JP2005341533A - Subscriber-side adsl device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a subscriber line transmission system capable of taking TCM Cross-talk measures by partially changing software, relative to hardware that adopts a standard scheme, without being considerably different from the standard scheme. <P>SOLUTION: In the subscriber-side ADSL device to which a burst phase of a TCM-ISDN is notified from a station side ADSL device, an ATU-R device includes a means for executing FFT of a received signal, recognizing an FEXTR phase or NEXTR phase from the phase of the FFT output and using information thereof to recognize a phase of 400HZ that is a burst cycle of the TCM-ISDN. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a digital subscriber line transmission system that uses an existing telephone line as a high-speed data communication line, and more particularly to an improvement in a modulation / demodulation method of a transmission apparatus used in the transmission system.

  In recent years, multimedia services such as the Internet have become widespread throughout society, including households, and early provision of an economical and reliable digital subscriber line transmission system for using such services. Is strongly demanded.

[1] Description of ADSL Technology xDSL (Digital Subscriber Line) is known as a technology for providing a digital subscriber line transmission system that uses an existing telephone line as a high-speed data communication line. xDSL is a transmission system using a telephone line and is one of modulation / demodulation techniques. This xDSL is broadly divided into an uplink transmission rate from a subscriber's house (hereinafter referred to as a subscriber side) to a receiving station (hereinafter referred to as a station side) and a downlink transmission rate from the station side to the subscriber side. It can be divided into symmetric and asymmetric.

Asymmetric xDSL includes ADSL (Asymmetric DSL), and G.A. DMT and a G.G. Both of them employ a DMT (Discrete Multiple Tone) modulation method as a modulation method.
[2] Explanation of DMT modulation method An example will be described with reference to FIG. In addition, this explanation and explanatory drawing will describe only the modulation / demodulation in the downlink direction from the station to the subscriber.

  First, transmission data is input to the apparatus and stored in the Serial to Parallel Buffer 10 for one symbol time (1/4 kHz). The stored data is divided for each transmission bit number per carrier determined in advance by a transmission bitmap 60 (described later), and is output to the encoder 20. Encoder 20 converts the input bit string into signal points for quadrature amplitude modulation and outputs the signal points to IFFT 30. The IFFT 30 performs inverse fast Fourier transform to perform quadrature amplitude modulation for each signal point, and outputs it to the Parallel to Serial Buffer 40. Here, 16 points of 240 to 255 points of IFFT output are added as the Cyclic Prefix to the head of the DMT symbol. The analog signal is converted from the parallel to serial buffer 40 to the D / A converter 50 at a sampling frequency of 1.104 MHz and transmitted to the subscriber side via the metallic line 100.

On the subscriber side, the A / D converter 110 converts the signal into a 1.104 MHz digital signal and stores it in the Serial to Parallel Buffer 120 for one DMT symbol. The Cyclic Prefix is removed in the same buffer and output to the FFT 130. The FFT 130 performs fast Fourier transform to generate (demodulate) signal points. The demodulated signal points are decoded by the decoder 140 according to the reception bitmap 160 holding the same value as the transmission bitmap 60. The decoded data is stored in the Parallel to Serial Buffer 150 and becomes received data as a bit string.
[3] Detailed Description of Bitmap A bitmap described in the DMT modulation method will be described in more detail with reference to FIG.

  The station-side device and the subscriber-side device measure the ratio of the modulated signal and noise (hereinafter referred to as S / N) on the line during training for communication, and determine the number of bits transmitted on each modulated carrier. decide. As shown in FIG. 12, a carrier having a large S / N is assigned a large number of transmission bits, and a carrier having a small S / N is assigned a small number of transmission bits.

  Thereby, a bit map indicating the number of transmission bits corresponding to the carrier number is created from the measured S / N on the receiving side.

On the receiving side, this bit map is notified to the transmitting side during training, so that it is possible to perform modulation / demodulation using the same bit map on both the transmitting and receiving sides during steady data communication.
[4] Measures against crosstalk from ISDN ping-pong transmission If there is cross-talk from ISDN ping-pong transmission (hereinafter referred to as TCM Cross-talk), ADSL will improve transmission characteristics by using two of the aforementioned bitmaps. I was trying. A method of using two bitmaps will be described with reference to FIG.

  In ISDN ping-pong transmission, the station side transmits downlink data in the first half cycle of 400 Hz in synchronization with 400 Hz shown in FIG. 13, and the subscriber side transmits uplink data after receiving downlink data. For this reason, ADSL on the station side is affected by near-end crosstalk from ISDN (hereinafter referred to as NEXT) in the first half cycle of 400 Hz, and far-end crosstalk (from the upstream data of subscriber-side ISDN in the second half cycle ( Hereinafter, it is called FEXT.)

  In the subscriber side ADSL, contrary to the station side, the first half of 400 Hz is affected by FEXT, and the second half is affected by NEXT.

  When the metallic cable between the station and the subscriber becomes longer, the S / N between the received signal and the NEXT becomes smaller, and in some cases, the NEXT becomes larger than the received signal.

  Even in this case, since there is not much influence of FEXT, conventionally, two bitmaps for receiving the NEXT interval (DMT symbol A) and two bitmaps for receiving the FEXT interval (DMT symbol B) are prepared. In the NEXT interval, A technique has been adopted in which the number of transmission bits is reduced to improve the S / N tolerance, and the number of transmission bits is increased in the FEXT section to increase the transmission capacity.

At this time, in order to match the TCM cross-talk period of 400 Hz, the original 16-point cyclic prefix is 246 μS per 1 DMT symbol, whereas the cyclic prefix is 20 points, and the TCM cross-talk is 250 μS per 1 DMT symbol. -Talk was synchronized with TCM Cross-talk for one cycle and 10 DMT symbols.
[5] FEXT and NEXT
FIG. 1 shows a timing chart of crosstalk received by ADSL from TCM-ISDN.

  TCM-ISDN operates at a frequency of 400 Hz and its period is 2.5 ms. Of the one TCM-ISDN cycle, the first half cycle is transmitted by the CO side, and the second half cycle is transmitted by the RT side.

  Therefore, in the first half of one TCM-ISDN cycle, the station side ADSL equipment (ATU-C) is affected by near-end cross-talk (NEXT: near end cross-talk) from TCM-ISDN. In the half cycle, it is affected by far end cross-talk (FEXT) from TCM-ISDN. On the other hand, the subscriber ADSL equipment (ATU-R) is affected by FEXT from TCM-ISDN in the first half of the TCM-ISDN cycle, and from TCM-ISDN to NEXT in the second half cycle. Receive. In this specification, such time regions affected by NEXT and FEXT are referred to as NEXT interval and FEXT interval, respectively.

  The station-side ADSL device (ATU-C) can estimate the FEXT interval and the NEXT interval in the subscriber-side ADSL device (ATU-R). Similarly, the subscriber-side ADSL apparatus (ATU-R) can estimate the FEXT section and the NEXT section in the station-side ADSL apparatus (ATU-C). Each section is defined as follows.

FEXTR: FEXT interval in ATU-R estimated by ATU-C
NEXTR: NEXT interval in ATU-R estimated by ATU-C
FEXTC: FEXT interval in ATU-C estimated by ATU-R
NEXTC: NEXT interval in ATU-C estimated by ATU-R Note that transmission delay is also considered in the above definition.
[6] Sliding window In order to provide a digital subscriber line transmission system capable of transmitting an ADSL signal satisfactorily under the above-described crosstalk environment from TCM-ISDN, the present applicant has first proposed. Proposed the introduction of "sliding window" in Japanese Patent Application No. 10-144913.

  According to the above-mentioned Japanese Patent Application No. 10-144913, in the case of the downlink direction in which the ADSL signal is transmitted from the station side ADSL device (ATU-C) to the subscriber side ADSL device (ATU-R), the cross from TCM-ISDN The state of the ADSL signal transmitted by the station side ADSL equipment (ATU-C) under the talk environment is defined as follows.

  That is, as shown in FIG. 2, when the transmission symbol is completely included in the FEXTR section, the station side ADSL apparatus (ATU-C) transmits the symbol as an inside symbol by the sliding window. When a part of the transmission symbol is included in the NEXTR section, the station ADSL device (ATU-C) transmits the symbol as an outside symbol.

The station ADSL equipment (ATU-C) transmits inside symbols using bitmap A, which is a bitmap for FEXTR section, and uses outside symbols using bitmap B, which is a bitmap for NEXTR section. Send (Dual Bitmap).
As in the downlink, in the uplink, the subscriber ADSL equipment (ATU-R) transmits an inside symbol using the bitmap A which is the bitmap for the FEXTC section, and the bitmap B which is the bitmap for the NEXTC section is transmitted. To send outside symbols.

Here, the station-side ADSL device (ATU-C) may not use the bitmap B (Single Bitmap). At this time, the station-side ADSL device (ATU-C) transmits only the pilot tone outside the sliding window. Similarly, the subscriber-side ADSL device (ATU-R) may not use the bitmap B, and the subscriber-side ADSL device (ATU-R) does not transmit anything outside the sliding window.
"A Study on ADSL Suitable for Crosstalk from TCM-ISDN" 1998 IEICE General Conference European Patent Application No. 841771 JP-A-8-331018 Special table hei 10-502511 Japanese Patent Application No. 10-304125 (Japanese Patent Laid-Open No. 2000-32096)

  As described above, an effective ADSL transmission technique in a noise environment from TCM-ISDN is provided by, for example, the above-mentioned Japanese Patent Application No. 10-144913 made by the present applicant. There is still room for consideration regarding the specific training method for ADSL transceivers and the means for implementing such training methods when adopting various transmission technologies.

  The present invention has been made on the basis of new knowledge and consideration on the above points, and in the ADSL transceiver in adopting effective transmission technology of ADSL signals under the noise environment from TCM-ISDN. It is an object of the present invention to provide a specific training method or a digital subscriber line transmission system provided with means for performing such a training method and a communication device used therefor.

  As described above in detail, according to the present invention, a specific training method for an ADSL transceiver, or such a training method, when adopting an effective transmission technology of an ADSL signal under a noise environment from TCM-ISDN. A digital subscriber line transmission system including means for implementing the above and a communication apparatus used therefor are provided.

  According to the first aspect of the present invention, in the subscriber-side ADSL device that is notified of the TCM-ISDN burst phase from the station-side ADSL device, the ATU-R device performs FFT of the received signal, and An ADSL device on the subscriber side has means for recognizing the FEXTR section or NEXTR section based on the phase of the FFT output, and using the information to recognize the phase of 400 Hz, which is the TCM-ISDN burst period. Provided.

  According to the second aspect of the present invention, the subscriber-side ADSL device that is notified of the TCM-ISDN burst phase from the station-side ADSL device demodulates the received signal from the station-side ADSL device, As a result, there is provided an ADSL apparatus on the subscriber side which has means for recognizing the FEXTR section or the NEXTR section and recognizing the phase of the TCM-ISDN 400HZ using the information.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.
[1] Initialization FIG. 3 shows an outline of a timing chart in the initialization of the ADSL transceiver. At the time of ADSL training, it is important to send the ADSL signal only in the section that does not cause NEXT noise to TCM-ISDN in consideration of the influence on TCM-ISDN for both uplink and downlink. Therefore, as shown in FIG. 3, initialization is performed with a single bitmap in transceiver training and exchange. For channel analysis (channel analysis), initialization is performed with a single bitmap in sequences other than C medley (C-MEDLEY) and R medley (R-MEDLEY), but C medley (C-MEDLEY) ) And R-MEDLEY only, channel quality with both inside and outside symbols when using Dual Bitmap, and only inside symbols when using Single Bitmap. Survey (measurement of S / N).
[2] Initialization Counter FIG. 4 shows an embodiment of the initialization counter of the present invention.

In ADSL, the station side ADSL equipment (ATU-C) and the subscriber side ADSL equipment (ATU-R) have independent counters. The hyperframe counter (501) has a function of counting the number of DMT symbols by continuously counting the DMT symbol clock (519) a predetermined number of times (for example, 345 times), and the sliding window DEC (503 ) To specify the FEXTR, NEXTR, FEXTC, and NEXTC sections of the sliding window (523).
In addition, in order to match the start of C-REVEILLE and C-RATES1 in the station side ADSL equipment (ATU-C) and the start of R-REBERB3 in the subscriber side ADSL equipment (ATU-R) to the phase of 400 Hz, a 400 Hz signal (517 ) Is set as a sequence transition condition (507) and the hyperframe counter (501) is cleared at that time. The symbol counter (505) counts the number of DMT symbols by counting the DMT symbol clock (519), and the sequence transition condition is that it matches the count value DEC (513) (509). (507) and decide the length of each initialization signal. The sequence counter information (521) such as the received signal detection signal from the outside and the CRC calculation result is used as the transition condition logic (507) and the enable signal of the sequence counter (511), so that the count value of the sequence counter (511) Is the code value representing the status of the initialization sequence, and initialization information (C-REVEILLE, C-PILOT1, C-REVERB1, etc.) is determined by initialization DEC using the count value of the sequence counter (511). 525).
This method is conscious of realization in hardware, but can be realized in the same configuration in software.

In C-PILOT1, the TCM-ISDN 400 Hz phase is reported from the station ADSL equipment (ATU-C) to the subscriber ADSL equipment (ATU-R), and the subscriber ADSL equipment (ATU-R) Is detected as a 400Hz signal (517). Although details of this method will be described later, this enables periodic crosstalk detection such as TCM-ISDN in the subscriber side ADSL equipment (ATU-R).
[3] Transceiver training
Transceiver training includes TEQ, FEQ, AGC, timing recovery, and frame synchronization training sequences. For these, training is performed only when the ADSL transceiver repeatedly sends out a pseudo-probability signal, for example, a synchronization symbol (S).
On the other hand, in transceiver training, initialization is performed using a single bitmap, so that these trainings are inevitably performed only in the FEXT interval.

  However, if initialization is performed with dual bitmap in transceiver training, training may be performed only in the FEXT section.

In TEQ and FEQ training, it is possible to continue training regardless of FEXT and NEXT. At this time, in the NEXT section, the TEQ and FEQ coefficient update step size is set to 0 or a very small value.
[4] Inverse Synchronization Symbol
As shown in FIGS. 8 and 9, there is one inverse synchronization symbol (I) in one hyper frame. In each training, this inverse synchronization symbol (I ) Is also used in conjunction with the synchronization symbol (S) as follows:

  When the inverse synchronization symbol (I) is received at the receiving side, the phase of each carrier except the pilot tone is rotated 180 degrees after the FFT (130) shown in FIG. As a result, the state is the same as when the synchronization symbol (S) is received, and then training is performed using the synchronization symbol (S) generated on the receiving side.

In phase detection for frame synchronization, when phase detection is performed using a synchronization symbol (S), confirmation is performed using the next inverse synchronization symbol (I). In addition, when phase detection is performed using the inverse synchronization symbol (I), confirmation is performed using the next synchronization symbol (S).
[5] Method of notifying the phase of TCM-ISDN 400 Hz from the station side ADSL equipment (ATU-C) to the subscriber side ADSL equipment (ATU-R) The TCM-ISDN 400 Hz phase is described below. -C) Details of how to notify the subscriber ADSL equipment (ATU-R).

  In addition to the pilot tone, C-PILOT1 transmits the 74th (319.125 kHz) carrier belonging to the frequency band with little crosstalk from TCM-ISDN. The TCM-ISDN 400 Hz phase is notified from the station side ADSL device (ATU-C) to the subscriber side ADSL device (ATU-R) using 2 bits on the 74th carrier as follows. This state is shown in FIG.

加入者側ADSL装置(ATU-R) は、局側ADSL装置(ATU-C) から送信された74番目のキャリアを受信し、以下の２種類の方法により、TCM-ISDN 400Hzの位相を認識する。
（１）FFT を行うことで、TCM-ISDN 400Hz の位相を認識する方法
加入者側ADSL装置(ATU-R) では、74番目のキャリアを受信してから、図１２に示されているFFT(130)を実行する。そして、そのFFT 出力の位相により、FEXTR 区間あるいはNEXTR 区間を認識する。そして、その情報を用いてTCM-ISDN 400 Hzの位相を認識する。

The subscriber side ADSL equipment (ATU-R) receives the 74th carrier transmitted from the station side ADSL equipment (ATU-C) and recognizes the phase of TCM-ISDN 400Hz by the following two methods. .
(1) Method of recognizing TCM-ISDN 400Hz phase by performing FFT The subscriber side ADSL equipment (ATU-R) receives the 74th carrier and then receives the FFT ( 130) is executed. The FEXTR section or NEXTR section is recognized according to the phase of the FFT output. And the phase of TCM-ISDN 400 Hz is recognized using the information.

However, with this method, the subscriber-side ADSL equipment (ATU-R) can recognize the phase of TCM-ISDN 400 Hz only with a relatively coarse accuracy every 256 points. Therefore, the following method is effective to obtain better accuracy.
(2) Method of recognizing the phase of TCM-ISDN 400Hz by performing QPSK demodulation The subscriber side ADSL equipment (ATU-R) receives the 74th carrier and then performs QPSK demodulation as shown in FIG. I do. Based on the result, the FEXTR section or NEXTR section is recognized. And the phase of TCM-ISDN 400Hz is recognized using the information.
In this method, the subscriber-side ADSL equipment (ATU-R) can recognize the TCM-ISDN 400 Hz phase with high accuracy for each point.
[6] TCM-ISDN 400 Hz burst clock PLL configuration method Figure 6 shows the ATU-C transmitter reference model. As shown in FIG. 6, an 8 kHz signal called NTR (Network Timing Reference) is always input from the outside to the station side ADSL device (ATU-C) device. In addition, a TCM-ISDN 400 Hz signal may be input from the outside. The TCM-ISDN 400 Hz signal may be generated inside the station-side ADSL device (ATU-C) without being input from the outside (see, for example, Japanese Patent Application No. 10-115223). At this time, TCM-ISDN 400 Hz and NTR 8 kHz are synchronized in frequency.

  In the station side ADSL equipment (ATU-C) equipment shown in Fig. 7, when TCM-ISDN 400 Hz (702) is input from the outside in order to recognize 400 Hz within the equipment, the TCM-ISDN 400 Hz is Rather than input to the ADSL equipment (ATU-C) internal oscillator (VCXO) (704) and synchronizing with the APLL (703), an external NTR 8 kHz (701) signal (TCM-ISDN 400 Hz and NTR 8kHz is synchronized to the oscillation frequency (704) of the ADSL equipment (ATU-C) on the station side and input to VCXO to generate 400 Hz (709) by dividing it. It is characterized by that.

First, normally, TCM-ISDN 400 Hz (702) is input to the PLL inside the station ADSL equipment (ATU-C), and the TCM-ISDN 400 Hz is synchronized with the frequency of the internal VCXO (704).
The oscillation frequency of VCXO of the normal station side ADSL equipment (ATU-C) is about 17.664 MHz. In this case, when synchronizing with TCM-ISDN 400 Hz, it is once in 44160 (17.664M / 400). The PLL synchronization operation is performed based on the phase comparison information. Usually, the greater the number of phase comparisons, the smaller the phase jitter and frequency error. However, with a 17.664 MHz clock, a phase comparison of 44160 times usually results in very large phase jitter and frequency error.

  To avoid this, NTR 8 kHz (701) synchronized with TCM-ISDN 400 Hz is always supplied from the outside to the station ADSL device (ATU-C). C) If the internal VCXO is PLL-synchronized, the number of phase comparisons will increase 20 times compared to TCM-ISDN 400 Hz, and phase comparison information will be obtained once in 2208, reducing phase jitter and frequency error. It becomes.

  The above embodiment is only one embodiment for carrying out the present invention, and various other modifications are considered. Needless to say, the effect of the present invention is not changed in any case.

It is a timing chart of TCM-ISDN crosstalk. It is a figure which shows a sliding window. It is a figure which shows the outline | summary of the timing chart in initialization. It is a figure which shows the embodiment of the initialization counter of this invention. It is a figure which shows QPSK demodulation. It is a figure which shows an ATU-C transmitter reference model. It is a figure which shows an ATU-C timing reproduction algorithm. It is a figure which shows the station side transmission pattern of SWB system. It is a figure which shows the subscriber side transmission pattern of SWB system. It is a figure which shows the transmission pattern for every DMT symbol. It is a figure which shows the SWB system in the case of using two bitmaps. It is a figure which shows the functional block of the subscriber transmission system by a DMT modulation system. It is a figure which shows the definition of a bitmap. It is a figure which shows a prior art example.

Explanation of symbols

501 ... Hyperframe counter 503 ... Sliding window DEC
505 ... Symbol number counter 507 ... Transition condition logic 511 ... Sequence counter 517 ... 400 Hz signal 519 ... DMT symbol clock 521 ... Sequence transition information 525 ... Initialization information

Claims (2)

In the ADSL equipment on the subscriber side where the burst phase of TCM-ISDN is notified from the ADSL equipment on the station side,
The ATU-R unit performs FFT on the received signal, recognizes the FEXTR section or NEXTR section based on the phase of the FFT output, and recognizes the phase of 400HZ, which is the TCM-ISDN burst period, using that information ADSL device on the subscriber side, characterized by having In the ADSL equipment on the subscriber side where the burst phase of TCM-ISDN is notified from the ADSL equipment on the station side,
It has the means to demodulate the received signal from the ADSL equipment on the station side, recognize the FEXTR section or NEXTR section based on the result, and recognize the TCM-ISDN 400HZ phase using that information ADSL equipment on the subscriber side.

Images