to other users of that frequency band, and will necessarily degrade the signal-to-noise ratio seen by those other users. Similarly, other users of a frequency (whether narrow-, wide-, or spread-spectrum) will at best look like white noise to the receiver in a spread spectrum system, and will degrade that receiver's SNR to some extent. Dave, great post.. I have one little nit to pick..... It is my understanding that 2 or more SS users that are synchronized to each other and are using __orthogonal__ spreading codes will not create and interference to each other, not even noise.... That is true to an extent. First, it only holds in single path conditions like a cable connection. In the real world radio waves arrive by multiple paths of different lengths, so destroying the coherence of the code edges. Otherwise orthogonal codes then look increasingly like plain old noise. Secondly, whatever the coding, there is still power there, which will cause blocking of the front end of the receiver. There is a big fallacy concerning spread spectrum that it somehow makes really efficient use of radio spectrum because if you spread far enough it "disappears". This is of course nonsense, it doesn't matter whether you use TDMA, FDMA, OFDM, SS or whatever. You still need the same bandwidth * SNR product to shift a given amount of data. Shannon will not be denied. d Pearce Consulting http://www.pearce.uk.com |
#17
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
Don,
I agree with you that SS does not give something for nothing. SS does NOT allow you to cram more data into a given BW... SS does NOT make more efficient use of the spectrum I never said it does... It is useful for cell phones becasue it is a good way for multiple users to share spectrum but the overall agreggate data throughput is not increased by the use of SS.. I agree with you! But my understadning is that synchronized SS users do not interfere with each other when using orthogonal spreading codes. Let me clarify two things... 1) There is a limit to the number of orthogonal codes so there is a limit to the number of users...you can NOT pack more and more aggregate data into the BW then you could without SS. 2) sychronized means that the SS spreading signals are synchronized ____when they arrive at the Rx____ the system must take the prop delay into account so that the signals are synched at the Tx. Mark |
#18
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
On 1 Dec 2005 07:09:44 -0800, "Mark" wrote:
But my understadning is that synchronized SS users do not interfere with each other when using orthogonal spreading codes. Let me clarify two things... 1) There is a limit to the number of orthogonal codes so there is a limit to the number of users...you can NOT pack more and more aggregate data into the BW then you could without SS. 2) sychronized means that the SS spreading signals are synchronized ____when they arrive at the Rx____ the system must take the prop delay into account so that the signals are synched at the Tx. Mark You are right. But as I said, in a real system, the degree of synchronization is way below optimum because of multipath. The signal from any phone to the base station travels by many paths, of unequal length and hence delay. This destroys the synchronization. You can only take propagation delay into account for one of those paths. As for the other direction, phones receive signals not only from their own base station, but also from those nearby. The base stations do their best to synchronize, but unless you are stood equidistant from all nearby base stations, you don't get synchronized interference. So although you do get a certain amount of benefit from RW (orthogonal) coding, it is in reality way beyond what theory would give you. d Pearce Consulting http://www.pearce.uk.com |
#19
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
On Thu, 01 Dec 2005 15:16:08 GMT, (Don Pearce)
wrote: On 1 Dec 2005 07:09:44 -0800, "Mark" wrote: But my understadning is that synchronized SS users do not interfere with each other when using orthogonal spreading codes. Let me clarify two things... 1) There is a limit to the number of orthogonal codes so there is a limit to the number of users...you can NOT pack more and more aggregate data into the BW then you could without SS. 2) sychronized means that the SS spreading signals are synchronized ____when they arrive at the Rx____ the system must take the prop delay into account so that the signals are synched at the Tx. Mark You are right. But as I said, in a real system, the degree of synchronization is way below optimum because of multipath. The signal from any phone to the base station travels by many paths, of unequal length and hence delay. This destroys the synchronization. You can only take propagation delay into account for one of those paths. As for the other direction, phones receive signals not only from their own base station, but also from those nearby. The base stations do their best to synchronize, but unless you are stood equidistant from all nearby base stations, you don't get synchronized interference. So although you do get a certain amount of benefit from RW (orthogonal) coding, it is in reality way beyond what theory would give you. d Pearce Consulting http://www.pearce.uk.com Sorry that last sentence should read way BELOW. d Pearce Consulting http://www.pearce.uk.com |
#20
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
Don Pearce wrote: On Thu, 01 Dec 2005 15:16:08 GMT, (Don Pearce) wrote: On 1 Dec 2005 07:09:44 -0800, "Mark" wrote: But my understadning is that synchronized SS users do not interfere with each other when using orthogonal spreading codes. Let me clarify two things... 1) There is a limit to the number of orthogonal codes so there is a limit to the number of users...you can NOT pack more and more aggregate data into the BW then you could without SS. 2) sychronized means that the SS spreading signals are synchronized ____when they arrive at the Rx____ the system must take the prop delay into account so that the signals are synched at the Tx. Mark You are right. But as I said, in a real system, the degree of synchronization is way below optimum because of multipath. The signal from any phone to the base station travels by many paths, of unequal length and hence delay. This destroys the synchronization. You can only take propagation delay into account for one of those paths. As for the other direction, phones receive signals not only from their own base station, but also from those nearby. The base stations do their best to synchronize, but unless you are stood equidistant from all nearby base stations, you don't get synchronized interference. So although you do get a certain amount of benefit from RW (orthogonal) coding, it is in reality way beyond what theory would give you. d Pearce Consulting http://www.pearce.uk.com Sorry that last sentence should read way BELOW. d Pearce Consulting http://www.pearce.uk.com ok, thanks Mark |
#21
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
Wow, again. You are making my head hurt...haven't thought much about
this stuff in YEARS. Yeah, DirectTV....did a single test with a straight drop to one TV...same issue. Unless the whole house is ringing back into the antenna, it looks like a signal problem and not local interferrence. I thought the old sound was OK but not great. Even the new stuff is good for low bandwidth stuff, which is anything but classical or...at the top of the heap...opera. This is a truly great discussion. One thing to consider is that if you have spread spectrum, which is a way of distributing the information across a wide bandwidth, you also have multicoding schema embeded. The good stuff used what NASA came up with...actually adapted...for deep space telemetry. They used what were called inner and outer codes and Gold codes to make exra-ordinary reduncancy in the signal. This let the S/N creep do damn near zero and with enough integration and processing, you could pull something out of nothing. Frankly, I had forgotten some of the stuff the others are talking about. But in my early years in lasers, we use to dsay, if you pump something hard enough, it will lase, even the fabled jello laser, which was based on lasing the florescent dyes in the damned stuff. The same types of truisms hold for signal processing. With enough crunch power, you can encode the hell out of stuff and basically recover a signal from nothing if it was there to begin with, even in the presences of all the stuff we dread, such as fading, multipath, interference, drops, etc. It is truly a matter of bandwidth and crunching power. Now, the trick is to get this crunch to real time at a no cost....GOTCHA!!! Henry On 30 Nov 2005 18:12:53 -0800, "Mark" wrote: None of them is, or can be, idea. The signal transmitted by a SS transmitter will, at best, look like purely random "white noise" to other users of that frequency band, and will necessarily degrade the signal-to-noise ratio seen by those other users. Similarly, other users of a frequency (whether narrow-, wide-, or spread-spectrum) will at best look like white noise to the receiver in a spread spectrum system, and will degrade that receiver's SNR to some extent. Dave, great post.. I have one little nit to pick..... It is my understanding that 2 or more SS users that are synchronized to each other and are using __orthogonal__ spreading codes will not create and interference to each other, not even noise.... To the OP If you have 3 TVs hooked up, your ghosts are probably from all the wiring..... try it with one TV directly connected ot the sat rcv. And also to the OP... I assumme you are talking about Direct TV which now has XM radio....how did you like the previous Music Choice Service... Mark |
#22
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
True, but there are scheme where Shannon is finessed. Bandwidth, my
man, and encoding. It is the encoding. You aren't cheating Dr. Shannon...who I think may have just died a few years ago...RIP great man....you are really being profligate with bandwidth, not chincing on it. That is the trade off. Redundancy bought with profiligate bandwidth. As I said before, the trick is to do things in real time and cheaply. We used to collect data and crunch for a week to pull the signal out. Henry On Thu, 01 Dec 2005 08:42:54 GMT, (Don Pearce) wrote: On 30 Nov 2005 18:12:53 -0800, "Mark" wrote: None of them is, or can be, idea. The signal transmitted by a SS transmitter will, at best, look like purely random "white noise" to other users of that frequency band, and will necessarily degrade the signal-to-noise ratio seen by those other users. Similarly, other users of a frequency (whether narrow-, wide-, or spread-spectrum) will at best look like white noise to the receiver in a spread spectrum system, and will degrade that receiver's SNR to some extent. Dave, great post.. I have one little nit to pick..... It is my understanding that 2 or more SS users that are synchronized to each other and are using __orthogonal__ spreading codes will not create and interference to each other, not even noise.... That is true to an extent. First, it only holds in single path conditions like a cable connection. In the real world radio waves arrive by multiple paths of different lengths, so destroying the coherence of the code edges. Otherwise orthogonal codes then look increasingly like plain old noise. Secondly, whatever the coding, there is still power there, which will cause blocking of the front end of the receiver. There is a big fallacy concerning spread spectrum that it somehow makes really efficient use of radio spectrum because if you spread far enough it "disappears". This is of course nonsense, it doesn't matter whether you use TDMA, FDMA, OFDM, SS or whatever. You still need the same bandwidth * SNR product to shift a given amount of data. Shannon will not be denied. d Pearce Consulting http://www.pearce.uk.com |
#23
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
I had forgotten this stuff. The problem really is that we are trying
to maximize information and noice/interference immunity in a too small bandwidth....and doing a really good job with what we have to work with. The irony is that with a full-fledged robust spread spectrum system, theoretically, bandwidth restrictions become moot. You can operate in any band without screwing up the owners of that band, but regulation won't allow it and nobody can really build these things cost effectively. This was the same thing with writing efficient code in the days when processors were slow and memory was dear. Anybody old enough to remember that pain. It is gone now as will the bandwidth concept in some near future...but those who have paid dearly for that bandwidth will stiffle any try attempt to move beyond that box. Does the wireless, cable, and phone regulatory mess ring any bells? Follow the money. Henry On Thu, 01 Dec 2005 15:16:08 GMT, (Don Pearce) wrote: On 1 Dec 2005 07:09:44 -0800, "Mark" wrote: But my understadning is that synchronized SS users do not interfere with each other when using orthogonal spreading codes. Let me clarify two things... 1) There is a limit to the number of orthogonal codes so there is a limit to the number of users...you can NOT pack more and more aggregate data into the BW then you could without SS. 2) sychronized means that the SS spreading signals are synchronized ____when they arrive at the Rx____ the system must take the prop delay into account so that the signals are synched at the Tx. Mark You are right. But as I said, in a real system, the degree of synchronization is way below optimum because of multipath. The signal from any phone to the base station travels by many paths, of unequal length and hence delay. This destroys the synchronization. You can only take propagation delay into account for one of those paths. As for the other direction, phones receive signals not only from their own base station, but also from those nearby. The base stations do their best to synchronize, but unless you are stood equidistant from all nearby base stations, you don't get synchronized interference. So although you do get a certain amount of benefit from RW (orthogonal) coding, it is in reality way beyond what theory would give you. d Pearce Consulting http://www.pearce.uk.com |
#24
Posted to rec.audio.tech
|
|||
|
|||
Recording XM signals...an issue
|
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
common mode rejection vs. crosstalk | Pro Audio | |||
Artists cut out the record biz | Pro Audio |