Patent Application
20100254467
The invention relates to a method and to a device for noise processing and to a communication system comprising such a device.
DSL or xDSL, is a family of technologies that provide digital data transmission over the wires of a local telephone network.
Asymmetric Digital Subscriber Line (ADSL) is a form of DSL, a data communications technology that enables faster data transmission over copper telephone lines than a conventional voice band modem can provide. Such fast transmission is achieved by utilizing frequencies that are normally not used by a voice telephone call, in particular, frequencies higher than normal human hearing.
VDSL (Very High Speed DSL) is an xDSL technology providing faster data transmission over a single twisted pair of wires. High bit rates are achieved at a range of about 300 meters (1000 ft), which allows for 26 Mbit/s with symmetric access or up to 52 Mbit/s in downstream-12 Mbit/s in upstream with asymmetric access.
According to its high bandwidth, VDSL is capable of supporting applications like HDTV, as well as telephone services (e.g., Voice over IP) and general Internet access, over a single connection.
VDSL2 (Very High Speed Digital Subscriber Line 2) is an access technology that exploits the existing infrastructure of copper wires that were originally used for plain old telephone service (POTS). It can be deployed from central offices, from fiber-fed cabinets preferably located near the customer premises, or within buildings.
VDSL2 is designed to support the wide deployment of Triple Play services such as voice, video, data, high definition television (HDTV) and interactive gaming. VDSL2 enables operators and carriers to gradually, flexibly, and cost efficiently upgrade existing xDSL infrastructure.
ITU-T G.993.2 (VDSL2) is an enhancement to G.993.1 (VDSL) that permits the transmission of asymmetric and symmetric (full duplex) aggregate data rates up to 200 Mbit/s on twisted pairs using a bandwidth up to 30 MHz.
The xDSL wide band modulation approaches are problematic relating to crosstalk interference that is introduced to the twisted pair transmission line and received by the modem.
Crosstalk occurs when wires are coupled, in particular between wire pairs of the same or a nearby bundle that are used for separate signal transmission. Hence, data signals from one or more sources can be superimposed on and contaminate a data signal. The crosstalk comprises a near-end crosstalk (NEXT) and a far-end crosstalk (FEXT).
Based on such crosstalk, data signals transmitted over twisted-pair lines can be considerably degraded by the crosstalk interference generated on one or more adjacent twisted-pair phone lines in the same and/or a nearby multi-core cable or bundle. With an increasing transmission speed, this problem even deteriorates, which may significantly limit a maximum data rate to be transmitted via a single line.
A multiple-input-multiple-output system (hereinafter referred to as MIMO system) is of significant importance in modern communication technology. Such MIMO system allows to model crosstalk interference of a telecommunication system.
An impact of NEXT can be reduced by utilizing frequency-division duplex between upstream and downstream direction. Current VDSL2 deployments are (indirectly) FEXT limited due to the choice of a relatively conservative Power Spectral Density (PSD) mask, thereby allowing to reduce NEXT in conjunction with a strong loop attenuation at frequencies beyond 1 MHz.
Downstream precompensation (achieved in particular by precoding) or upstream cancellation may allow utilization of more aggressive PSD masks (in particular with full binder deployments); at least, improvements regarding the reach and/or the data rate can be achieved by reducing crosstalk at existing PSD levels.
A prerequisite for the successful employment of DSM methods is the identification of the FEXT-channel. A particularly important and difficult aspect of FEXT-channel estimation concerns a feedback from the receiver.
Disadvantageously, existing systems do not provide such feedback in a timely fashion.
The problem to be solved is to avoid the disadvantages as set forth above and to provide an efficient input from the receiver side that can be used in particular for coherent and noncoherent wireline MIMO processing.
This problem is solved according to the features of the independent claims. Further embodiments result from the depending claims.
In order to overcome this problem, a method for noise processing is provided comprising the steps:
In particular, the SNR-data may in comprise link state information and/or link quality information.
Such noise processing may in particular comprise noise indication to the precoder and/or to the noise canceller, it may also comprise noise detection and/or reduction conducted at the precoder and/or the noise canceller.
It is to be noted that the precoder allows in particular to process data prior to its transmission via at least one channel or via at least one line such that a noise, e.g., interference and/or crosstalk, is reduced when the data arrive at the receiver(s).
A noise canceller may be associated with a receiver and it may be used to reduce and/or cancel noise, e.g., interference and/or crosstalk by processing data arriving at this receiver.
In particular, the precoder and/or the noise canceller may be associated with or located at a Central Office (CO), to which several lines are attached, wherein said lines are at least partially combined by at least one cable binder.
The precoder as well as the noise canceller may be referred to as noise reduction means in particular processing a precoding matrix and/or a coefficient matrix by utilizing the SNR-data received.
Advantageously, this approach provides a fast (i.e. with low latency) and reliable (i.e. statistically unbiased and with low variance) information about a Signal-to-Noise Ratio (SNR).
Within so-called non-coherent MIMO processing, a correlative analysis of link state and SNR across all managed ports may be utilized in order to perform a binder group classification.
Further, in so-called coherent MIMO processing it is advantageous to detect “leaving” ports, i.e., ports with unexpectedly broken connections due to switching-off of Customer Premises Equipments (CPEs) or due to impulse noise. Such detection may avoid the precoder and/or noise canceller to adjust to a faulty state and hence affect and/or impair other ports as well.
With the receiver providing SNR-data, advantageously nearly instantaneous SNR-information, such impairment of other ports (at the CO) can be avoided.
In an embodiment, the transmitter comprises the precoder that performs precoding of the data to be conveyed pursuant to the SNR-data received.
In another embodiment, the receiver comprises the noise canceller that performs noise cancellation on the data received pursuant to the SNR-data received.
In a further embodiment, wherein the SNR-data is generated based on at least one error detection code.
In a next embodiment, the error detection code is a CRC-Code.
It is also an embodiment that the SNR-data is generated based on at least one error correction code.
Pursuant to another embodiment, the error correction code is a Reed-Solomon-Code.
According to an embodiment, the precoder and/or the noise canceller each comprises a precoding matrix and/or a cancellation matrix comprising coefficients. Said coefficients may in particular be updated pursuant to a given schedule, e.g., at a given rate.
According to another embodiment, the coefficients are frozen in case the SNR-data fulfill a first criterion.
The first criterion may in particular comprise the property that a SNR per carrier (e.g., per tone) is determined.
Hence, such first criterion may stop an adaptation of the coefficients. E.g., in case the SNR-data reach a certain threshold that indicates an impairment of a particular line within the cable (binder), the Central Office may freeze further coefficient adaptation based on this SNR-data from this particular line. As the Central Office may adapt a cancellation matrix and/or a precoding matrix, the impact from this distinct line of low SNR may be disregarded, the coefficients that were frozen may be used for a further restart or reset.
In yet another embodiment, a reset or restart is conducted in case the SNR-data fulfill a second criterion, wherein the frozen coefficients are used for initializing purposes.
Hence, pursuant to the second criterion, a restart or reset may be initiated. Such second criterion can be represented by a threshold of SNR-data showing a high degree of deterioration (e.g., noise).
According to a next embodiment, the first criterion and/or the second criterion comprise at least one of the following:
Pursuant to yet an embodiment, the precoder is used in downstream direction, in particular from a Central Office to several Customer Premises Equipments.
It is also an embodiment that the noise canceller is used in upstream direction, in particular from several Customer Premises Equipments towards the Central Office.
At least one cable binder may be connected to the Central Office.
It is a further embodiment that the precoder and/or the noise canceller are located and/or associated with the Central Office.
According to yet another embodiment, a channel between the at least one transmitter and the at least one receiver is represented by a multiple-input-multiple-output (MIMO) system.
According to a next embodiment, the noise comprises at least one of the following components:
It is also an embodiment that the at least one transmitter and the at least one receiver are connected by at least one cable binder.
The problem stated above is also solved by a device for noise processing comprising a processor unit that is equipped/arranged such that the method as described is executable on said processor unit.
The device may in particular be used to reduce nose comprising in particular interference and/or crosstalk by precoding and/or noise cancellation.
According to an embodiment, said device is a communication device, in particular a Central Office (CO), a Digital Subscriber Line Access Multiplexer (DSLAM) or a Customer Premises Equipment (CPE).
The problem stated supra is further solved by a device for noise processing comprising a precoder and/or a noise canceller processing SNR-data that is provided by a receiver.
It is an embodiment that the device further comprises a S/H-Unit (Sample-Hold-Unit) that obtains the SNR-data.
According to another embodiment, the device is at least a part of a Central Office or it comprises a Central Office.
The problem stated above is also solved by a communication system comprising the device as described herein.
Embodiments of the invention are shown and illustrated in the following figures:
The transmitter comprises a processing stage 201 performing bit operations that are linear over F2 (binary numbers): A framer 205 adds a CRC frame to the data. Next, in a block 206, the data are scrambled and encoded by an error-detection code like the Reed-Solomon-Code (RS). The output of the block 206 is fed to an Interleaver & Tone Ordering unit 207.
The data are further processed by the transmitter in a processing stage 202 performing operations that are linear over C (complex numbers): A QAM Mapper 208 maps the data to symbols (constellations) in the QAM plane and a subsequent block 209 performs an inverse Fast Fourier Transform (FFT) and adds a cyclic prefix (CP).
The output of the block 209 is converted to analog signals by a Digital/Analog Converter DAC 210. Subsequently, data is conveyed via an Analog Channel 211 and at the receiver, data is converted to digital signals by an Analog/Digital Converter ADC 212.
A processing stage 203 at the receiver performs the inverse operations to the processing stage 202 and a processing stage 204 at the receiver performs the inverse operations to processing stage 201.
The output of ADC 212 is fed to a block 213 performing a Time Equalization (TEQ) with skipping of cyclic prefixes (CP). Next, a Fast Fourier Transform FFT is performed in a block 214 and the output is forwarded to a Frequency Domain Equalizer (FEQ) & Slicer 215. In the next stage, a Deinterleaver and Tone Decoder 216 performs the inverse operation to block 207 and a subsequent RS-Decoder & Descrambler 217 performs the inverse operation of block 206. The block 218 performs a CRC check and removes the frame to forwards the (user) data.
The output of the Transmitter K (as well as all other Transmitters 1, 2, . . . , K−1) is forwarded to a Downstream Prefilter 303. The Downstream Prefilter 303 comprises a Precoding Matrix 303, a Coefficient Update Function 305 and a Sample/Hold-Function 306. An output from the Framer 301 conveys a Link State information to the Sample/Hold-Function 306. The Sample/Hold-Function 306 conveys information to the Coefficient Update Function 305 and vice versa. An output from the QAM mapping unit 302 is connected to the Coefficient Update Function 305, the latter being further connected to the Precoding Matrix 303 for updating the matrix' coefficients.
The Transmitters 1 to K as well as the Downstream Prefilter 303 may be located in and/or associated with a Central Office (CO) or a DSLAM.
A channel, that may be at least partially realized by at least one cable binder, is depicted by a DMT-MIMO Channel 307 comprising or represented by a Transfer Matrix 308.
It is to be noted that the DMT-MIMO Channel 307 may comprise the functionalities of boxes 209 to 214 as depicted in
The DMT-MIMO Channel 307 is further connected to several Receivers, i.e. CPE 1 to CPE K. Each receiver comprises a QAM Detection Unit 309 and a CRC Check Unit 310. Additional units similar to the units 216 and 217 shown in
The QAM Detection Unit 309 provides an output signal to the Coefficient Update Function 305 indicating a Decision Error (e.g., a difference between the position received in the QAM plane and the actual constellation). The CRC Check Unit 310 provides SNR-data to the Sample/Hold Function 306.
Hence, in case the SNR-data indicate noise that is above a predetermined threshold, the Sample/Hold Function 306 may freeze the actual coefficients. In case the noise gets worse or it stays above said threshold (for a predetermined period of time), a restart is conducted utilizing the coefficients as stored (frozen).
A channel, that may be at least partially realized by at least one cable binder, is depicted by a DMT-MIMO Channel 403 comprising or represented by a Transfer Matrix 404.
The output of the Transmitter CPE K (as well as all other Transmitters CPE 1, 2, . . . , K−1) is forwarded to the DMT-MIMO Channel 403.
The DMT-MIMO Channel 403 is further connected to an Upstream Canceller 405 comprising a Cancellation Matrix 406, the Upstream Canceller 405 being further connected to several Receivers 1 to K. The Upstream Canceller 405 and the Receivers 1 to K may be located within the Central Office or the DSLAM.
Each receiver 1 to K comprises a QAM Detection Unit 407 and a CRC Check Unit 408. Additional units similar to the units 216 and 217 shown in
The Upstream Canceller 405 comprises the Cancellation Matrix 406, a Coefficient Update Function 409 and a Sample/Hold-Function 410. An output from the Framer 401 conveys a Link State information to the Sample/Hold-Function 410. The Sample/Hold-Function 410 conveys information to the Coefficient Update Function 409 and vice versa. An output from the QAM mapping unit 402 is connected to the Coefficient Update Function 409, the latter being further connected to the Cancellation Matrix 406 for updating the matrix' coefficients.
The QAM Detection Unit 407 provides an output signal to the Coefficient Update Function 409 indicating a Decision Error (e.g., a difference between the position received in the QAM plane and the actual constellation). The CRC Check Unit 408 provides SNR-data to the Sample/Hold Function 410.
Hence, in case the SCR-data indicate noise that is above a predetermined threshold, the Sample/Hold Function 410 may freeze the actual coefficients. In case the noise gets worse or it stays above said threshold (for a predetermined period of time), a restart is conducted utilizing the coefficients as stored (frozen).
This approach allows to utilize SNR-data, in particular a CRC (Cyclic Redundancy Check) code or information for almost instantaneous SNR measurement. Advantageously, the CRC code provides feedback in case Forward Error Correction (FEC) still cannot cope with a failure rate. In other words, the CRC may indicate deterioration of SNR if the SNR is below a certain threshold.
However, other kinds of SNR-data can be utilized as well, e.g., failure detection codes or failure correction codes. In particular, a Reed-Solomon-Code can be used for obtaining SNR-data.
The CRC is a linear code with a fixed block of redundancy bits and arbitrary length message bits. The CRC aims at error detection rather than error correction. Ethernet uses the CRC as its only error detection method (without further fault error correction, FEC) [IEEE 8023].
In DMT-based multicarrier systems an 8 bit CRC over one superframe can be used, i.e., the CRC is updated at an interval of approximately 10 ms to 60 ms depending on FEC parametrization such as interleaver depth and code rate.
Advantageously, the CRC (or even a factor 10 time-averaged statistic thereof) can be suited to provide a reliable trigger for non-stationary noise affecting a loop.
In particular, the block structure according to
A CRC check information provided by CRC Check Units 310 or 408 may be a single bit, e.g., “0” if the check was positive and “1” if the check went wrong. Such single bit may be transferred via an overhead channel from the CPE to the CO.
The CO may use this information by a sliding memory. At the CO, a DSM controller unit performs a continuous update of the Precoding Matrix 304 based on, e.g., knowledge of input symbol (provided by the QAM Mapping Unit 302) and received QAM decision error (provided by the QAM Detection Unit 309). If a number of consecutive CRC detects and/or CRC events reach the DSM controller unit at the CO, the state of the Precoder Matrix 304 is frozen (hold position).
Further to the continuous registration of the CRC information it is advantageous to register the link state information (provided by the Framer CRC 301) to be able to return back from the hold state to the sample state and hence to be able to continue the adaptation of the coefficients in particular with an optimal speed.
It is to be noted that referring to the Central Office (CO) also implies a reference to a portion thereof or a DSLAM or a portion thereof. In addition, the CO as well as the DSLAM may be part of a device comprising the functionality as described herein.
| Number | Date | Country | Kind |
|---|---|---|---|
| 07018853.7 | Sep 2007 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP08/62704 | 9/23/2008 | WO | 00 | 6/4/2010 |
