Pulse or digital communications – Systems using alternating or pulsating current – Plural channels for transmission of a single pulse train
Reexamination Certificate
2000-07-26
2004-03-09
Le, Amanda T. (Department: 2634)
Pulse or digital communications
Systems using alternating or pulsating current
Plural channels for transmission of a single pulse train
C370S431000
Reexamination Certificate
active
06704367
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the optimization of transmission bandwidth or discrete multitone modulation (DMT), and more specifically to a discrete loading algorithm for maximizing the system performance or data rate of DMT transmission.
REFERENCES TO RELATED ART
The following is a list of references cited in the disclosure of this invention. For convenience, the reference number of a prior art in the list will be included when the prior art is referred to in the following description.
[1] J. M. Cioffi, “Asymmetric digital subscriber lines,”
Chapter
34
of the Communications Handbook
, Editor-in-Chief, J. D. Gibson, CRC Press in Cooperation with IEEE Press, 1997.
[2] J. M. Cioffi et al., “Very-high-speed digital subscriber lines,”
IEEE Commun. Mag
., vol. 37, pp. 72-79, Apr. 1999.
[3] A. Ruiz, J. M. Cioffi, and S. Kasturia, “Discrete multiple tone modulation with coset coding for the spectrally shaped channel,”
IEEE Trans. Commun
., vol. 40, pp. 1012-1029, Jun. 1992.
[4] T. M. Cover and J. A. Thomas, Elements of Information Theory. Wiley, N.Y., 1991.
[5] J. Campello, “Optimal discrete bit loading for multicarrier modulation system,” in
Proc
. 1998
IEEE Int. Symp. Inform. Theory
, MIT, pp. 193, Aug. 1998.
[6] J. Campello, “Practical bit loading for DMT,” in
Conf. Rec
. 1999
IEEE Int. Conf. Commun
. (ICC '99), Vancouver, Canada, Jun. 1999, pp. 801-805.
[7] A. Leke and J. M. Cioffi, “A maximum rate loading algorithm for discrete multitone modulation systems,” in
Conf. Rec
. 1997
IEEE Global Telecommun. Conf
. (
GLOBECOM
'97), Phoenix, Ariz., November 1997, pp. 1514-1518.
[8] D. Hughes-Hartogs, “Ensemble modem structure for imperfect transmission media,” U.S. Pat. Nos. 4,679,227, (July 1987), 4,731,816 (March 1988), and 4,833,796 (May 1989).
[9] P. S. Chow,
Bandwidth Optimized Digital Transmission Techniques for Spectrally Shaped Channels with Impulse Noise
. Ph.D. Dissertation, Stanford University, May 1993.
[10] P. S. Chow, J. M. Cioffi, and J. A. C. Bingham, “A practical discrete multitone transceiver loading algorithm for data transmission over spectrally shaped channels,”
IEEE Trans. Commun
., vol. 43, pp. 773-775, Febuary/March/April 1995.
[11] R. F. H. Fischer and J. B. Huber, “A new loading algorithm for discrete multitone transmission,” in
Conf. Rec
. 1996
IEEE Global Telecommun. Conf
. (
GLOBECOM
'96), London, pp. 724-728, November 1996.
[12] American National Standards Institute (ANSI), “Network and Customer Installation Interfaces-Asymmetric Digital Subscriber Line (ADSL) Metallic Interface,”
Draft American National Standard for Telecommunications
, Jun. 12, 1998.
[13] L. M. C. Hoo, J. Tellado, and J. M. Cioffi, “Dual Qos loading algorithms for multicarrier systems,” in
Conf. Rec
. 1999
IEEE Int. Conf. Commun
. (
ICC
'99), Vancouver, Canada, June 1999, pp. 796-800.
BACKGROUND OF THE INVENTION
As the Internet gradually becomes part of our life, the demand for high-speed transmission networks is ever increasing. There have been a number of approaches proposed for improving/constructing the information infrastructure. Among them, asymmetric digital subscriber line (ADSL) technology [1] and very-high-speed digital subscriber line (VDSL) technology [2] are two of the most promising solutions that can provide data rates up to 8 megabits per second and 52 megabits per second respectively over ordinary twisted-pair phone lines from a central office to a customer's premise. For the ADSL service, the discrete multitone modulation (DMT) [1], [3] has been selected as a standard by various standards institutes. Transmission technique using DMT is now also being considered as an international standard for the future VDSL service.
A basic DMT structure [1] is shown in FIG.
1
. At first, the input data stream is encoded, including the use of forward error correcting (FEC) codes, trellis codes (optional), and interleaving. The usable bandwidth of a channel is divided into N subchannels (or tones) that are assumed to be independent. With an appropriate loading algorithm, the data bits to be transmitted are assigned to these N subchannels for transmission, where the bit loading algorithm is trying to optimize the transmission bandwidth based on all the subchannels' conditions.
The data bits assigned to each subchannel are mapped onto QAM constellation to form a complex sample, and then the resulting N complex samples from the N subchannels are extended to be a 2N-point complex-conjugate symmetric sequence. This 2N-point complex sequence is further modulated by the 2N-point inverse fast Fourier transform (IFFT) to generate 2N-point real samples for transmission through the channel. In order to overcome severely intersymbol interference (ISI) and to make the transmitted signals look periodic, the last v samples of each 2N-sample block are circularly wrapped to prefix the block itself. After receiving the signals transmitted, the receiver discards the first v samples, and then the remaining received samples are demodulated by the 2N-point FFT. The resulting transform samples are further processed by a frequency-domain equalizer (FEQ) and a memoryless decoder to recover the original data bits.
The problem of optimizing the transmission bandwidth was first solved by Shannon in 1948 [4], known as “water-filling” method, but the corresponding method is impractical to be implemented for the DMT system due to its high complexity and infinite-granularity in the constellation size. To overcome this problem, several discrete loading algorithms have been proposed for an optimal or suboptimal solution in the finite-granularity constellation case. Basically, these algorithms can be classified into two categories [5], [6]:
(1) Bit Rate Maximization Problem (BRMP)—Maximize the data rate subject to power and system performance margin constraints, i.e.,
max
∑
n
=
1
N
b
n
(
1
)
subject to
∑
n
=
1
N
p
n
(
b
n
)
≤
p
(
2
)
where b
n
is the number of bits that are transmitted on the n
th
tone, p
n
id the power distribution required to transmit b
n
bits on the n
th
tone, and p is the total power constraint.
(2) Margin Maximization Problem (MMP)—Maximize the system performance margin subject to power and data rate constraints, i.e.,
min
∑
n
=
1
N
p
n
(
b
n
)
(
3
)
subject to
∑
n
=
1
N
b
n
=
B
(
4
)
where p
n
is a function of b
n
and B is the data rate constraint.
The optimal discrete loading algorithm presented by Leke and Cioffi [7] is of BRMP type. It utilizes the water-filling solution to determine the turned-on subchannels first, and then assigns energy to each turned-on subchannel using the water-filling distribution to maximize the data rate. The Hughes-Hartogs loading algorithm [8] can be regarded as BRMP type or MMP type, depending on the constraint used. This algorithm assigns one additional bit to the subchannel that needs the least energy until the data rate or power constraint is met. It gives an optimal discrete solution but its computational complexity becomes impractical when the number of bits to be transmitted per DMT symbol is large.
The suboptimal discrete loading algorithm proposed by Chow, Cioffi, and Bingham [9], [10] is of MMP type. It distributes the data bits among all the usable subchannels according to a well-known formula, and then assigns energy to each usable subchannel with the flat distribution. Due to the use of the flat-energy distribution, this loading algorithm suffers some performance degradation as compared to the Hughes-Hartogs algorithm but it involves less computational complexity. In another prior art [11], Fischer and Huber paid their attention on minimizing the error probability of transmission (equivalent to MMP) and deriv
Chang Yuan-Shuo
Wang Chin-Liang
Le Amanda T.
Proscend Communications Inc.
LandOfFree
Optimal discrete loading algorithm for DMT modulation does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Optimal discrete loading algorithm for DMT modulation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Optimal discrete loading algorithm for DMT modulation will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3201567