(19)
(11)EP 1 430 679 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
08.04.2020 Bulletin 2020/15

(21)Application number: 02778202.8

(22)Date of filing:  30.08.2002
(51)International Patent Classification (IPC): 
H04L 1/00(2006.01)
H04L 27/26(2006.01)
H04L 25/03(2006.01)
H04L 5/00(2006.01)
H04L 1/16(2006.01)
(86)International application number:
PCT/US2002/027568
(87)International publication number:
WO 2003/021846 (13.03.2003 Gazette  2003/11)

(54)

DATA COMMUNICATION USING AT LEAST TWO WIDEBAND SUBCHANNELS

DATENKOMMUNIKATION MIT MINDESTENS ZWEI BREITBAND-SUBKANALEN

COMMUNICATION DE DONNEES UTILISANT AU MOINS DEUX SOUS-CANAUX A LARGE BANDE


(84)Designated Contracting States:
AT BE BG CH CY CZ DE DK EE ES FI FR GB GR IE IT LI LU MC NL PT SE SK TR

(30)Priority: 31.08.2001 US 315950 P

(43)Date of publication of application:
23.06.2004 Bulletin 2004/26

(73)Proprietors:
  • Adaptive Networks, Inc.
    Newton, MA 02459 (US)
  • New Visual Corporation
    San Diego, CA 92108 (US)

(72)Inventors:
  • PROPP, Michael, B.
    Newton, MA 02459 (US)
  • SAAB, Khaled
    Randolph, MA 02368-3726 (US)

(74)Representative: Wynne-Jones IP Limited 
2nd Floor, 5210 Valiant Court Gloucester Business Park Gloucester
Gloucestershire GL3 4FE
Gloucestershire GL3 4FE (GB)


(56)References cited: : 
EP-A- 1 041 731
US-A- 3 655 917
US-A- 6 097 763
US-A1- 2001 005 395
WO-A-02/11343
US-A- 5 852 630
US-A- 6 141 356
US-B1- 6 279 158
  
  • PROAKIS: "Digital Communications, Chapter 12, Multichannel and multicarrier systems" 2000, MCGRAW-HILL , NEW YORK, US , XP002488884 * page 680, paragraph 2 * * page 686, last paragraph * * page 687, last paragraph - page 688, paragraph 1 * * page 688, paragraph 3 * * page 689, paragraph 3 * * page 692, paragraph 3 *
  • CZYLWIK: "Comparison between adaptive OFDM and single carrier modulation with frequency domain equalization" IEEE VEHICULAR TECHNOLOGY CONFERENCE, vol. 2, 4 May 1997 (1997-05-04), pages 865-869, XP010228966 New York, US ISBN: 978-0-7803-3659-9
  • MOROSI ET AL.: "A rate adaptive bit-loading algorithm for a DMT modulation system for in-building power-line communications" IEEE GLOBAL TELECOMMUNICATIONS CONFERENCE, vol. 1, 28 November 2005 (2005-11-28), pages 403-407, XP010879112 Piscataway, US ISBN: 978-0-7803-9414-8
  
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description

TECHNICAL FIELD



[0001] This document relates to communicating data using wideband communications.

BACKGROUND



[0002] In data communications, wideband transmission may be used. However, the received signal may be impaired by noise and frequency-dependent channel attenuation. For example, an entire portion of the transmitted signal may fall into an attenuation null and be severely attenuated. In addition, the intersymbol interference (ISI) and/or interchannel interference (ICI) could degrade the signal, causing a high bit error rate and rendering an error correction engine less impactful.

[0003] The book PROAKIS: "Digital Communications", Chapter 12, Multichannel and multicarrier systems, 2000, MCGRAW-HILL, NEW YORK, US, discloses methods and systems for communicating using multiple channels.

[0004] Other methods and systems for communicating using multiple channels are disclosed in US2001/005395 A1 (JIN) 28 June 2001 (2001-06-28) and in US6279158 B1 (GEILE ET AL.) 21 August 2001 (2001-08-21).

SUMMARY



[0005] The invention is defined by the appended claims. All the embodiments that are not defined in the appended claims are regarded as pure examples for better understanding the claimed invention.

[0006] In one general aspect, communicating data includes identifying a static component of a transmission medium for wideband communications having at least two subchannels. A function of noise or attenuation versus frequency is determined for the static component. Wideband modulation schemes are selected for the subchannels based on the function determined and a data signal is communicated over the subchannels. The transmission medium includes a medium other than a copper telephone wire between and including a carrier facility and a termination of the copper telephone wire at an access point at a user premise and copper telephone wire in a cable riser in multi-tenant units and multi-dwelling buildings.

[0007] Implementations may include one or more of the following features. For example, the transmission medium may include a copper telephone wire between the access point at the user premise and a termination of the copper telephone wire within the user premise, a power line, a cable line, a telephone wire other than a copper telephone wire, and/or a radio frequency medium.

[0008] The selected wideband modulation schemes may be independently applied to each of the subchannels. The selected wideband modulation schemes may be applied using separate modulators for each of the subchannels.

[0009] In one implementation, the static component may be independently identified on each of the subchannels. The function of noise or attenuation versus frequency may be determined for the static component of each of the subchannels. A wideband modulation scheme may be selected to optimize throughput on each of the subchannels based on the function determined on each of the subchannels.

[0010] A measurement of the noise or attenuation versus frequency may be taken at a time interval other than a frame-by-frame time interval and the function of the noise or the attenuation versus the frequency may be determined for the static component based on the measurement. The wideband modulation scheme may be selected at a time of the measurement based on the function determined. The function of the noise or the attenuation versus the frequency may be determined by averaging a function of noise or attenuation versus frequency for the static component over a period of time.

[0011] The static component may include noise as a relatively static function of frequency, attenuation as a relatively static function of frequency, or noise and attenuation as a relatively static function of frequency.

[0012] A dynamic component of the transmission medium for the wideband communications may be identified. An adaptive filter may be created based on the dynamic component and the adaptive filter may be applied to the communicated data signal at a receiver. The adaptive filter may be applied to the communicated data signal on each of the subchannels for the wideband communication. The adaptive filter may include a noise filter, a channel filter, and/or a frequency-domain equalizer.

[0013] The static component may be static relative to the dynamic component. The dynamic component may include noise that varies as a function of frequency, attenuation that varies as a function of frequency, or noise and attenuation that vary as a function of frequency.

[0014] A channel filter may be created by averaging elements of the channel filter based on the static component and the channel filter may be applied to the communicated data signal at a receiver. A noise filter may be created by averaging elements of the noise filter based on the static component and the noise filter may be applied to the communicated data signal at a receiver. A frequency-domain equalizer may be created by averaging elements of the frequency-domain equalizer based on the static component and the frequency-domain equalizer may be applied to the communicated data signal at a receiver.

[0015] A filter may be created by averaging the function of noise versus the frequency and the filter may be applied to the communicated data signal at a receiver. A filter may be created by averaging the function of attenuation versus the frequency and the filter may be applied to the communicated data signal at a receiver. A filter may be created by averaging the function of noise and attenuation versus the frequency and the filter may be applied to the communicated data signal at a receiver.

[0016] Independent demodulation functions may be applied for at least two of the subchannels. Separate demodulators may be used to apply the demodulation functions for the subchannels. Separate demodulation algorithms may be used to apply the demodulation functions for the subchannels.

[0017] Modulation schemes may be selected for at least two subchannels to optimize orthogonality between each of the subchannels. The modulation schemes may include offset quadrature amplitude modulation (OQAM) schemes.

[0018] A forward error correction code may be selected based on a length of a frame and a rate of retransmission. An error detection code may be selected to determine a level of error detection and a rate of retransmission.

[0019] An acknowledgement for a retransmission of previously transmitted data communicated across the transmission medium may be attached to new data. A data block that requires retransmission also may be attached to the new data. An acknowledgement for correct reception of previously transmitted data communicated across the transmission medium may be attached to new data.

[0020] In another general aspect, communicating data includes in a first transmitter on a first channel, detecting a state of communications by a second transmitter over a second channel and synchronizing transmissions by the first transmitter with transmissions by the second transmitter based on the state detected. The first channel and the second channel may be included in a transmission medium other than a copper telephone wire between and including a carrier facility and a termination of the copper telephone wire at an access point at a user premise and copper telephone wire in a cable riser in multi-tenant units and multi-dwelling buildings.

[0021] Implementations may include one or more of the following features. For example, the first channel may include a first wire and the second channel may include a second wire that differs from the first wire. The transmission medium may include a copper telephone wire between the access point at the user premise and a termination of the copper telephone wire within the user premise, a power line, a cable line, a telephone wire other than a copper telephone wire, and/or a radio frequency medium. The state of communications may include a receiving state and/or a transmitting state.

[0022] In another general aspect, communicating data includes identifying frequencies in a wideband communication channel that require reduced emissions and applying a digital frequency-domain filter to a transmitted signal to reduce the emissions of the identified frequencies. The wideband communication channel is over a transmission medium that includes a medium other than a copper telephone wire between and including a carrier facility and a termination of the copper telephone wire at an access point at a user premise and copper telephone wire in a cable riser in multi-tenant units and multi-dwelling buildings.

[0023] Implementations may include one or more of the following features. For example, the transmission medium may include a copper telephone wire between the access point at the user premise and a termination of the copper telephone wire within the user premise, a power line, a cable line, a telephone wire other than a copper telephone wire, and/or a radio frequency medium. The identified frequencies may be identified by a regulatory agency for reduced emissions. The identified frequencies may include frequencies that interfere with other devices or communications.

[0024] These general and specific aspects may be implemented using a system, a method, or a computer program, or any combination of systems, methods, and computer programs.

[0025] Other features will be apparent from the description and drawings, and from the claims.

DESCRIPTION OF DRAWINGS



[0026] 

Fig. 1 is a flow chart of an exemplary process for communicating data.

Fig. 2 is a graph of an exemplary curve of attenuation versus frequency with a single channel of wideband modulation. Fig. 3 is a graph of an exemplary curve of attenuation versus frequency with multiple channels of wideband modulation.

Fig. 4 is a graph of an exemplary curve of noise amplitude versus time with a constant perceived signal-to-noise ratio (SNR).

Fig. 5 is a graph of an exemplary curve of noise amplitude versus time with a dynamically adapted SNR.

Fig. 6 is a table illustrating throughput versus dynamically varying bit error rates (BERs) for an exemplary set of parameters.

Fig. 7 is a block diagram of an exemplary transmit path.

Fig. 8 is a block diagram of an exemplary receive path. Fig. 9 is a flow chart of an exemplary process for communicating data.

Fig. 10 is a flow chart of an exemplary process for communicating data.

Like reference symbols in the various drawings may indicate like elements.

DETAILED DESCRIPTION



[0027] For certain media, each of the noise and attenuation, as functions of frequency, may exhibit both a static and a dynamic component. The static component typically refers to a predictable or average function that is independent or relatively ' independent of time and possibly location (given the same cable parameters, such as type and length of cable at each location). For example, the typical signal attenuation on a transmission medium (e.g., a power line or a telephone wire) of a static configuration and long length (for example, several thousand feet) is an increasing function of frequency. This average, static function of signal attenuation versus frequency may be predictable based upon models of the particular cable. Similarly, there may be typical time-independent or static values of background noise as a function of frequency. The static component may include noise and/or attenuation as a relatively static function of frequency. The dynamic component generally refers to variations in both time and location of noise or attenuation versus frequency due to, for example, different devices or taps being connected to the wire at different times or locations. For example, for a measurement of noise and signal attenuation of a particular cable having a given type and length, the static function may be a useful estimate of future noise and signal attenuation that does not require further measurements; it also may correspond to an estimate of a time-averaged measurement.

[0028] A transmission medium may include a power line, a telephone line, a cable line, a digital subscriber line (DSL), an integrated services digital network (ISDN) line, a radio frequency (RF) medium, and/or other transmission media. In one exemplary implementation, a transmission medium may be identified in different segments and span from within a carrier's facility to access points outside of buildings and dwellings, to a cable riser in multi-tenant units and multi-dwelling buildings, and to one or more termination points within the buildings and dwellings.

[0029] In data communication systems, it may be desirable to increase or maximize the data throughput for signaling between particular locations. Two approaches to modulation for noisy media with frequency-dependent attenuation include (1) multicarrier modulation such as, for example, orthogonal frequency-division multiplexing (OFDM) or discrete multitone (DMT), which consists of many narrowband subchannels, or (2) wideband modulation. Narrowband modulation includes several different defining characteristics. In one defining characteristic, narrowband describes modulated channels where the frequency bandwidth includes small relative changes of attenuation versus frequency such that the attenuation versus frequency over the channel appears relatively flat. In another defining characteristic, narrowband describes modulated channels having a bandwidth that is relatively small compared to the carrier frequency. More specifically, in narrowband, one symbol interval occupies a large number of cycles of the carrier (e.g., approximately greater than 50 cycles). In yet another defining characteristic, narrowband describes transmissions with a channel bandwidth having a frequency typically ranging from 50Hz to 64kHz, which generally may be used for carrying voice information.

[0030] Wideband modulation also includes several different defining characteristics. In one defining characteristic, wideband describes modulated channels where the frequency bandwidth includes large relative changes of attenuation versus frequency such that the attenuation versus frequency over the channel bandwidth varies and does not appear relatively flat. In another defining characteristic, wideband describes modulated channels having a bandwidth that is wide relative to the carrier frequency. More specifically, in wideband, one symbol interval occupies a small number of cycles of the carrier (e.g., approximately less than 50 cycles). In yet another defining characteristic, wideband describes transmissions with a channel bandwidth greater than 64kHz.

[0031] OFDM or DMT modulation typically use many subchannels such that each subchannel has a band that is sufficiently narrow to minimize the variation of attenuation or noise over the frequency band of that subchannel. The aggregate of all the subchannels forms a symbol, which is sufficiently long in length to accommodate the slow signaling rates of the narrowband subchannels (where the signaling rate of each subchannel is slow as it is narrowband). If the subchannel signal-to-noise ratio (SNR) can be calculated, then the optimal modulation for that subchannel can be determined. The SNR for any particular subchannel generally is made available to train the transmitter how to load that subchannel. In order to maximize the data throughput, the transmitter typically is updated periodically with the SNR or similar information for each subchannel. Such updates typically are performed using transmissions back from the receiver at a rate that exceeds the time variation of the SNR. For example, if a time-variation in the noise or attenuation impairs the performance of a particular subchannel, the data on that subchannel may be subject to a high error rate which may compromise the ability to recover the data in the symbol until the transmitter can be updated with the lower SNR for that channel. Consequently, in the case of significant time-variation of noise and/or attenuation characteristics, the performance of OFDM or DMT may be compromised by the long symbol lengths and the requirements of transmitter training.

[0032] OFDM or DMT may be used when significant variations of attenuation as a function of frequency (i.e., a high frequency-selectivity of the channel) would result in intersymbol interference (ISI) spread over large numbers of symbols when using high data rates. With single carrier wideband modulation, a time-domain equalizer may not be feasible for such a channel due to the large number of taps that are necessary for the single carrier wideband modulation. By contrast, with frequency-domain equalization (FDE), single carrier wideband modulation may become feasible under such conditions. With wideband modulation using FDE, the energy of an individual symbol generally is distributed over the whole available frequency band and consequently, narrowband noise or narrowband notches in the channel transfer function have only a small impact on the error rate. Such impairments may affect OFDM or DMT without training, but with training may be overcome. However, rapid time variations in the frequencies of narrowband noise or notches in frequency-selective channels typically do not affect the performance of wideband modulation with FDE, but may affect OFDM or DMT even with training as previously noted.

[0033] To optimize throughput at a high SNR, a modulation scheme typically is selected with the largest number of bits per symbol providing an acceptable error rate. For example, in a channel where the attenuation and/or noise is a static function of frequency and the attenuation generally increases linearly with frequency, consequently the SNR decreases as a function of frequency. In this example, if single-carrier wideband modulation is used, the selection of the modulation scheme with the consequent number of bits per symbol and the resulting throughput becomes a compromise. In this example, if OFDM or DMT are used, the modulation scheme for each subchannel may be optimized for high throughput and the number of bits per symbol may decrease as a function of frequency for each subchannel.

[0034] However, when the noise and attenuation include more than just static functions of frequency and also are dynamic and highly sensitive to frequency selection, then the performance of OFDM or DMT modulation may become impaired for the reasons noted above. If the modulation schemes for each subchannel were selected based only on the static function that did not account for the dynamics of the noise and attenuation, then the available bandwidth for using OFDM or DMT may be compromised by not being able to take advantage of the full available bandwidth. The available bandwidth may compromised as the performance of the selected OFDM or DMT modulation becomes highly susceptible to further noise and/or attenuation as a dynamic function of frequency. The noise and/or attenuation as a dynamic function of frequency reduces the SNR below the threshold needed for the selected modulation scheme, which was selected based only on the static function, to remain operative in the higher frequency subchannels.

[0035] For media with noise and attenuation that vary with frequency as both static and dynamic functions, one optimal approach may include two elements. First, wideband modulation may be used to handle noise and/or attenuation as a dynamic and highly selective function of frequency to increase or maximize available bandwidth. Second, a modulation scheme with as large as possible number of bits per symbol may be used to maximize throughput for noise and/or attenuation as a static function of frequency. This may be achieved by wideband modulation not with a single carrier but with a small number of carriers, for example, approximately 2 to 12, rather than the approximately 100 to 1000 used in OFDM or DMT. For example, in a 10 MHz channel, each carrier could be spaced a few megahertz apart rather than a few kilohertz apart for OFDM and DMT. Improved or optimal modulation for each of these few carriers may be selected to achieve improved or optimal throughput given the SNR due to the static function over the bandwidth of that modulated carrier. Due to the modulation being wideband, there typically is immunity to frequency-selective noise and attenuation and the ability to optimally handle rapid time variation of such frequency-selective noise and attenuation and consequently the availability of much wider usable bandwidth.

[0036] Fig. 1 illustrates a process 100 for communicating data using wideband modulation having at least two subchannels. Process 100 typically includes identifying a static component of a transmission medium for wideband communications having at least two subchannels (step 110). A function of noise and/or attenuation versus frequency is determined for the static component (step 120). A wideband modulation scheme is selected for each of the subchannels based on the function determined (step 130), the wideband modulation schemes are applied to the subchannels (step 140), and a data signal is communicated over the subchannels (step 145). A data signal may include data, data gaps, noise, data frames, inter-frame gaps, and preambles.

[0037] In one exemplary implementation, the static component of the transmission medium for the wideband communications may be identified prior to dividing the transmission medium into the at least two wideband subchannels (step 110), and a function of noise and/or attenuation versus frequency is determined for the static component (step 120). Then a wideband modulation scheme is selected independently for the subchannels based on the function determined (step 130), applied to the subchannels (step 140), and a data signal is communicated over the subchannels (step 145).

[0038] In another exemplary implementation, the transmission medium for the wideband communications may be separated into at least two wideband subchannels before the static component is identified independently for the resulting subchannels (step 110). In this implementation, functions of noise and/or attenuation versus frequency are determined independently for the static component of subchannels (step 120) and then wideband modulation schemes are selected independently for the subchannels based on the determined functions (step 130), applied to the subchannels (step 140), and a data signal is communicated over the subchannels (step 145).

[0039] In another exemplary implementation, the function of noise and/or attenuation versus frequency may be determined based on an infrequent measurement of the noise and/or attenuation versus frequency over the usable bandwidth (step 120). The wideband modulation scheme may be selected at the time of the infrequent measurement (step 130), applied to the subchannels (step 140), and a data signal is communicated over the subchannels (step 145). The infrequent measurement includes measurements that are taken on a non-regular basis at a time interval greater than a frame-by-frame interval. Additionally or alternatively, the function of noise and/or attenuation versus frequency may be determined by averaging the function of noise and/or attenuation versus frequency for the static component over a period of time.

[0040] In one implementation, wideband modulation schemes may be selected independently for several or each of the subchannels to improve or optimize orthogonality between the subchannels (step 130). Each such modulated carrier (individual subchannel) may be made non-interfering or orthogonal to the others (for example, with each carrier being an integer multiple of cycles over a symbol period) and resultant short symbols may result from the combination of all such individual channels. This approach allows the modulation schemes for each of the carriers to be improved or optimized, for example, to achieve an increased or highest total throughput at a required error rate based upon the static functions of noise and attenuation versus frequency. The improvement or optimization also may include the constraint of maximum total power or power spectral density. One exemplary modulation scheme to optimize orthogonality between the subchannels includes offset quadrature amplitude modulation (OQAM).

[0041] In one exemplary implementation, the wideband modulation schemes may be independently selected for the subchannels (step 130) and the selected modulations schemes may be applied to multiple data frames (step 140). When the wideband modulation schemes are selected based on the static component (step 130), then new modulation schemes may not be selected until a change is recognized with respect to the static component. Thus, once the modulation schemes are selected (step 130), those same modulation schemes maybe applied to multiple data frames (step 140).

[0042] A dynamic component of the transmission medium for wideband communications also may be identified (step 150). Typically, the dynamic component includes noise and/or attenuation that varies as a function of frequency.

[0043] A filter may be created or applied to the communicated data signal at a receiver based on the dynamic and/or the static components (step 160). For example, in one implementation, the dynamic component may be used to create and apply an adaptive filter to the communicated data signal. The adaptive filter may include a channel filter, a noise filter, and/or a frequency-domain equalizer. The adaptive filter may be created by averaging the function of noise and/or attenuation versus frequency. Additionally or alternatively, the adaptive filter may be created by taking an average of the elements of the adaptive filters over a period of time. This approach may optimally handle channels with dynamic functions of noise and attenuation versus frequency that may be highly frequency-selective by preserving short symbol times with the energy of each individual modulated carrier distributed over a wide spectrum.

[0044] In another implementation, the static component may be used to create and apply a filter to the communicated data signal at a receiver (step 160). The filter may include a channel filter, a noise filter, and/or a frequency-domain equalizer. The filter may be created by averaging the function of noise and/or attenuation versus frequency. Additionally or alternatively, the filter may be created by taking an average of the elements of the filters over a period of time.

[0045] Independent demodulation functions may be applied to each of several subchannels based on the selected wideband modulation schemes (step 170). Applying independent demodulation functions may include using separate demodulators for each subchannels and/or separate demodulation algorithms for each of the subchannels.

[0046] Fig. 2 illustrates a graph 200 of the steep curve 210 of attenuation increasing with frequency resulting in a higher attenuation (HIGH) at higher frequencies. Systems that experience similar characteristics include long distance telephone wiring, which may encounter attenuation versus frequency that is a steep curve increasing with frequency. As illustrated, using wideband modulation as represented by power spectral density (PSD) curve 220 with a single channel 230 leads to an inefficient match of the modulation 220 to the steepness of the attenuation curve 210. For example, the attenuation experienced by different frequency channels generally tends to vary in this manner based on the length of the wire.

[0047] Fig. 3 illustrates a graph 300 showing that improved communications may be achieved by choosing a few individual channels 330A, 330B, 330C, 330D to preserve wideband modulation and by varying modulation schemes 320A, 320B, 320C, 320D according to the predicted attenuation curve 310 of the corresponding individual channel (e.g., the most complex modulation may be applied to the lowest frequency channel), while preserving the wideband characteristics of the modulation in order to handle the dynamic noise characteristics of the transmission medium (e.g., telephone wiring) over the widest possible bandwidth, thereby achieving a highest possible throughput.

[0048] Fig. 4 illustrates a graph 400 of dynamic noise amplitude over time with a constant perceived SNR 410, which may result in lower speed data communications.

[0049] Fig. 5 illustrates a graph 500 of the noise amplitude over time with a perceived SNR 510 that is dynamically adapted by identifying the dynamic component (e.g., the dynamic noise component), creating and applying an adaptive filter based on the dynamic component. This "on-the-fly" frequency-domain equalization improves the speed of the data communications.

[0050] Referring back to Fig. 1, process 100 may further include selecting error correction and/or error detection codes (step 180). Forward error correcting codes typically are applied to the transmitted data to achieve a particular bit error rate. The forward error correction (FEC) is typically selected to achieve an error rate that can handle worst-case channel conditions. However, in dynamic channels, there may be times when the correcting bounds of the FEC are exceeded. With the short symbol lengths and low latency processing enabled by this approach, the data may be transmitted into short frames composed of blocks incorporating both forward error correction and error detection, which may be acknowledged by the receiver using the error-detecting code to determine which blocks were not corrected by the FEC. This allows for a low-level link protocol that may achieve a lower bit error rate through retransmission by the transmitter of any blocks that were unable to be corrected by the FEC, as determined by the acknowledgements received from the receiver.

[0051] In addition, if a retransmission is necessary (step 185), an acknowledgement for a retransmission of previously transmitted data may be attached to new data (step 195). When a retransmission is required, the block of data that requires retransmission to the opposite node also may be attached to the new data along with the acknowledgement for the data transmitted by the opposite node. Using "piggyback" acknowledgements enables data to be sent in both directions at higher efficiencies. More specifically, for example, higher efficiencies may be realized because the number of different transmissions between a sender and receiver are reduced by attaching to new data both acknowledgements for retransmissions and acknowledgements that the data received was correct. The combination of (i) piggyback acknowledgements and (ii) short frame lengths and low-latency processing made possible by using multiple wideband channels enables a communication system incorporating these features to provide low latency communications to an application. If an acknowledgement for a retransmission of previously transmitted data is not necessary, then an acknowledgement indicating that the data transmitted was correct may be attached to and transmitted with new data (step 190).

[0052] Fig. 6 illustrates an exemplary table 600 illustrating throughput versus dynamically varying bit error rates (BERs) for an exemplary set of parameters. The parameters may include a raw bit rate (e.g., 100Mbps), which is the speed at which bits are transmitted over a communications path, that is typically expressed in bits per second. The FEC overhead (e.g., 2:1) includes additional redundant portions of a frame that is added by the FEC code and that is expressed as a ratio of total information transmitted to actual information presented for delivery. The propagation delay (e.g., 11.43 µs) includes a time for transit of a signal from one point on a circuit, link, network, or system to the other point. The block size (header, body, CRC) (e.g., 2, 64, 2) includes the size of a data block prior to FEC encoding that is typically expressed in bytes. The blocks per frame (e.g., 32) includes the number of data blocks contained within a single frame. The BER includes the ratio of the mean number of bit errors in any given interval to the total number of bits in that interval. The block error rate includes the ratio of the mean number of block errors in any given interval to the total number of blocks in that interval. The throughput includes the average rate at which application information is communicated during a specified period of time. Throughput may be used as a figure of merit for a system as it excludes the overhead introduced by the system such as error control coding and protocol overhead. The post-FEC BER includes a bit error rate as applied to the bit stream received after FEC decoding has been performed.

[0053] For example, as the noise may vary significantly over time, transmitter training on a fast enough time scale may not be relied upon. Under a conventional approach, a modulation and FEC must be used for the worst case to always deliver, for example, a BER of 10-9 or less, which compromises throughput. However, using short frames and low-latency processing of both the FDE and the FEC, dynamic fluctuations in the post-FEC BER, can still deliver a final BER of 10-9 or less while using a more complex modulation and less coding overhead for higher throughput, through a low-latency ARQ using piggyback acknowledgements. Table 600 illustrates the maintenance of the delivered throughput under dynamically-varying BERs for the given set of parameters.

[0054] Fig. 7 illustrates an exemplary transmit path 700 used to implement the features described above with respect to Figs. 1-6. Transmit path 700 typically includes a processor 710, error correction and/or error detection encoders 720, one or more modulators 730, a frequency domain shaping filter 740, a digital analog converter (DAC) 750, a low pass filter 760, and a line driver 770.

[0055] Processor 710 interfaces the data communications system to the user. It may run the communications protocols, for example, segmentation and reassembly, packetizing data, retransmitting blocks of data that were received incorrectly as determined by the error detecting code, and sending acknowledgements (ACKs or NAKs), such as piggyback ACKs or NAKs to determine whether blocks of data were received correctly or incorrectly. Depending on the application and on the processor, the task list may include any functionality deemed suitable to be performed by a CPU.

[0056] Error correction and/or detection encoders 720 performs the data encoding. For example, it may apply an error detecting code to a block of data, followed by a forward error correcting (FEC) code to words of data within the block. The complexity of the FEC code can be dependent on the expected quality of the channel or subchannels (i.e., if the channel or subchannel is characterized by a high signal-to-noise ratio (SNR), a simple code, if any, could suffice), the coding overhead, and the desired Symbol Error Rate (SER) or Bit Error Rate (BER) at the receiver.

[0057] The input from the processor 710 to the encoders 720 may be one or more bit streams. The stream(s) could be encoded either as one stream or segmented as separate independent streams and then encoded for each subchannel to be then modulated by the modulators 730. Where only one bit stream is encoded, segmentation into multiple streams for each subchannel may be performed before sending that subchannel to the modulators 730. More specifically, the segmentation into multiple streams could occur before applying the error detecting code(s), after applying the error detecting code(s) and before applying the forward error correcting code(s), or after applying both the error detecting code(s) and the forward error correcting code(s).

[0058] The input data streams are modulated by the modulators 730 for each subchannel with the modulation constellation/complexity determined for each subchannel (e.g., BPSK, QAM, QAM 8, QAM 16, QAM 32, QAM 64, QAM 128, QAM 256, QAM 1024). The modulated streams are then combined/added together and sent to the frequency-domain shaping filter 740.

[0059] The frequency-domain shaping filter 740 can be used to reduce emissions in particular parts of the frequency band. For example, the transmitted signal can be notched such that it will not interfere with the amateur radio (HAM) bands, amplitude modulation (AM) bands, and even with other modems. The filter can be used to adjust the PSD of the transmitted signal.

[0060] DAC 750 converts the digital signal to an analog waveform. Low pass filter 760 smoothes the DAC output and removes the undesired high frequency bands. Line driver 770 boost the signal power to an appropriate level for transmission.

[0061] Fig. 8 illustrates an exemplary receive path 800 used to implement the features described above with respect to Figs. 1-6. Receive path 800 typically includes a low pass filter 810, an analog digital converter (ADC) 820, a synchronizer 830, an equalizer 840, one or more demodulators 850, error correction and/or error detection decoders 860, and a processor 870.

[0062] Low pass filter 870 acts as an anti-aliasing filter. ADC 820 converts the incoming analog signal to a digital signal. Synchronizer 830 determines the beginning of a received data frame. In addition, it 830 can be used to maintain proper bit timing at the receiver location. To determine if a frame is received, the incoming signal can be correlated to a predefined and stored transmitted preamble. Decision logic analyzes the correlation results and determines if the received sequence is a match to the transmitted preamble. This process can be repeated until a match is found.

[0063] Equalizer 840, for example an adaptive frequency-domain equalizer constructed from both a channel and noise filter, may be used to recover the transmitted data over a noisy transmission channel with frequency-domain attenuation. The equalizer may be designed to analyze the channel and adapt to the channel noise and to the channel attenuation/distortion. The equalizer reduces or eliminates noise and corrects for the channel attenuation/distortion to reduce or eliminate inter-symbol interference (ISI) and/or inter-channel interference (ICI). In one implementation, equalizer 840 may be constructed using two filters: a noise filter and a channel filter. The combination of the two filters can be used to generate a filter such as a Wiener filter or another filter as an alternative, or as capable of being used in combination, or as an optimal filter. Constructing the frequency-domain equalization channel and noise filter may include using one or more samples of the noise in the inter frame gap (IFG), the received preamble, and the transmitted preamble.

[0064] Demodulators 850 receives the input stream, subdivides or segments the incoming stream into subchannels, and then demodulates each subchannel. The demodulators 850 have knowledge of the modulation constellation for each subchannel.

[0065] Error correction and detection decoders 860 used is described above with regard to the transmitter in Fig. 7. The processing of the decoding and reassembly of the streams is in the required order and sequence as determined by the transmitter construction.

[0066] Processor 870 interfaces the data communications system to the user and can also run the communications protocols, including whether acknowledgements (ACKs or NAKs) should be sent as to whether blocks of data were received correctly or incorrectly. Depending on the application and on the processor, the task list could include any functionality deemed suitable to be performed by a CPU.

[0067] Once the data has been received, demodulated, decoded, and reassembled from the received blocks into application data it will be sent to the user.

[0068] Fig. 9 illustrates a process 900 for communicating data. Process 900 typically includes using a first transmitter on a first channel to detect a state of communications by a second transmitter over a second channel (step 910). Transmissions by the first transmitter are synchronized with transmissions by the second transmitter based on the state detected (step 920). The first channel may include a first wire and the second channel may include a second wire that differs from the first wire. The state of communications includes a receiving state and a transmitting state. In general, process 900 may be applied when the frame sizes are fixed.

[0069] More specifically, for media with high crosstalk such as telephone wiring, each transmitter may synchronize its transmission to that of other wire pairs by first detecting transmissions on other channels at its end of the loop and then synchronizing its transmission to coincide in time with such transmissions (to reduce or eliminate the effect of crosstalk on reception that would occur if reception coincided with transmission on other pairs).

[0070] Fig. 10 illustrates a process 1000 for communicating data that includes identifying frequencies in a wideband communication channel that require reduced emissions (step 1010). A digital frequency-domain filter may be applied to a transmitted signal to reduce the emissions of the identified frequencies (step 1020). In one exemplary implementation, the frequencies that require reduced emissions may be identified by a regulatory agency, such as, the Federal Communications Commission (FCC). Additionally or alternatively, the frequencies that require reduced emissions may be identified so as to reduce interference with other devices and communications.

[0071] More specifically, as wideband modulation can handle notches, multiple, precisely-located notches may be applied to the transmitted signal by digital frequency-domain filtering to reduce emissions in particular parts of the frequency band and thus to minimize interference with multiple existing RF transmissions, such as amateur radio bands.

[0072] The described systems, methods, and techniques may be implemented in digital electronic and/or analog circuitry, computer hardware, firmware, software, or in combinations of these elements. Apparatus embodying these techniques may include appropriate input and output devices, a computer processor, and a computer program product tangibly embodied in a machine-readable storage device for execution by a programmable processor. A process embodying these techniques may be performed by a programmable processor executing a program of instructions to perform desired functions by operating on input data and generating appropriate output. The techniques may be implemented in one or more computer programs that are executable on a programmable system including at least one programmable processor coupled to receive data and instructions from, and to transmit data and instructions to, a data storage system, at least one input device, and at least one output device. Each computer program may be implemented in a high-level procedural or object-oriented programming language, or in assembly or machine language if desired; and in any case, the language may be a compiled or interpreted language. Suitable processors include, by way of example, both general and special purpose microprocessors. Generally, a processor will receive instructions and data from a read-only memory and/or a random access memory. Storage devices suitable for tangibly embodying computer program instructions and data include all forms of non-volatile memory, including by way of example semiconductor memory devices, such as Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), and flash memory devices; magnetic disks such as internal hard disks and removable disks; magneto-optical disks; and Compact Disc Read-Only Memory (CD-ROM). Any of the foregoing may be supplemented by, or incorporated in, specially-designed ASICs (application-specific integrated circuits).


Claims

1. A method (100) for communicating data over a transmission medium with noise and attenuation that exhibit, as functions of frequency, a static component, which is defined as a predictable or average function that is independent of time, and a dynamic component, varying in both time and location, a transmission channel of the transmission medium being divided into at least two wideband subchannels, wherein one symbol transmitted over a wideband subchannel occupies less than 50 cycles of a transmission carrier, wherein the data is transmitted over the transmission medium,
the method comprising:

identifying the static component of the transmission medium (110) for wideband communications having at least two subchannels;

determining a function of noise and/or attenuation versus frequency for the static component (120);

selecting wideband modulation schemes for the subchannels (130), which are made non-interfering or orthogonal to the other subchannels, to achieve an increased or highest total throughput at a required error rate, based upon the static functions of noise and attenuation versus frequency;

transmitting a data signal over the subchannels (145);

identifying a dynamic component of the transmission medium for the wideband communications (150);

creating an adaptive filter based on the dynamic component (160); and

applying the adaptive filter to a received data signal at a receiver (160).


 
2. The method as in claim 1, further comprising independently applying the selected wideband modulation schemes to each of the wideband subchannels (170).
 
3. The method as in claim 2, wherein independently applying the selected wideband modulation schemes (170) includes using separate modulators for each of the wideband subchannels to apply the selected modulation schemes.
 
4. The method as in claim 1, wherein:

identifying the static component includes independently identifying the static component on each of the wideband subchannels;

determining the function includes determining the function of noise or

attenuation versus frequency (120) for the static component of each of the wideband subchannels; and

selecting the wideband modulation scheme (170) includes selecting a wideband modulation scheme (170) to optimize throughput on each of the wideband subchannels based on the function determined on each of the wideband subchannels.


 
5. The method as in claim 1, wherein:

determining the function includes taking a measurement of the noise or

attenuation versus frequency (120) during a time interval other than a frame-by-frame time interval and determining the function of the noise or the attenuation versus the frequency (120) for the static component based on the measurement; and

selecting the wideband modulation scheme includes selecting the
wideband modulation scheme at a time of the measurement based on the function determined.


 
6. The method as in claim 1, wherein determining the function includes averaging a function of noise or attenuation versus frequency (200) for the static component over a period of time.
 
7. A computer program comprising code segments for causing a computer (710) to carry out the method according to anyone of the preceding claims.
 
8. A system for communicating data over a transmission medium with noise and attenuation that exhibit, as functions of frequency, a static component, which is defined as a predictable or average function that is independent of time, and a dynamic component, varying in both time and location, a transmission channel of the transmission medium being divided into at least two wideband subchannels, wherein one symbol transmitted over a wideband subchannel occupies less than 50 cycles of a transmission carrier, wherein the data is transmitted over the transmission medium,
the system comprising:

a processor module that is structured and arranged to

identify the static component of the transmission medium (110) for wideband communications having at least two subchannels,

determine a function of noise and/or attenuation versus frequency for the static component (120),

select wideband modulation schemes for the subchannels (130), which are made non-interfering or orthogonal to the other subchannels, to achieve an increased or highest total throughput at a required error rate, based upon the static functions of noise and attenuation versus frequency;

identify a dynamic component of the transmission medium for the wideband communications (150);

a transmit module that is structured and arranged to communicate a data signal over

the subchannels (145); and

a receiver including an adaptive filter that is based on the dynamic component and applied to a received data signal at a receiver (160).


 
9. The system of claim 8, further comprising a modulator module having one or more modulators that are structured and arranged to independently apply the selected wideband modulation schemes (170) to each of the wideband subchannels.
 
10. The system of claim 9, wherein the modulator module uses separate modulators for each of the subchannels to apply the selected modulation schemes (170).
 
11. The system of claim 8, wherein the processor is structured and arranged to: independently identify the static component on each of the subchannels;
determine the function of noise or attenuation versus frequency (120) for the static component of each of the subchannels; and
select a wideband modulation scheme (170) to optimize throughput on each of the subchannels based on the function determined on each of the
subchannels.
 
12. The system of claim 8, wherein the processor is structured and arranged to:

take a measurement of the noise or attenuation versus frequency (120) during a time interval other than a frame-by-frame time interval and determine the function of the noise or the attenuation versus the frequency (120) for the static component based on the measurement; and

select the wideband modulation scheme (170) at a time of the measurement based on the function determined.


 
13. The system of claim 8, wherein the processor is structured and arranged to average a function of noise or attenuation versus frequency (200) for the static component over a period of time.
 
14. The system of claim 8, wherein the adaptive filter (160) is applied to the communicated data signal on each of the subchannels.
 
15. The system of claim 8, wherein the adaptive filter (160) includes a noise filter, channel filter, or a frequency-domain equalizer.
 


Ansprüche

1. Verfahren (100) zum Übertragen von Daten über ein Übertragungsmedium mit Rauschen und einer Dämpfung, die als Funktionen der Frequenz eine statische Komponente, die als vorhersagbare oder durchschnittliche Funktion festgelegt ist, die zeitunabhängig ist, und eine dynamische Komponente, die sowohl bezüglich der Zeit als auch des Orts variiert, aufweisen, wobei ein Übertragungskanal des Übertragungsmediums in mindestens zwei Breitbandteilkanäle aufgeteilt ist, wobei ein Symbol, das über einen Breitbandteilkanal übertragen wird, weniger als 50 Zyklen eines Übertragungsträgers einnimmt, wobei die Daten über das Übertragungsmedium übertragen werden, wobei das Verfahren umfasst:

Identifizieren der statischen Komponente des Übertragungsmediums (110) für eine Breitbandkommunikation, die mindestens zwei Teilkanäle aufweist;

Bestimmen einer Funktion eines Rauschens und/oder einer Dämpfung in Bezug auf eine Frequenz für die statische Komponente (120);

Auswählen von Breitbandmodulationsschemata für die Teilkanäle (130), die nicht-interferierend oder orthogonal zu den anderen Teilkanälen ausgebildet sind, so dass ein erhöhter oder höchster Gesamtdurchsatz bei einer erforderlichen Fehlerrate auf der Basis der statischen Funktionen des Rauschens und der Dämpfung in Bezug auf die Frequenz erreicht wird;

Übertragen eines Datensignals über die Teilkanäle (145);

Identifizieren einer dynamischen Komponente des Übertragungsmediums für die Breitbandübertragung (150);

Erzeugen eines adaptiven Filters auf der Basis der dynamischen Komponente (160); und

Anwenden des adaptiven Filters auf ein empfangenes Datensignal bei einem Empfänger (160).


 
2. Verfahren nach Anspruch 1, das ferner das unabhängige Anwenden der ausgewählten Breitbandmodulationsschemata auf jeden der Breitbandteilkanäle (170) umfasst.
 
3. Verfahren nach Anspruch 2, bei dem das unabhängige Anwenden der ausgewählten Breitbandmodulationsschemata (170) das Verwenden von getrennten Modulatoren für jeden der Breitbandteilkanäle zum Anwenden der ausgewählten Modulationsschemata umfasst.
 
4. Verfahren nach Anspruch 1, bei dem:

das Identifizieren der statischen Komponente das unabhängige Identifizieren der statischen Komponente auf jedem der Breitbandteilkanäle umfasst;

das Bestimmen der Funktion das Bestimmen der Funktion des Rauschens oder der Dämpfung in Bezug auf die Frequenz (120) für die statische Komponente von jedem der Breitbandteilkanäle umfasst; und

das Auswählen des Breitbandmodulationsschemas (170) das Auswählen eines Breitbandmodulationsschemas (170) zum Optimieren des Durchsatzes auf jedem der Breitbandteilkanäle auf der Basis der Funktion umfasst, die auf jedem der Breitbandteilkanäle bestimmt worden ist.


 
5. Verfahren nach Anspruch 1, bei dem:

das Bestimmen der Funktion das Durchführen einer Messung des Rauschens oder der Dämpfung in Bezug auf die Frequenz (120) während eines Zeitintervalls, das von einem Rahmen-für-Rahmen-Zeitintervall verschieden ist, und das Bestimmen der Funktion des Rauschens oder der Dämpfung in Bezug auf die Frequenz (120) für die statische Komponente auf der Basis der Messung umfasst; und

das Auswählen des Breitbandmodulationsschemas das Auswählen des Breitbandmodulationsschemas zu einem Zeitpunkt der Messung auf der Basis der bestimmten Funktion umfasst.


 
6. Verfahren nach Anspruch 1, bei dem das Bestimmen der Funktion eine Durchschnittsbildung einer Funktion eines Rauschens oder einer Dämpfung in Bezug auf die Frequenz (200) für die statische Komponente während eines Zeitraums umfasst.
 
7. Computerprogramm, das Kodesegmente zum Bewirken umfasst, dass ein Computer (710) das Verfahren nach einem der vorhergehenden Ansprüche ausführt.
 
8. System zum Übertragen von Daten über ein Übertragungsmedium mit Rauschen und einer Dämpfung, die als Funktionen der Frequenz eine statische Komponente, die als vorhersagbare oder durchschnittliche Funktion festgelegt ist, die zeitunabhängig ist, und eine dynamische Komponente, die sowohl bezüglich der Zeit als auch des Orts variiert, aufweisen, wobei ein Übertragungskanal des Übertragungsmediums in mindestens zwei Breitbandteilkanäle aufgeteilt ist, wobei ein Symbol, das über einen Breitbandteilkanal übertragen wird, weniger als 50 Zyklen eines Übertragungsträgers einnimmt, wobei die Daten über das Übertragungsmedium übertragen werden, wobei das System umfasst:

ein Prozessormodul, das strukturiert und ausgebildet ist zum

Identifizieren der statischen Komponente des Übertragungsmediums (110) für eine Breitbandkommunikation, die mindestens zwei Teilkanäle aufweist;

Bestimmen einer Funktion eines Rauschens und/oder einer Dämpfung in Bezug auf eine Frequenz für die statische Komponente (120);

Auswählen von Breitbandmodulationsschemata für die Teilkanäle (130), die nicht-interferierend oder orthogonal zu den anderen Teilkanälen ausgebildet sind, so dass ein erhöhter oder höchster Gesamtdurchsatz bei einer erforderlichen Fehlerrate auf der Basis der statischen Funktionen des Rauschens und der Dämpfung in Bezug auf die Frequenz erreicht wird;

Identifizieren einer dynamischen Komponente des Übertragungsmediums für die Breitbandübertragung (150);

ein Übertragungsmodul, das zum Übertragen eines Datensignals über die Teilkanäle (145) strukturiert und ausgebildet ist; und

einen Empfänger, der einen adaptiven Filter umfasst, der auf der dynamischen Komponente basiert und auf ein empfangenes Datensignal bei einem Empfänger (160) angewandt wird.


 
9. System nach Anspruch 8, das ferner ein Modulatormodul mit einem oder mehreren Modulator(en) umfasst, der oder die zum unabhängigen Anwenden der ausgewählten Breitbandmodulationsschemata (170) auf jeden der Breitbandteilkanäle strukturiert und ausgebildet ist oder sind.
 
10. System nach Anspruch 9, bei dem das Modulatormodul getrennte Modulatoren für jeden der Teilkanäle zum Anwenden der ausgewählten Modulationsschemata (170) umfasst.
 
11. System nach Anspruch 8, bei dem der Prozessor strukturiert und ausgebildet ist zum: unabhängigen Identifizieren der statischen Komponente auf jedem der Teilkanäle; Bestimmen der Funktion des Rauschens oder der Dämpfung in Bezug auf die Frequenz (120) für die statische Komponente von jedem der Teilkanäle; und Auswählen eines Breitbandmodulationsschemas (170) zum Optimieren des Durchsatzes auf jedem der Teilkanäle auf der Basis der Funktion, die auf jedem der Teilkanäle bestimmt worden ist.
 
12. System nach Anspruch 8, bei dem der Prozessor strukturiert und ausgebildet ist zum:

Durchführen einer Messung des Rauschens oder der Dämpfung in Bezug auf die Frequenz (120) während eines Zeitintervalls, das von einem Rahmen-für-Rahmen-Zeitintervall verschieden ist, und Bestimmen der Funktion des Rauschens oder der Dämpfung in Bezug auf die Frequenz (120) für die statische Komponente auf der Basis der Messung; und

Auswählen des Breitbandmodulationsschemas (170) zu einem Zeitpunkt der Messung auf der Basis der bestimmten Funktion.


 
13. System nach Anspruch 8, bei dem der Prozessor für eine Durchschnittsbildung einer Funktion eines Rauschens oder einer Dämpfung in Bezug auf die Frequenz (200) für die statische Komponente während eines Zeitraums strukturiert und ausgebildet ist.
 
14. System nach Anspruch 8, bei dem der adaptive Filter (160) auf das übertragene Datensignal auf jedem der Teilkanäle angewandt wird.
 
15. System nach Anspruch 8, bei dem der adaptive Filter (160) einen Rauschfilter, einen Kanalfilter oder einen Frequenzbereichentzerrer umfasst.
 


Revendications

1. Procédé (100) de communication de données sur un support de transmission avec bruit et atténuation qui présentent, en fonction de la fréquence, une composante statique, qui est définie en tant que fonction prédictible ou moyenne qui est indépendante du temps, et une composante dynamique, variant à la fois dans le temps et l'emplacement, un canal de transmission du support de transmission étant divisé en au moins deux sous-canaux à large bande, dans lequel un symbole transmis sur un sous-canal à large bande occupe moins que 50 cycles d'une porteuse de transmission, dans lequel les données sont transmises sur le support de transmission,
le procédé comprenant :

identifier la composante statique du support de transmission (110) pour communications à large bande présentant au moins deux sous-canaux ;

déterminer une fonction de bruit et/ou d'atténuation en fonction de la fréquence pour la composante statique (120) ;

sélectionner des techniques de modulation à large bande pour les sous-canaux (130), qui sont rendus non brouillants ou orthogonaux aux autres sous-canaux, pour atteindre un débit total accru ou le plus élevé à un taux d'erreurs requis, sur la base des fonctions statiques de bruit et d'atténuation en fonction de la fréquence ;

transmettre un signal de données sur les sous-canaux (145) ;

identifier une composante dynamique du support de transmission pour les communications à large bande (150) ;

créer un filtre adaptatif sur la base de la composante dynamique (160) ; et

appliquer le filtre adaptatif à un signal de données reçu au niveau d'un récepteur (160).


 
2. Procédé selon la revendication 1, comprenant en outre l'application de manière indépendante des techniques de modulation à large bande sélectionnées à chacun des sous-canaux à large bande (170).
 
3. Procédé selon la revendication 2, dans lequel l'application de façon indépendante des techniques de modulation à large bande sélectionnées (170) comprend l'utilisation de modulateurs distincts pour chacun des sous-canaux à large bande pour appliquer les techniques de modulation sélectionnées.
 
4. Procédé selon la revendication 1, dans lequel :

l'identification de la composante statique comprend l'identification de façon indépendante de la composante statique sur chacun des sous-canaux à large bande ;

la détermination de la fonction comprend la détermination de la fonction de bruit ou d'atténuation en fonction de la fréquence (120) pour la composante statique de chacun des sous-canaux à large bande ; et

la sélection de la technique de modulation à large bande (170) comprend la sélection d'une technique de modulation à large bande (170) pour optimiser un débit sur chacun des sous-canaux à large bande sur la base de la fonction déterminée sur chacun des sous-canaux à large bande.


 
5. Procédé selon la revendication 1, dans lequel :

la détermination de la fonction comprend la prise d'une mesure du bruit ou de l'atténuation en fonction de la fréquence (120) durant un intervalle temporel autre qu'un intervalle temporel trame-par-trame et la détermination de la fonction du bruit ou de l'atténuation en fonction de la fréquence (120) pour la composante statique sur la base de la mesure ; et

la sélection de la technique de modulation à large bande comprend la sélection de la technique de modulation à large bande à un instant de la mesure sur la base de la fonction déterminée.


 
6. Procédé selon la revendication 1, dans lequel la détermination de la fonction comprend le moyennage d'une fonction de bruit ou d'atténuation en fonction de la fréquence (200) pour la composante statique sur une période temporelle.
 
7. Programme d'ordinateur comprenant des segments de code pour amener un ordinateur (710) à réaliser le procédé selon l'une quelconque des revendications précédentes.
 
8. Système de communication de données sur un support de transmission avec bruit et atténuation qui présentent, en fonction de la fréquence, une composante statique, qui est définie en tant que fonction prédictible ou moyenne qui est indépendante du temps, et une composante dynamique, variant à la fois dans le temps et l'emplacement, un canal de transmission du support de transmission étant divisé en au moins deux sous-canaux à large bande, dans lequel un symbole transmis sur un sous-canal à large bande occupe moins que 50 cycles d'une porteuse de transmission, dans lequel les données sont transmises sur le support de transmission,
le système comprenant :

un module de processeur qui est structuré et agencé pour

identifier la composante statique du support de transmission (110) pour des communications à large bande présentant au moins deux sous-canaux,

déterminer une fonction de bruit et/ou d'atténuation en fonction de la fréquence pour la composante statique (120),

sélectionner des techniques de modulation à large bande pour les sous-canaux (130), qui sont rendus non brouillants ou orthogonaux par rapport aux autres sous-canaux, pour atteindre un débit total accru ou le plus élevé à un taux d'erreurs requis, sur la base des fonctions statiques de bruit et d'atténuation en fonction de la fréquence ;

identifier une composante dynamique du support de transmission pour les communications à large bande (150) ;

un module de transmission qui est structuré et agencé pour communiquer un signal de données sur les sous-canaux (145) ; et

un récepteur comprenant un filtre adaptatif qui est basé sur la composante dynamique et appliqué à un signal de données reçu au niveau d'un récepteur (160).


 
9. Système selon la revendication 8, comprenant en outre un module de modulateur présentant un ou plusieurs modulateurs qui sont structurés et agencés pour appliquer de façon indépendante les techniques de modulation à large bande sélectionnées (170) à chacun des sous-canaux à large bande.
 
10. Système selon la revendication 9, dans lequel le module de modulateur utilise des modulateurs distincts pour chacun des sous-canaux pour appliquer les techniques de modulation sélectionnées (170).
 
11. Système selon la revendication 8, dans lequel le processeur est structuré et agencé pour :

identifier de manière indépendante la composante statique sur chacun des sous-canaux ;

déterminer la fonction de bruit ou d'atténuation en fonction de la fréquence (120) pour la composante statique de chacun des sous-canaux ; et

sélectionner une technique de modulation à large bande (170) pour optimiser un débit sur chacun des sous-canaux sur la base de la fonction déterminée sur chacun des sous-canaux.


 
12. Système selon la revendication 8, dans lequel le processeur est structuré et agencé pour :

prendre une mesure du bruit ou de l'atténuation en fonction de la fréquence (120) durant un intervalle temporel autre qu'un intervalle temporel trame-par-trame et déterminer la fonction du bruit ou de l'atténuation en fonction de la fréquence (120) pour la composante statique sur la base de la mesure ; et

sélectionner la technique de modulation à large bande (170) à un instant de la mesure sur la base de la fonction déterminée.


 
13. Système selon la revendication 8, dans lequel le processeur est structuré et agencé pour moyenner une fonction de bruit ou d'atténuation en fonction de la fréquence (200) pour la composante statique sur une période temporelle.
 
14. Système selon la revendication 8, dans lequel le filtre adaptatif (160) est appliqué au signal de données communiqué sur chacun des sous-canaux.
 
15. Système selon la revendication 8, dans lequel le filtre adaptatif (160) comprend un filtre de bruit, un filtre de canal ou un égaliseur de domaine fréquentiel.
 




Drawing



































Cited references

REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description




Non-patent literature cited in the description