Re: UVS : Amati Market Cap when I bought ?
Not sure,
I purchased at $5.00 a share.
The chart went up & down until TI boght them for $20. a Share.
Amati, Western and PairGain Share prices doubled on the first Hype of DSL.
The whole story is laid out on this Board.
https://www.siliconinvestor.com/subject.aspx?subjectid=1256
Which is close to or Surpassed by the DD that Has been found and documented here.
I was deep into the technology and posted often.in fact when I began researching ADSL you got a whole 6 hits on the search engine of the day, HOTBOT.COM.
How Deep ?
I tried explaining in Layman's terms why Amati's DMT Discreet Multi-Tone Line Code was far Superior to AT&T'store CAP Line Code
Why DMT ?
https://www.siliconinvestor.com/readmsg.aspx?msgid=555777
There are two types of transmission systems for Asymmetric digital subscriber line (ADSL).
Discrete multi-tone modulation (DMT), and Carrierless amplitude modulation (CAP). The CAP transmission scheme is similar to Quadrature amplitude modulation (QAM).
Here in Florida the ANSI working group T1E1.4 in 1993 agreed to base it's interface definition on the DMT modulation method, after considerable debate, including what became known as the ADSL Olympics. Not a Bake-off, as some have come to call it.
This telco Olympic Event involved BT, Bellcore and Nynex, with performance measuring on early (93) versions of DMT, CAP and QAM prototypes.
The European Standards working group ETSI STC TM3 RG12 were highly instrumental in providing ANSI with information and text for this annex, which was edited by BT.
Why DMT?
There are several types of Noise that can impair digital data transmission on POTS (plain old telephone system) which in the US can be referred to as Copper Twisted Pair.
NOISE OR IMPAIRMENTS INTERFERING WITH
THE TRANSMISSION OF DATA FLOW ON POTS:
Near-End Cross-talk (NEXT)
Far-End Cross-talk (FEXT)
Radio Frequency Interference (RFI)
Impulsive Noise
All of these noise sources will, in combination with each other, or as a solo interference, to a lesser or greater extent, will effect the data carrying capacity of any or all channels.
NEXT is caused by signals from transmitters or other pairs operating in the same multi-pair cable.
These signals interfere with the input of a transceiver at the same end.
NEXT in most cases, is the most significant noise source which limits the reach of high-rate duplex data transmission systems.
FEXT occurs when signals from transmitters on other pairs in the same cable leak into the input of the wrong transceiver.
FEXT does not interfere, for the most part, with symmetrical digital transmission systems i.e. HDSL because the FEXT always suffers more attenuation than the signal passing through the channel.
Though for ADSL FEXT is of concern where the different directions of transmission occupy different frequency bands. As with NEXT, FEXT is usually worst when the interfering signals on the pairs are of the same kind as the system being interfered with. This phenomenon is known as self-FEXT.
RFI, this interference does not bother ADSL to a great degree because ADSL is frequency spectrum is low, near the 20khz range.
Radio Frequency Interference (RFI)
RFI is a noise source that effects most access wire pairs. It can vary in considerable amounts from pair to pair, plus RFI has characteristics which show temporal variations.
Digital transmission systems must tolerate certain levels of RFI, in the UK this is prescribed by current European legislation on electromagnetic compatibility (EMC). ( statement from BT)
Various amounts of RFI will ultimately exist in the same frequency range as our transmitted signal.
RFI interference is in band, and cannot be filtered out. Copper wire transmission systems should not interfere with radio signals. This puts a limit on the transmitted power spectral density. I could go off here and get deep, but I'm trying to keep this material on the light side, so I'll skip it.
As I mentioned above RFI is a significant noise source for VDSL that occupy the same bands as amateur radio enthusiasts (1.8 - 30mhz). Peaking your interest now, I hope. This frequency range of RFI is difficult to cancel adaptively because of the on/off nature of the side-band surpressed carrier transmission.
More later..
Impulse Noise - DMT's Friend, CAP's arch enemy.
Impulse Noise is major impairment for digital access transmission systems running on copper twister pair.
There are a variety of sources that produce short electrical transients, known as impulse noise.
Household appliances switching on and off (e.g.) Air Conditioner, Furnace, fluorescent lights, refrigerators, etc.)
The telephone itself ( on-hook/off-hook and ringing.)
These transient disturbances can be electromagneticly coupled into the access network and may cause error bursts in digital transmission.
Amati's DMT uses forward error correction (FEC) and data interleaving in ADSL and VDSL to reduce the effects of impulse noise. The higher the frequency transmission the more these effects are felt, and need to be addressed. Thus VDSL is more susceptible than ADSL.
Let me take a minute to discuss more reasons why DMT is the better digital transmission technology.
As most know ADSL & VDSL transmit the Downstream and upstream paths at the same time. There is need for a way to separate these two signals so they do not interfere with each other.
CAP uses frequency division multiplexing (FDM) The low bit rate control channel occupies the frequency range immediately above the voice-band region of telephony. The high bit rate downstream channel occupies the higher frequencies. The total frequency range for ADSL is limited to approximately 1 Mhz, by encoding several bits per symbol. Use of FDM forces the downstream channel to occupy higher frequencies which causes increased attenuation and hence a reduction in range.
DMT uses asymmetric echo cancellation which models and cancels the interference between Upstream and downstream channels. Echo cancellation yields about a 2dB of performance improvement over the FDM method.
CAP is a 2-D transmission scheme similar to QAM, one upstream channel, one downstream channel.
Discrete Multi-tone modulation (DMT) in it's current form for ADSL consists of 256 channels. It might be easier for you to think of DMT as a bank of CAP/QAM systems operating simultaneously in parallel, each with a carrier frequency corresponding to a DMT sub-channel tone frequency. In other words the DMT transmitter modulates data by forming tone bursts for a number of frequencies, combining them together and sending them as a DMT symbol.
Fast Fourier transform (FFT)
The multi-carrier modulation (and demodulation) requires orthogonality between various sub-channels.
This can be accomplished in an efficient all-digital realization by exploiting Fast Fourier transform (FFT) methods. ( J. Bingham - Multi-carrier modulation from data transmission over the local cable networkIEEE Communication Magazine)
Energy from an impulse noise event will modify a received symbol but the FFT "spreads the effect over the large number of sub-channels within the FTT window. The event is therefore less likely to cause an error.
The use of FEC and interleaving gives DMT transceivers resilience against even longer impulse noise bursts. The current ANSI standard for DMT ADSL transceivers will protect up to 3 DMT symbols.
CROSS-TALK
Another problem would be continuous cross-talk. A steady state of background noise such as cross-talk can be handled by the use of trellis coding. This will add an additional 2.5 dB performance gain when used in conjunction with FEC. This is another option for ANSI standard-compliant systems.
A DMT system has an even greater flexibility to deal with multiple RFI tones, this is achieved by simply not assigning information to those sub-channels where RFI exists.
VDSL AND RFI
With a DMT-based VDSL system it is not necessary to have a transceiver "react" to RFI by adapting to it during operation. One example: Transmission in the DMT sub-channels that correspond to amateur radio bands can be prohibited permanently by taking this into account in the design of the transceiver. ( P. Chow and John Cioffi : "Broadband digital subscriber lines" IIR Maximizing Copper Conference, London (March 1995). This principle of blanking out transmission at certain frequencies has had to be used by some cable companies to avoid interference (to and from) amateur radio enthusiasts.
As with ADSL, the performance of a Basic DMT VDSL system can be increased by around 6dB via the use of trellis coding, echo cancellation, optimized transmit power distribution and larger DMT sub-channel constellations. Although greater transmit power would further increase performance, the scope for this is limited with VDSL, since transmit power significantly greater than + 10dBm would be prohibited by EMC emission constraints and worst case pair balance. This is lower than ADSL transmit power levels because VDSL operates at frequencies where the copper cable becomes a more efficient radiator. (note: a point where I believe CAP leaks, effecting amateur radio and short-wave transmissions).
In the presence of a very pessimistic noise source (HDSL and ISDN cross-talk plus RFI), DMT ADSL is anticipated to be able to operate (with a 6dB performance margin) over customer lines having a insertion loss of up to about 51dB at 300 kHz (R. Kirkby "loop set for 2.048Mbits ADSL with noise model A"
ETSI STC TM3 WGI(RG12) TD No. 12 Ipswitch, UK (June 94).
This is the consensus of a variety of telcos and manufactures all performing independent performance analysis for ADSL standardization work. P.J. Etter "Modelling results for DMT ADSL transmission": ETSI STC TM3 WG1 (RG12) TD No. 12, Ipswitch, UK (June 94) and M.J. Genossar "Simulation results for 2.048 kbit/s ADSL with noise model A", ETSI STC TM3 WGI(RG12) TD No. 12 Ipswitch, UK (June 94).
DMT Vs CAP
1. DMT's ability to adaptively shape the Allocation of information and transmit power across a given bandwidth enables it to better approach optimum performance than is theoretically possible with CAP/QAM
2. The complex startup procedure required for DMT could result in longer activation times than is possible with CAP/QAM
3. DMT offers greatest performance advantages over CAP/QAM at high transmission rates --- the shorter loops at higher rates have more usable bandwidth giving DMT a greater degree of freedom ( more usable sub-channels for flexible adaptation.
4. DMT is more adept than CAP/QAM at coping with multiple RF interferes
5. It is simple for DMT to meet an arbitrary transmit power mask specification to meet spectral compatibility requirements.
6. echo- cancellation in DMT is non-trivial
7. DMT has greater inherent immunity to impulsive noise than CAP/QAM
8. CAP/QAM can use a simpler forward error correction than DMT
9. DMT requires minimal equalization with slower signal processing rates than CAP/QAM DMT hardware can be more easily programmed to support a variety of upstream and downstream data rates with potential for on-line configuration.
10. DMT need be no more complex to implement than CAP (in terms of signal processing MIPS per area of VLSI real estate), although it may be more complicated to understand and design, in particular the Startup sequence.
11. DMT has the potential for more sophisticated embedded line monitoring functionally (e.g. for cable sheath management).
Primary Reference Source
The BT Technical Journal vol 13 No 4 October 1995
"Broadband multimedia delivery over copper"
by G. Young, K.T. Foster and J.W. Cook
This document contains a summation of only a fraction of the BT Technical Journal article on " Broadband multimedia delivery over copper"
This documents only intent was to highlight the need for, the technologies of the current ANSI Standard DMT, to insure the customer buys a product that can handle the real world environment, due to the poor state of today's dirty old copper, which has been in place for many decades.
If there are any engineers who find fault in my analysis in the above, who may not wish to post in a public forum, you may contact me at "jw@nexusprime.org" as I realize some of my sources of information are dated and improvements in both technologies has been achieved, and that my interpretation of the information may be incorrect. I will gladly post any constructive corrections you may wish to add.
All comments are welcomed.
closing I would like to thank all those who have worked in the development of DMT. Through your dedicated efforts, the teleco's and the customers will have a reliable system of digital transmission on dirty old copper wire, if only Alexander Gramm Bell could see you now. He would not roll over in his grave, he would be doing back flips.
Disclaimer :
This document is my analysis of Reference sources stated. It may have inaccuracies or interpretations which may not be actuate. Always verify information for yourself, before investing based on any information in an on-line forum. Impostors, personnel from rival companies, broker houses, MM firms, could be posting bias or dis-information campaigns. This possibly in order to manipulate the price of a particular stock in order to increase the chances of making a larger profit.
This document does not contain any information that is not available to the public.
Any information used was either researched by myself or provided to me upon my specific request.
Background
I am not a communications engineer, and the closest I have come to telecommunications before January 1996, was to debate buying MCI stock in December of 1974, on a tip from a friend in the HR office at MCI. "Long term" I recall were his words. As with Benjamin Franklin and his "Shock " of discovering electricity, some of us must learn the hard way.
I hope this document will help those who wish to further understand why I sign-off with,
I Want My DMT
JW@KSC
Not sure,
I purchased at $5.00 a share.
The chart went up & down until TI boght them for $20. a Share.
Amati, Western and PairGain Share prices doubled on the first Hype of DSL.
The whole story is laid out on this Board.
https://www.siliconinvestor.com/subject.aspx?subjectid=1256
Which is close to or Surpassed by the DD that Has been found and documented here.
I was deep into the technology and posted often.in fact when I began researching ADSL you got a whole 6 hits on the search engine of the day, HOTBOT.COM.
How Deep ?
I tried explaining in Layman's terms why Amati's DMT Discreet Multi-Tone Line Code was far Superior to AT&T'store CAP Line Code
Why DMT ?
https://www.siliconinvestor.com/readmsg.aspx?msgid=555777
There are two types of transmission systems for Asymmetric digital subscriber line (ADSL).
Discrete multi-tone modulation (DMT), and Carrierless amplitude modulation (CAP). The CAP transmission scheme is similar to Quadrature amplitude modulation (QAM).
Here in Florida the ANSI working group T1E1.4 in 1993 agreed to base it's interface definition on the DMT modulation method, after considerable debate, including what became known as the ADSL Olympics. Not a Bake-off, as some have come to call it.
This telco Olympic Event involved BT, Bellcore and Nynex, with performance measuring on early (93) versions of DMT, CAP and QAM prototypes.
The European Standards working group ETSI STC TM3 RG12 were highly instrumental in providing ANSI with information and text for this annex, which was edited by BT.
Why DMT?
There are several types of Noise that can impair digital data transmission on POTS (plain old telephone system) which in the US can be referred to as Copper Twisted Pair.
NOISE OR IMPAIRMENTS INTERFERING WITH
THE TRANSMISSION OF DATA FLOW ON POTS:
Near-End Cross-talk (NEXT)
Far-End Cross-talk (FEXT)
Radio Frequency Interference (RFI)
Impulsive Noise
All of these noise sources will, in combination with each other, or as a solo interference, to a lesser or greater extent, will effect the data carrying capacity of any or all channels.
NEXT is caused by signals from transmitters or other pairs operating in the same multi-pair cable.
These signals interfere with the input of a transceiver at the same end.
NEXT in most cases, is the most significant noise source which limits the reach of high-rate duplex data transmission systems.
FEXT occurs when signals from transmitters on other pairs in the same cable leak into the input of the wrong transceiver.
FEXT does not interfere, for the most part, with symmetrical digital transmission systems i.e. HDSL because the FEXT always suffers more attenuation than the signal passing through the channel.
Though for ADSL FEXT is of concern where the different directions of transmission occupy different frequency bands. As with NEXT, FEXT is usually worst when the interfering signals on the pairs are of the same kind as the system being interfered with. This phenomenon is known as self-FEXT.
RFI, this interference does not bother ADSL to a great degree because ADSL is frequency spectrum is low, near the 20khz range.
Radio Frequency Interference (RFI)
RFI is a noise source that effects most access wire pairs. It can vary in considerable amounts from pair to pair, plus RFI has characteristics which show temporal variations.
Digital transmission systems must tolerate certain levels of RFI, in the UK this is prescribed by current European legislation on electromagnetic compatibility (EMC). ( statement from BT)
Various amounts of RFI will ultimately exist in the same frequency range as our transmitted signal.
RFI interference is in band, and cannot be filtered out. Copper wire transmission systems should not interfere with radio signals. This puts a limit on the transmitted power spectral density. I could go off here and get deep, but I'm trying to keep this material on the light side, so I'll skip it.
As I mentioned above RFI is a significant noise source for VDSL that occupy the same bands as amateur radio enthusiasts (1.8 - 30mhz). Peaking your interest now, I hope. This frequency range of RFI is difficult to cancel adaptively because of the on/off nature of the side-band surpressed carrier transmission.
More later..
Impulse Noise - DMT's Friend, CAP's arch enemy.
Impulse Noise is major impairment for digital access transmission systems running on copper twister pair.
There are a variety of sources that produce short electrical transients, known as impulse noise.
Household appliances switching on and off (e.g.) Air Conditioner, Furnace, fluorescent lights, refrigerators, etc.)
The telephone itself ( on-hook/off-hook and ringing.)
These transient disturbances can be electromagneticly coupled into the access network and may cause error bursts in digital transmission.
Amati's DMT uses forward error correction (FEC) and data interleaving in ADSL and VDSL to reduce the effects of impulse noise. The higher the frequency transmission the more these effects are felt, and need to be addressed. Thus VDSL is more susceptible than ADSL.
Let me take a minute to discuss more reasons why DMT is the better digital transmission technology.
As most know ADSL & VDSL transmit the Downstream and upstream paths at the same time. There is need for a way to separate these two signals so they do not interfere with each other.
CAP uses frequency division multiplexing (FDM) The low bit rate control channel occupies the frequency range immediately above the voice-band region of telephony. The high bit rate downstream channel occupies the higher frequencies. The total frequency range for ADSL is limited to approximately 1 Mhz, by encoding several bits per symbol. Use of FDM forces the downstream channel to occupy higher frequencies which causes increased attenuation and hence a reduction in range.
DMT uses asymmetric echo cancellation which models and cancels the interference between Upstream and downstream channels. Echo cancellation yields about a 2dB of performance improvement over the FDM method.
CAP is a 2-D transmission scheme similar to QAM, one upstream channel, one downstream channel.
Discrete Multi-tone modulation (DMT) in it's current form for ADSL consists of 256 channels. It might be easier for you to think of DMT as a bank of CAP/QAM systems operating simultaneously in parallel, each with a carrier frequency corresponding to a DMT sub-channel tone frequency. In other words the DMT transmitter modulates data by forming tone bursts for a number of frequencies, combining them together and sending them as a DMT symbol.
Fast Fourier transform (FFT)
The multi-carrier modulation (and demodulation) requires orthogonality between various sub-channels.
This can be accomplished in an efficient all-digital realization by exploiting Fast Fourier transform (FFT) methods. ( J. Bingham - Multi-carrier modulation from data transmission over the local cable networkIEEE Communication Magazine)
Energy from an impulse noise event will modify a received symbol but the FFT "spreads the effect over the large number of sub-channels within the FTT window. The event is therefore less likely to cause an error.
The use of FEC and interleaving gives DMT transceivers resilience against even longer impulse noise bursts. The current ANSI standard for DMT ADSL transceivers will protect up to 3 DMT symbols.
CROSS-TALK
Another problem would be continuous cross-talk. A steady state of background noise such as cross-talk can be handled by the use of trellis coding. This will add an additional 2.5 dB performance gain when used in conjunction with FEC. This is another option for ANSI standard-compliant systems.
A DMT system has an even greater flexibility to deal with multiple RFI tones, this is achieved by simply not assigning information to those sub-channels where RFI exists.
VDSL AND RFI
With a DMT-based VDSL system it is not necessary to have a transceiver "react" to RFI by adapting to it during operation. One example: Transmission in the DMT sub-channels that correspond to amateur radio bands can be prohibited permanently by taking this into account in the design of the transceiver. ( P. Chow and John Cioffi : "Broadband digital subscriber lines" IIR Maximizing Copper Conference, London (March 1995). This principle of blanking out transmission at certain frequencies has had to be used by some cable companies to avoid interference (to and from) amateur radio enthusiasts.
As with ADSL, the performance of a Basic DMT VDSL system can be increased by around 6dB via the use of trellis coding, echo cancellation, optimized transmit power distribution and larger DMT sub-channel constellations. Although greater transmit power would further increase performance, the scope for this is limited with VDSL, since transmit power significantly greater than + 10dBm would be prohibited by EMC emission constraints and worst case pair balance. This is lower than ADSL transmit power levels because VDSL operates at frequencies where the copper cable becomes a more efficient radiator. (note: a point where I believe CAP leaks, effecting amateur radio and short-wave transmissions).
In the presence of a very pessimistic noise source (HDSL and ISDN cross-talk plus RFI), DMT ADSL is anticipated to be able to operate (with a 6dB performance margin) over customer lines having a insertion loss of up to about 51dB at 300 kHz (R. Kirkby "loop set for 2.048Mbits ADSL with noise model A"
ETSI STC TM3 WGI(RG12) TD No. 12 Ipswitch, UK (June 94).
This is the consensus of a variety of telcos and manufactures all performing independent performance analysis for ADSL standardization work. P.J. Etter "Modelling results for DMT ADSL transmission": ETSI STC TM3 WG1 (RG12) TD No. 12, Ipswitch, UK (June 94) and M.J. Genossar "Simulation results for 2.048 kbit/s ADSL with noise model A", ETSI STC TM3 WGI(RG12) TD No. 12 Ipswitch, UK (June 94).
DMT Vs CAP
1. DMT's ability to adaptively shape the Allocation of information and transmit power across a given bandwidth enables it to better approach optimum performance than is theoretically possible with CAP/QAM
2. The complex startup procedure required for DMT could result in longer activation times than is possible with CAP/QAM
3. DMT offers greatest performance advantages over CAP/QAM at high transmission rates --- the shorter loops at higher rates have more usable bandwidth giving DMT a greater degree of freedom ( more usable sub-channels for flexible adaptation.
4. DMT is more adept than CAP/QAM at coping with multiple RF interferes
5. It is simple for DMT to meet an arbitrary transmit power mask specification to meet spectral compatibility requirements.
6. echo- cancellation in DMT is non-trivial
7. DMT has greater inherent immunity to impulsive noise than CAP/QAM
8. CAP/QAM can use a simpler forward error correction than DMT
9. DMT requires minimal equalization with slower signal processing rates than CAP/QAM DMT hardware can be more easily programmed to support a variety of upstream and downstream data rates with potential for on-line configuration.
10. DMT need be no more complex to implement than CAP (in terms of signal processing MIPS per area of VLSI real estate), although it may be more complicated to understand and design, in particular the Startup sequence.
11. DMT has the potential for more sophisticated embedded line monitoring functionally (e.g. for cable sheath management).
Primary Reference Source
The BT Technical Journal vol 13 No 4 October 1995
"Broadband multimedia delivery over copper"
by G. Young, K.T. Foster and J.W. Cook
This document contains a summation of only a fraction of the BT Technical Journal article on " Broadband multimedia delivery over copper"
This documents only intent was to highlight the need for, the technologies of the current ANSI Standard DMT, to insure the customer buys a product that can handle the real world environment, due to the poor state of today's dirty old copper, which has been in place for many decades.
If there are any engineers who find fault in my analysis in the above, who may not wish to post in a public forum, you may contact me at "jw@nexusprime.org" as I realize some of my sources of information are dated and improvements in both technologies has been achieved, and that my interpretation of the information may be incorrect. I will gladly post any constructive corrections you may wish to add.
All comments are welcomed.
closing I would like to thank all those who have worked in the development of DMT. Through your dedicated efforts, the teleco's and the customers will have a reliable system of digital transmission on dirty old copper wire, if only Alexander Gramm Bell could see you now. He would not roll over in his grave, he would be doing back flips.
Disclaimer :
This document is my analysis of Reference sources stated. It may have inaccuracies or interpretations which may not be actuate. Always verify information for yourself, before investing based on any information in an on-line forum. Impostors, personnel from rival companies, broker houses, MM firms, could be posting bias or dis-information campaigns. This possibly in order to manipulate the price of a particular stock in order to increase the chances of making a larger profit.
This document does not contain any information that is not available to the public.
Any information used was either researched by myself or provided to me upon my specific request.
Background
I am not a communications engineer, and the closest I have come to telecommunications before January 1996, was to debate buying MCI stock in December of 1974, on a tip from a friend in the HR office at MCI. "Long term" I recall were his words. As with Benjamin Franklin and his "Shock " of discovering electricity, some of us must learn the hard way.
I hope this document will help those who wish to further understand why I sign-off with,
I Want My DMT
JW@KSC
3
