Pulse or digital communications – Systems using alternating or pulsating current – Plural channels for transmission of a single pulse train
Reexamination Certificate
1998-10-09
2001-12-04
Vo, Don N. (Department: 2631)
Pulse or digital communications
Systems using alternating or pulsating current
Plural channels for transmission of a single pulse train
C375S270000
Reexamination Certificate
active
06327311
ABSTRACT:
BACKGROUND
1. Field of the Invention
The present invention relates to high-speed data transmission over unconditioned metallic wiring and specifically to an efficient modulation technique for robust data transmission over severely distorted channels.
2. Background
Characteristics of Wiring in Homes and Buildings
In many instances, it is desirable to install communications networks in homes and businesses using the pre-existing wiring. Utilizing the pre-existing wiring allows the homeowner or business owner to network the building using the existing copper infrastructure without a major investment in the installation of optical fiber or other network transmission media. However, the network also needs to be capable of transmitting data at high data rates.
The pre-existing wiring (i.e., telephone wiring and power wiring) of most homes and other buildings is not of uniform type and may consist of 24 gauge twisted quad wiring, unshielded flat pair, or other miscellaneous types of wiring. This wiring can produce severely distorted transmission channels.
FIG. 1
shows an example of a network
100
using existing 24 gauge twisted copper, such as the existing telephone lines in a home or business. Network
100
includes a main line
101
and trunk lines
102
,
103
and
104
, which are each coupled at one end to main line
101
. Main line
101
includes a signal source
105
at one end and a receiver terminator
106
at the opposite end. Receiver terminator
106
provides main line
101
with a 100 Ohm termination. In
FIG. 1
, main line
101
is 360 feet long. Trunk line
102
is 80 feet long and is coupled to main line
101
at a point 170 feet from signal source
105
. Trunk line
103
is 25 feet long and is coupled to main line
101
at a point 90 feet from receiver terminator
106
. Trunk line
104
is 25 feet long and is coupled to main line
101
at a point 40 feet from receiver terminator
106
. Trunk lines
103
and
104
each have open, unterminated ends (i.e., infinite termination) opposite the end that is coupled to main line
101
. Trunk line
102
includes a 100 ohm terminator at an end opposite the end of trunk line
102
that is coupled to main line
101
. Other examples of networks can include any number of terminated, unterminated or improperly terminated lines.
FIG. 2
shows the frequency response of the transmission channel between signal source
105
and receiver terminator
106
of network
100
shown in FIG.
1
. The unterminated trunk lines, trunk lines
103
and
104
, cause a deep null in the spectrum of the frequency response. Other networks may have multiple spectral nulls or a differently shaped frequency response.
Other sources of spectral nulls or distortions in the frequency response of a transmission channel include filters to reject interference from HAM radio bands.
FIG. 3
shows the combined response of transmit and receive filters in a passband modulated transceiver, including RFI suppression filters, for a transmission band of between 4 MHz and 10 MHz. The spectral null in the center of the spectrum suppresses the 40 meter HAM band.
As long as the signal-to-noise ratio of a received signal is sufficiently high, channel distortion can be corrected by equalization. Near-optimal throughput can be achieved by using a decision-feedback equalizer or equivalent structure. (See G. D. Forney, Jr., and M. V. Eyuboglu,
Combined Equalization and Coding Using Precoding,
IEEE COMM. MAG., vol. 29, no. 12, pp. 25-34, December 1991.) An ideal decision-feedback equalizer (DFE) or equivalent precoding structure, in combination with a fractionally-spaced feedforward equalizer (FSE), can correct the distortion from a transmission channel in an optimal manner, enabling the achievable throughput to approach the theoretical channel capacity arbitrarily closely with the use of sufficiently complex coding schemes (See J. M. Cioffi, et al.,
MMSE Decision
-
Feedback Equalizers and Coding—Part I: Equalization Results
, IEEE TRANS COMM., vol. 43, no. 10, pp. 2582-2594, October 1995; J. M. Cioffi, et al.,
MMSE Decision
-
Feedback Equalizers and Coding—Part II: Coding Results
, IEEE TRANS COMM., Vol. 43, no. 10, p. 2595-2604, October 1995).
However, when the transmission band of the channel contains deep spectral nulls and the signal-to-noise ratio is low, a large part of the transmission band may become unusable. This can easily happen when the transmitted signal power is limited and the spectrum of the transmitted signal is constrained within a narrow bandwidth to allow spectral compatibility with other signals on the transmission channel. In cases in which the power spectral density (PSD) is constrained or in which the SNR is limited by self-crosstalk, the frequency-dependent SNR is fixed and the SNR cannot be improved by increasing the transmit PSD.
FIG. 4
shows a combined response of the transmit and receive filters of FIG.
3
and the transmission channel between signal source
105
and receiver termination
106
of FIG.
1
. In
FIG. 4
, much of the spectrum is unusable because it is near or below the noise floor of −120 dBm/Hz.
In such cases where the signal-to-noise ratio is relatively low and the channel contains large spectral nulls, the achievable throughput for traditional single-carrier modulation using integral bits per symbol may be zero. For example, a single-carrier transceiver operating with a baud rate of 4 Mhz, a 15 dB gap (a measure of the difference between the theoretical channel capacity and the achievable channel capacity) and integer bits per symbol on a channel having the power spectral density shown in
FIG. 4
has an achievable capacity of zero bits per symbol. Therefore, traditional single-carrier modulation fails. Single carrier modulation schemes are further discussed below. Theoretical capacity, achievable capacity, and the gap between them are further discussed below.
The problem of transmitting data through noisy channels having large spectral nulls is often solved by using transceivers that utilize either multicarrier modulation or frequency diverse modulation schemes. Multicarrier modulation or frequency diverse modulation schemes may provide acceptable throughput in such cases, but these schemes have additional implementation complexity and other practical disadvantages in comparison with single-carrier modulation transceivers.
Multi-Carrier Modulation
Multi-carrier modulation is a popular solution in some applications. The most common type of multi-carrier modulation is Discrete Multi-Tone (DMT) modulation. See J. A. Bingham, et al.,
Multicarrier Modulation for Data Transmission: An Idea Whose Time Has Come,
IEEE COMM. MAG., May 1990, 5-14; I. Kalet,
The Multitone Channel,
IEEE TRANS. COMM., Vol. 37, No.2, February 1989. On typical subscriber-loop channels, for example, DMT modulation generally achieves the same throughput as single-carrier modulation, assuming equivalent coding methods and properly optimized parameters. On severely distorted channels with large unusable spectral regions, however, DMT modulation transceivers may achieve better throughput than single-carrier modulation transceivers, especially when the capacity gap (see discussion of channel capacity below) is large.
DMT modulation transceivers have some disadvantages, however, as compared to single carrier modulation transceivers. A first disadvantage is that DMT modulation requires that the transmitter be informed of the transmission channel response. Therefore, DMT requires significant amounts of information flow from the receiver to the transmitter as well as data flow from the transmitter to the receiver. In addition, DMT modulation has a much higher peak-to-average ratio than single-carrier modulation, requiring the use of more expensive analog-to-digital and digital-to-analog converters with greater dynamic ranges than is required in single-carrier systems. DMT modulation also has less natural immunity to narrowband interference than single-carrier modulation. In addition, DMT modulation has a more complex transceiver structure compared to single-
Broadcom HomeNetworking, Inc.
Christie Parker & Hale LLP
Vo Don N.
LandOfFree
Frequency diverse single carrier modulation for robust... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Frequency diverse single carrier modulation for robust..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Frequency diverse single carrier modulation for robust... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2578929