JP3803668B2 - ADSL modem apparatus and pilot tone selection optimization method - Google Patents

Description

  The present invention relates to an ADSL (Digital Subscriber Line Access Multiplexer) provided in a telephone station for performing communication via a telephone line with a DSLAM (Digital Subscriber Line Access Multiplexer) device. Asymmetric digital subscriber line: ADSL modem apparatus capable of optimizing selection of pilot tone associated with various ADSL modulation / demodulation methods according to line conditions, and pilot tone selection optimization method thereof About.

  In recent years, ADSL communication systems have been widely used as services for connecting to the Internet at high speed. In this ADSL communication system, a DSLAM device provided in a telephone office building and an ADSL modem device provided in each subscriber's home are connected using a telephone line. Various methods are used as connection modes for connecting the ADSL modem apparatus and the DSLAM apparatus. For example, there are a method in which the transmission distance that can be connected is not so long but the maximum transfer rate is large, and a method in which the maximum transfer rate is not so large but the transmission distance that can be connected is long. And with the implementation of a number of modulation / demodulation methods such as the deeming method accompanying the recent ITU-T recommendation and technical disclosure, the ADSL service increases the freedom of connection mode selection according to each line distance and environment, and improves the transmission speed. It has been drastically planned.

  In communication using ADSL, a plurality of carrier frequencies are set in a frequency band to be used, and data modulation is performed for each carrier frequency. Therefore, in order to specify which carrier frequency is used, a frequency number (also called a carrier frequency number) is used.

  The ADSL modem apparatus operates the modulation / demodulation function using the pilot tone transmitted from the DSLAM apparatus as a reference. The frequency of the pilot tone to be used is determined by the modulation / demodulation method to be used, but a plurality of pilot tones may be allowed for one modulation / demodulation method. The frequency will be selected. The modulation / demodulation method used is ITU-T recommendation G.264. G.994.1 (hereinafter simply referred to as “G.994.1”). It is selected by determination of the line status using the hs tone (frequency: 384 KHz (frequency number 12), 448 KHz (frequency number 14), 2048 KHz (frequency number 64)), mainly by distance determination. In selecting the modulation / demodulation method, G. The pilot tone is selected from the detection status of the hs tone and the range allowed for the determined modulation / demodulation method. Information about the selection result in the ADSL device is sent from the modem device side to the DSLAM device side as selection information in order to request communication using the selected modulation / demodulation method and pilot tone. This request is based on G.C. This is implemented using a message recommended in 994.1.

  However, the conventional G.P. In the determination using the three carrier frequencies in the hs tone, due to the relationship that these carrier frequencies are concentrated in a relatively low frequency band among the frequency bands used in ADSL, errors due to line conditions such as impedance imbalance and signal reflection are caused. Judgment is likely to occur and the optimum pilot tone may not always be selected. As a result, the training sequence can be continued at the preparation stage of the modulation / demodulation processing after the pilot tone selection (for example, the training sequence according to ITU-T recommendation G.992.1 (hereinafter simply referred to as “G.992.1”)). There is a case where the state does not change until data communication (link establishment) fails.

  The object of the present invention is to determine the situation in the training sequence of each modulation / demodulation method, detect erroneous setting of pilot tone selection, and change to an optimal pilot tone, so that robust modulation / demodulation suitable for all line conditions An object of the present invention is to provide an ADSL modem apparatus that realizes the execution of the system and a pilot tone selection optimization method therefor.

  An ADSL modem apparatus according to the present invention is an ADSL modem apparatus that communicates with a DSLAM apparatus provided in a telephone station via a telephone line, and includes at least a modulation / demodulation unit that performs a training sequence, a modulation / demodulation unit, In cooperation, a selection unit that selects a modulation / demodulation method that seems to be optimal in the current line situation and a training sequence that uses a pilot tone that is allowed in the selected modulation / demodulation method are executed by the modulation / demodulation unit, so that the training sequence fails. And a control unit that detects and changes the pilot tone in response to the failure and causes the modem unit to re-execute the training sequence.

The pilot tone selection optimization method of the present invention is a pilot tone selection optimization method for optimizing the selection of pilot tones used between a DSLAM device and an ADSL modem device provided in a telephone office building, Select the modulation / demodulation method that seems to be optimal in the current line conditions, and perform a training sequence using the pilot tones that are allowed in the selected modulation / demodulation method. And changing the pilot tone in response to re-execution of the training sequence.

  That is, in the ADSL modem apparatus of the present invention, G. 994.1 and G.I. The pilot tone selection function used in the process of transition of modulation / demodulation by the 992.1 recommendation and the deemed system is provided. This makes it possible to realize robust communication quality.

  The pilot tone, which is a low frequency band and a narrow band, is in a situation where the carrier frequency band is used for a higher frequency in accordance with the recent high speed modulation / demodulation method, so in particular, the bridge tap (BT; bridge tap) The condition is likely to be affected by line noise and the like, and there is a possibility that the failure probability of training increases. According to the present invention, by appropriately selecting each pilot tone, the G. The execution of robust training can be complemented in the modulation / demodulation method determination algorithm using the hs tone, the transition success probability to the data communication state can be dramatically improved, and the robustness of the wideband transmission method can be enhanced. .

  Next, a preferred embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of an ADSL modem apparatus according to an embodiment of the present invention.

  As shown in FIG. 1, the ADSL modem apparatus 2 is connected to a DSLAM apparatus 1 in a telephone station via a telephone line 3 and transmits / receives data to / from the DSLAM apparatus 1. . The ADSL modem apparatus 2 is a G. An hs unit 4, a modem unit 5, a control unit 6, and a memory 7 are provided. The memory 7 stores a pilot tone frequency number associated with each modulation / demodulation method, the number of training failures, and the like. G. The hs portion 4 is a G. As defined in G.994.1. By using the hs tone signal, the G.G. signal in the downstream direction (the direction from the DSLAM device 1 toward the ADSL modem device 2). The frequency component of the hs tone is discriminated in cooperation with the modulation / demodulation unit 5 to select a modulation / demodulation method that seems to be optimal in the current line situation and to determine a handshake tone. The control unit 6 controls the functional state of the ADSL modem apparatus 2. In particular, the control unit 6 monitors the state of the modem unit 5 during training, and selects an appropriate pilot tone by selecting a pilot tone and conditional branching. . Specifically, the control unit 6 performs G. The modulation / demodulation unit 5 performs a training sequence using a pilot tone allowed in the modulation / demodulation method selected by the hs unit 4, detects a failure of the training sequence, changes the pilot tone in accordance with the failure, and causes the modulation / demodulation unit 5 to Re-execute the training sequence.

  In this ADSL modem apparatus 2, G. The selected frequency numbers (carrier frequency numbers) of various pilot tones after the training sequence determined in association with each modulation / demodulation method recommended in 992.1 or the like are held. When the ADSL modem apparatus 2 is powered on, first the G. A modulation / demodulation method is determined by the hs unit 4. Thereafter, training is executed in both the ADSL modem apparatus 2 and the DSLAM apparatus 1 by the initial setting pilot tone. In this embodiment, it is determined whether the pilot tone is suitable for the line condition by monitoring the failure during the training. In addition, a pilot tone having a different frequency is executed as a recovery process at the time of retraining after a failure, thereby complementing robust training execution. That is, in this embodiment, G.I. In an ADSL modem apparatus standardized by 992.1 or using a related disregarding system, The carrier frequency recommended in 994.1, ie G. For the purpose of complementing the selection of C-Pilot 1 (ie, pilot tone) determined by this selection function by selecting a modulation / demodulation method such as Annex I or Annex C using hs tone, Corrects pilot tone selection errors due to signal reflection and other effects.

  Hereinafter, processing for optimizing the pilot tone in the ADSL modem apparatus 2 based on the present embodiment will be described with reference to FIG.

  Assuming that the DSLAM device 1 on the telephone station side is always operating, when the ADSL modem device 2 is activated, first, in step 101, the G. The hs unit 4 receives the G. The frequency component of the hs tone is determined in cooperation with the modulation / demodulation unit 5 and the modulation / demodulation method is determined in accordance with the line condition. At that time, training is executed from pilot tones having a higher frequency among pilot tones stored in the memory 7 as an initial setting under the conditions allowed for the determined modulation / demodulation scheme. Here, pilot tone A is selected (step 102), and training is executed (step 103). For example, when the selected modulation / demodulation method is the Annex I method, since carrier frequency numbers 128 and 64 are defined as pilot tones, the carrier frequency number 128 is selected as an initial value (pilot tone A) and training is performed. Is executed. The control unit 6 monitors whether the training has succeeded or failed in the middle (step 104). If the training is not unsuccessful, that is, if the training has succeeded, the pilot tone A selected in step 102 is finally used as a pilot tone. And the process proceeds to step 112 to shift to a state in which data communication is executed.

  On the other hand, if the training fails in step 104, the number of times the training has failed so far is counted (step 105), and it is determined whether the number of failures N has reached a preset number x (step 105). Step 106). If N has not reached x, the process returns to step 103 in order to execute training with the same pilot tone again. On the other hand, if the training fails x times or more, it is determined that the channel status has a significant adverse effect on the pilot tone frequency, and the pilot tone of another frequency band prepared next is selected ( Step 107) and the training is re-executed (Step 108). In the illustrated example, the pilot tone B located on the lower frequency side than the pilot tone A is selected. The number of times x can be arbitrarily set as desired.

  Even in the execution of the training in step 108, the control unit 6 monitors whether the training by the pilot tone B is successful or has failed in the middle (step 109). The selected pilot tone B is finally determined as a pilot tone, and the process proceeds to step 112 to shift to a state where data communication is executed. On the other hand, if the training fails in step 109, the number of times the training has failed so far is counted (step 110), and it is determined whether the number of failures N has reached a preset number x (step 110). Step 111). If N does not reach x, the process returns to step 108 in order to execute training with the same pilot tone again. On the other hand, if training fails x times or more, it is determined that the channel status has a significant adverse effect on the pilot tone frequency, and training is performed again with the default pilot tone (pilot tone A). Therefore, the process returns to step 102.

  FIG. 3 shows the G.G. of the telephone station DSLAM apparatus 1 connected to the telephone station from the ADSL modem apparatus 2. The sequence of the mode request | requirement using the message of 994.1 is shown. First, all the pilot tones that can be implemented using CLR (Capabilities List Request) are presented from the ADSL modem apparatus 2 to the DSLAM apparatus 1 (step 151). Then, the ADSL modem apparatus 2 receives a CL (Capabilities List) from the DSLAM apparatus 1 (step 152), a series of protocols of MP (Mode Proposal) (step 153), MS (Mode Selection Message) (step 154). A sequence that requires a single pilot tone in transmission and reception is used. These sequences are pre-training G.P. This is implemented as a mutual processing between the DSLAM device 1 and the ADSL modem device 2 by hs tone.

  FIG. 4 illustrates a state transition in a sequence between the ADSL modem apparatus and the DSLAM apparatus in the training in Step 103 or Step 108. As shown, this sequence is G. Although it is based on 992.1, it is different in that “C-Pilot 2” on the ADSL modem apparatus side is equal to “C-Pilot 1” with the pilot tone selection function. ing.

  Although the preferred embodiments of the present invention have been described above, the present invention is not limited to the ADSL system in a narrow sense, but also the ITU-T recommendation G.264. It is also applicable to the 993.1 VDSL (very high bit rate digital subscriber line) system.

It is a block diagram which shows the structure of the ADSL modem apparatus of one Embodiment of this invention. It is a flowchart which shows a pilot tone selection process. G. between the ADSL modem device and the DSLAM device. It is a figure which shows the sequence of hs communication. It is a figure explaining the state transition in the sequence between an ADSL modem apparatus and a DSLAM apparatus.

Explanation of symbols

1 DSLAM equipment 2 ADSL modem 3 Telephone line 4 G. hs section 5 modulation / demodulation section 6 control section 7 memory

Claims (8)

An ADSL modem device that communicates with a DSLAM device provided in a telephone station via a telephone line,
A modem unit that performs at least a training sequence;
In cooperation with the modem unit, a selection unit that selects a modulation / demodulation method that seems to be optimal in the current line situation;
The modulation / demodulation unit performs a training sequence using a pilot tone allowed in the selected modulation / demodulation method, detects a failure of the training sequence, changes the pilot tone according to the failure, and causes the modulation / demodulation unit to A control unit to re-execute the training sequence;
An ADSL modem device. The selection unit is an ITU-T recommendation G.264. As defined in G.994.1. G. Selecting modulation / demodulation method by determining line condition using hs tone signal hs part,
2. The ADSL modem according to claim 1, wherein the control unit selects a pilot tone in order from a pilot tone having a higher frequency among pilot tones allowed in the selected modulation / demodulation scheme, and causes the training sequence to be executed. apparatus.   As the pilot tone change request to the DSLAM device, the control unit sends an ITU-T recommendation G. The ADSL modem apparatus according to claim 1 or 2, wherein CLR (Capabilities List Requirement) and MS (Mode Request Message), which are messages defined in 994.1, are used.   4. The control unit according to any one of claims 1 to 3, wherein the control unit instructs the modulation / demodulation unit to forcibly shift to a trial of the next pilot tone when the number of training sequence failures exceeds a predetermined number during training with each pilot tone. The ADSL modem apparatus according to claim 1. A pilot tone selection optimization method for optimizing selection of pilot tones used between a DSLAM device and an ADSL modem device provided in a telephone office building,
Selecting a modulation / demodulation method that seems to be optimal for the current line conditions;
When a training sequence using a pilot tone allowed in the selected modulation / demodulation method is performed and a failure of the training sequence is detected, the pilot tone is changed according to the failure and the training sequence is re-executed And the stage of
A pilot tone selection optimization method comprising: ITU-T Recommendation G. As defined in G.994.1. The modulation / demodulation method is selected by determining the line status using the hs tone signal,
6. The pilot tone selection optimization method according to claim 5, wherein pilot training is performed by selecting pilot tones in order from a pilot tone having a higher frequency among pilot tones allowed in the selected modulation / demodulation scheme. .   As a pilot tone change request from the ADSL modem apparatus to the DSLAM apparatus, an ITU-T recommendation G. The pilot tone selection optimization method according to claim 5 or 6, wherein CLR (Capabilities List Requirement) and MS (Mode Request Message), which are messages specified in 994.1, are used.   The pilot tone selection optimization method according to any one of claims 5 to 7, wherein, when training with each pilot tone, the training sequence with the next pilot tone is executed when the number of training sequence failures exceeds a certain number.

Images