Pulse or digital communications – Receivers – Interference or noise reduction
Reexamination Certificate
2000-03-01
2001-09-25
Pham, Chi (Department: 2631)
Pulse or digital communications
Receivers
Interference or noise reduction
C333S02800T
Reexamination Certificate
active
06295326
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to digital communications systems, and more specifically, to an approach for removing distortion in multi carrier digital communications systems such as Discrete Multi Tone (DMT) and Orthogonal Frequency Division Multiplexing (OFDM) systems.
BACKGROUND OF THE INVENTION
One of the most important concerns in digital communications systems is minimizing distortion introduced by a communications channel, sometimes referred to as “channel effects.”
FIG. 1
is a block diagram
100
of a conventional digital communications system. Data is transmitted from a transmitter
102
to a receiver
104
over a communications channel (“channel”)
106
. Channel
106
may be any type of communication medium for transferring data between transmitter
102
and receiver
104
. For example, channel
106
may be one or more network connections, wires, fiber-optic links or a wireless digital communications link.
Ideally, data is transmitted between transmitter
102
and receiver
104
over channel
106
without being distorted. That is, the data retrieved by receiver
104
from channel
106
is identical to the data placed onto channel
106
by transmitter. However, in practice, channel
106
introduces distortion that can corrupt data transmitted over channel
106
. The distortion introduced by channel
106
can cause successive transmitted symbols to interfere with each other, otherwise known as inter-symbol interference (ISI). ISI can cause severe corruption of digital data, resulting in a very high bit error rate (BER). The corruption of data occurs while being transmitted across communications channel
106
. In multi-carrier systems ISI must be removed before transformation of time domain data to frequency domain data, e.g., via a fast Fourier transform (FFT), which causes ISI to be spread across all frequency bins, producing significant signal to noise degradation. The common solution to this problem is to remove ISI from the sampled data using time domain equalization (TDEQ) before transforming the sampled data from the time domain data to the frequency domain. This allows the original data to be recovered from the communications channel.
FIG. 2
is a block diagram
200
illustrating an example implementation of transmitter
102
and receiver
104
of FIG.
1
. Transmitter
102
typically includes an encoder
202
, a digital-to-analog converter
204
and a transmit filter
206
and a line driver
207
. Encoder
202
encodes the original digital data to generate encoded digital data. The encoded digital data is converted to analog encoded data by digital-to-analog converter
204
. Transmit filter
206
removes unwanted components of the original data from the encoded analog data to generate filtered data. Line driver
207
amplifies the signal to transmit the signal across channel
106
. The filtered data is transmitted over channel
106
to receiver
104
.
Receiver
104
includes a differential amplifier
208
, one or more receive filters
209
, an analog-to-digital converter
210
and an equalizer
212
. After data is received by receiver
104
from channel
106
, receive filter
209
removes unwanted components from the encoded analog data received from transmitter
102
over channel
106
. Analog-to-digital converter
210
converts the encoded analog data received from transmitter
102
to encoded digital data. Equalizer
212
processes the encoded digital data to remove ISI. The encoded digital data is further processed by converting the data to the frequency domain via a Fast Fourier Transform to recover the modulated tones. Residual frequency domain equalization is performed in the frequency domain before the data is transferred to a decoder
214
that recovers the original digital data. Decoder
214
may be separate from receiver
104
as illustrated, or may be incorporated into receiver
104
.
Most conventional approaches for removing ISI have significant limitations that result in degraded performance. Many of the limitations in conventional approaches are attributable to the nature of ISI and the nature of other external noise sources. Conventional approaches address each problem as independent, i.e., ISI removal and noise mitigation. Thus, solving one problem can exacerbate the effects of another. For example, some TDEQ mechanisms do not account for noise sources such as thermal noise and crosstalk. In addition, the characteristics (transfer function) of a communications channel are not necessarily static and can change over time. As a result, conventional TDEQ mechanisms suffer from a number of drawbacks generally including not removing all of the channel-induced ISI or having stability problems. For example, decision feedback TDEQ mechanisms can provide improved performance, but are difficult to apply to discrete multi-tone (DMT) transmission system applications, which limits their use. Furthermore, decision feedback TDEQ mechanisms can suffer from error propagation problems, wherein the occurrence of one error leads to an increased probability of the occurrence of an error in the subsequent data symbols. As another example, transmission precoding is also used to compensate for known channel effects. However, transmission precoding requires an accurate characterization of the channel and a cooperative transmitter. Furthermore, transmission precoding cannot be implemented in all types of transmission systems. For example, transmission precoding cannot be implemented within the current asynchronous digital subscriber line (ADSL) standards ANSI T1.413 and ITU992.2.
Therefore, based on the need to reduce ISI in digital communications systems and the limitations in the prior approaches, an approach for reducing ISI in digital communications systems, and in particular DSL communications systems, that does not suffer from limitations inherent in conventional ISI removal approaches is highly desirable.
SUMMARY OF THE INVENTION
According to one aspect of the invention, a computer-implemented method is provided for processing data received from a communications channel. First, the data is received from the communications channel, wherein the transmitted data consists of data modulated onto an arbitrary number of carriers. Then, a minimum mean square estimate of the data before the data was transmitted on the communications channel is determined. Finally, the minimum mean square estimate is used to compensate for the distortion introduced into the data by the communications channel and also the external noise sources.
According to another aspect of the invention, a computer-implemented method is provided for equalizing data received from a digital subscriber line, DMT or OFDM based communication system, to remove distortion in the data introduced by transmitting the data along a communications channel, e.g., a digital subscriber line. First, the data is received from the digital subscriber line. Then, a minimum mean square estimate of the data is determined using a Kalman filter. The minimum mean square estimate is used to compensate for the distortion in the data introduced by transmitting the data on the digital subscriber line.
According to another aspect of the invention, a receiver is provided for equalizing analog data received from a communications channel, wherein the analog data includes distortion introduced by transmitting the analog data on the communications channel. The receiver comprises an analog front end (including filtering), an analog-to-digital converter and an equalizer. The analog-to-digital converter is configured to convert the analog data received from the communications channel to digital data. The equalizer is configured to determine a minimum mean square estimate of the analog data received from the communications channel and to use the minimum mean square estimate of the analog data to compensate for distortion introduced into the analog data by transmitting the analog data on the communications channel.
REFERENCES:
patent: 5068873 (1991-11-01), Murakami
patent: 5111481 (1992-05-01), Chen et al.
patent: 543
Logothetis Andrew
Skafidas Efstratios
Tonissen Shane Michael
Bandspeed, Inc.
Becker Edward A.
Hickman Palermo & Truong & Becker LLP
Pham Chi
Tran Khai
LandOfFree
Kalman filter based equalization for digital multicarrier... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Kalman filter based equalization for digital multicarrier..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Kalman filter based equalization for digital multicarrier... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2444735