j,
<< I have no motivation to prtest too much. Based on
what I know about signal processing and the differences
between CAP and DMT you are very wrong, but we can agree
to disagree.>>
Since you have knowledge of signal processing, I would like to point you to several papers I read while I worked at Bell Laboratories. The inventors of both DMT and CAP came from the same group (the same one within which I worked) in the Labs, so I am very familiar with both technologies. To wit:
I. Kalet, "The Multitone Channel," IEEE Trans. Comm., vol. 37, pp. 119-124, Feb. 1989.
J. A. C. Bingham, "Multicarrier Modulation for Data Transmission: An Idea Whose Time Has Come," IEEE Communications Magazine, pp. 5-14, May 1990.
I. Kalet and N. A. Zervos, "Optimized Decision Feedback Equalization vs. Optimized Orthogonal Frequency Division Multiplexing for High-Speed Data Transmission Over the Local Cable Network," Proceedings of the IEEE Int'l Conference on Communications, pp. 1080-1085, Sept. 1989. Also to appear in IEEE Trans. Comm.
J. Cioffi, G. P. Dudevoir, M. Eyuboglu, G. D. Forney, "MMSE Decision-Feedback Equalizers and Coding - Part 2: Coding Results," IEEE Trans. Comm., Vol 43, pp. 2582-2594, No. 10, October 1995.
J. Cioffi, G. P. Dudevoir, M. Eyuboglu, G. D. Forney, "MMSE Decision-Feedback Equalizers and Coding - Part 1: Equalization Results," IEEE Trans. Comm., Vol 43, pp. 2595-2604, No. 10, October 1995.
These papers point out that there is no theoretical performance difference between DMT and CAP. The reason the Labs chose CAP is that implementation is fundamentally different (read the white paper I referenced previously).
If you read (and thoroughly understand) these papers (including the white paper) I think we may not disagree nearly as much.
Hope this helps straighten things out a bit.
Luke |
