Oh, I forgot, here is your McDMT:
I guess simple answer is Amati won the battle and ANSI set DMT(Discrete Multi-Tone modulation) as ADSL standard.
Do you work at NASA? I used to live in Florida, only a few hours from the space center. And now, I live right next to JPL in California. We try to visit JPL's open house party every year. Our kids love to visit JPL.
I did some extra credit homework:
cs.tamu.edu
Modulation
While the ADSL standard specifies only one type of modulation, researchers have tested and
implemented three different modulation techniques for use with ADSL: Quadrature Amplitude
Modulation (QAM), Carrierless Amplitude-Phase Modulation (CAP), and Discrete Multi-Tone
Modulation (DMT).
Quadrature Amplitude Modulation (QAM)
The binary data stream is split into two sub-streams and separately modulated onto
orthogonal versions of the same carrier frequency. The two modulated signals are then
added and low pass filtered before transmission to the network[1]. This is the least used
modulation technique for ADSL.
Carrierless Amplitude-Phase Modulation (CAP)
Developed by AT&T Paradyne, CAP ADSL equipment is available as an alternative to the
industry standard DMT. CAP products were available before DMT equipment, and some
vendors invested in CAP technology instead of waiting for the DMT equipment to
mature[4]. The bit stream is divided into two symbol streams and fed to two parallel
in-phase and quadrature digital-shaping filters. The transfer functions of the two filters
have the same amplitude characteristics, but the phase characteristics differ by 90
degrees[6]. The outputs are added, passed through a digital-to-analog converter and
filtered before being sent to the transmission network[1]. Data encoded using CAP has the
same spectral characteristics and provides the same theoretical performance as QAM, but
does so without a carrier frequency. Eliminating carrier modulation (mixing) provides two
key advantages. First, it facilitates a less expensive all-digital transceiver implementation
by eliminating the computation-intensive multiply operations that are needed for carrier
modulation and demodulation. Second, the absence of a carrier provides increased
flexibility, because changing the spectrum's center frequency, symmetry and shape can be
done easily by downloading a new set of filter coefficients[6].
Discrete Multi-Tone Modulation (DMT)
This modulation technique was decided upon as the ADSL standard. DMT was developed
by Amati Communications and Stanford University, and commercialized by Northern
Telecom[4]. Each serial digital input signal is first encoded into a parallel form and then
passed through a fast-Fourier-transform processor. This converts the frequency domain
samples into time domain values with a sliding time-window effect. These values are then
transcoded into a serial format and converted from digital to analog before transmission
through the ADSL circuit[1].
Some of the most important parameters for standardized ADSL DMT are described
below[5].
Pilot
Carrier 64 (f=276 kHz) is reserved for a pilot. The data modulated onto the pilot
subcarrier shall be constant 0,0. Use of this pilot allows resolution of sample timing
in a receiver modulo-8 samples.
Nyquist Frequency
The carrier at the Nyquist Frequency (256) may not be used for data.
Synchronization Symbol
The synchronization symbol permits recovery of the frame boundary after the
micro-interruptions that might otherwise force retraining.
------------------
bugs.wpi.edu:8080/EE535/hwk97/hwk3cd97/ryant/ryant.html
Advantages of DMT
As mentioned previously, DMT eliminates the need for most adaptive equalization
based on the condition that the frequency response of the channel be nearly constant
across any subchannel. While this is one advantage of DMT, there are others.
By (carefully) dividing the channel into subchannels and allocating the maximum
allowable number of bits to each subchannel, DMT can be shown to be an optimum
code ([4]). This means that this system can perform at the theoretical limit and that no
other system can exceed its performance. While the proof is beyond the scope of this
paper, the reader is encouraged to see [7], [8] for a formal description and proof.
Since the data rate depends on the SNR of the channel, the channel can be monitored
for changes and the rate dynamically updated so as not to disturb the system
performance. The rate adaptation capability of DMT makes it well suited for a wide
variety of channel conditions and increases the resistance to bursty, high-power noise
(since a channel can be "shut off" for certain periods).
Frequency division multiplexing can be easily integrated into a DMT system by
grouping subchannels. Through this implementation, multiple users can be merged
onto the same channel, each with his/her own transmission rate (based on the number
of subchannels assigned to that user).
*******I still don't get it though <ggg>************* |
