JP4139961B2 - Code identification method, code identification circuit, and receiver circuit for optical transmission system - Google Patents

Description

【００３０】
表１に示すように、符号間干渉が小さい場合は、１サンプル前の識別結果に関係なく、中心しきい値に基づいて被識別信号の符号を識別する。これに対して、符号間干渉が大きい場合、即ち、切り替え信号が１の場合、１サンプル前の識別結果に応じて、上位しきい値との比較結果（比較器１０８の出力）、及び、下位しきい値との比較結果（比較器１０９の出力）を選択する。
【００３１】
符号間干渉がある場合、振幅が落ちて波形が広がる。このため、符号‘０’の次にくる符号‘１’の振幅は低くなる傾向がある。逆に、符号‘１’の後にくる符号‘０’の振幅は高くなる傾向がある。上記のように識別しきい値を切り替えることにより符号誤りを減らすことができる。
【００３２】
図２に、被識別信号、選択された比較器の入力しきい値、識別結果の時間波形を示す（識別結果は、‘０’、‘１’であるが図ではわかりやすいように調整して表示している）。振幅０をしきい値に選んだ場合に対して、本発明のしきい値切り替えを用いることにより符号誤りがなくなっている。
【００３３】
次にしきい値の調整方法を説明する。
【００３４】
ゲインの自動設定回路などを用いて符号識別回路の手前で信号振幅がほぼ一定となる場合、各しきい値は、あらかじめ評価を行った結果をもとに固定してもよい。
【００３５】
また、以下に示すようなしきい値調整回路を用いてもよい。図３はその一例であるしきい値調整回路３０１を示すブロック図である。本回路は、特願２００１−３７７９８９「受信品質監視回路」にも記載されている。
【００３６】
比較器３０６は、しきい値制御回路３０９とあわせ、符号‘１’の信号基準を求めるために用いられる。比較器３０７としきい値制御回路３１０は信号レベルの平均値を求めるために用いられる。また、比較器３０８としきい値制御回路３１１は、符号‘０’の基準レベルを求めるために用いられる。
【００３７】
比較器３０６、３０７、３０８は、被識別信号３０２がしきい値制御回路３０９、３１０、３１１の出力より大きい場合に‘１’を出力し、小さい場合に‘０’を出力する。しきい値制御回路３０９、３１０、３１１は、入力が‘１’の場合、しきい値を上げ、‘０’の場合しきい値を下げる。しきい値を上げる制御幅と下げる制御幅を変えることで、所望のレベルを得ることができる。しきい値をあげる制御幅が大きい場合にはしきい値は符号‘１’のレベルに近づき、下げる制御幅が大きい場合には符号‘０’に近づく。たとえば、制御幅の比は下記表２のように選択することが考えられる。
【００３８】
【表２】

【００３９】
しきい値算出回路３１２は、しきい値制御回路３０９としきい値制御回路３１０の中間の値をしきい値３０３として、しきい値制御回路３１０の出力をしきい値３０４として、しきい値制御回路３１０としきい値制御回路３１１の中間の値をしきい値３０５として出力する。
【００４０】
図４は、しきい値調整回路の第二の例であるしきい値調整回路４０１を示すブロック図である。識別結果ごとに信号振幅の平均をとり、各平均値にゲインをかけて、上位しきい値４０４、下位しきい値４０５を生成する。中心しきい値は、決まった固定値（容量結合されていて、直流が遮断されていれば０としてもよい）としても、上位しきい値と下位しきい値の中間値としてもよい。
【００４１】
次に切り替え信号の算出方法について説明する。切り替え信号は、符号間干渉が大きくなった場合に、しきい値の選択方法を切り替えるために用いられる。
【００４２】
図５は、波形劣化を推定する回路の一例を示すブロックである。波形劣化推定回路５０１は、被識別信号５０２を動作クロック（正相）５０４でサンプリングするサンプラ５０６と、被識別信号５０２を動作クロック（逆相）５０３でサンプリングするサンプラ５０７と、サンプラ５０６の出力を入力としてその平均パワーを算出するパワー平均化回路５０８と、サンプラ５０７の出力を入力としてその平均パワーを算出するパワー平均化回路５０９と、パワー平均化回路５０８とパワー平均化回路５０９の出力を入力とする比較器５１０で構成される。
【００４３】
波形が劣化していない場合、最適識別位相でサンプリングされた信号パワーの平均値は、逆相でサンプリングされたものより大きい。波形が劣化し、符号間干渉が大きくなると逆相でサンプリングされた信号パワーが、最適識別位相でサンプリングされた信号パワーの平均より大きくなるため、これを比較器５１０で検知して、検知結果を波形劣化有無信号、即ち、切り替え信号として出力する。
【００４４】
以上、本発明を実施の形態に基づいて説明したが、本発明はこれに限定されるものではなく、当業者の通常の知識の範囲内でその変更や改良が可能であることは勿論である。
【００４５】
【発明の効果】
本発明の符号識別回路を用いることにより、２６００ｐｓ／ｎｍまでエラーを抑えることができることを、シミュレーションにより確認した。
【００４６】
これは、ＳＭＦファイバの１ｋｍあたりの分散値が約２０ｐｓ／ｎｍ／ｋｍであることから換算すると、ファイバ長２０ｋｍの改善効果である（あくまでも一例である。送信チャープ、送信波形、ファイバ入力パワーなどにより、受信波形が変化するため、改善効果はばらつく。光領域（例えば光アンプ）、電気回路で発生する雑音は考慮していない）。
【図面の簡単な説明】
【図１】本発明の一実施の形態である符号識別回路１０１の機能ブロック図である。
【図２】符号識別回路１０１による符号の識別結果を説明するためのグラフである。
【図３】しきい値調整回路３０１の機能ブロック図である。
【図４】しきい値調整回路４０１の機能ブロック図である。
【図５】波形劣化推定回路５０１の機能ブロック図である。
【図６】光伝送システム用受信回路６０１の機能ブロック図である。
【図７】従来の符号識別回路７０１の機能ブロック図である。
【図８】ＮＲＺランダムパターンを分散のある光ファイバに伝送させた場合の符号識別器入力でのアイパターンを説明するグラフである。
【符号の説明】
１０１ 符号識別回路
１０７、１０８、１０９、３０６、３０７、３０８、５１０ 比較器
１１０、１１２、４０６ セレクタ
１１１ パタン識別回路
３０１、４０１ しきい値調整回路
３０９、３１０、３１１ しきい値制御回路
３１２ しきい値算出回路
４０７、４０８ 振幅平均化回路
４０９、４１０ ゲイン回路
５０１ 波形劣化推定回路
５０６、５０７ サンプラ
５０８、５０９ パワー平均化回路[0001]
BACKGROUND OF THE INVENTION
The present invention relates to code identification of a transmission signal, and more particularly to code identification in a receiving circuit of an optical transmission system.
[0002]
[Prior art]
The code identification circuit is, for example, one of the circuits constituting the optical transmission system reception circuit, and is used to identify the received electrical signal as “1” or “0”.
[0003]
FIG. 6 is a block diagram illustrating an example of a reception circuit 601 for an optical transmission system. The optical transmission system receiving circuit 601 receives an optical signal 602 transmitted through an optical fiber, and outputs an identification result 603. The optical signal 602 is converted into an electrical signal by the photodiode 604, the level is adjusted by the reception amplifier 605, the waveform is shaped by the reception filter 606, and the code identification circuit 607 identifies the codes “1” and “0”. And output as an identification result 603.
[0004]
FIG. 7 is a block diagram showing a configuration example of a conventional code identification circuit 701. The code identification circuit 701 uses the comparator 704 to compare the threshold 702 and the identified signal 703, and when the identified signal 703 is larger than the threshold 702, the code identification circuit 701 sets “1” as the identification result 705. If it is smaller, '0' is output.
[0005]
In the conventional coding identification circuit, the threshold value 703 is adjusted and fixed in advance. Further, as described in Patent Documents 1 and 2, there is a technique that uses a plurality of threshold values. Further, as shown in Non-Patent Document 1, there is also a method of automatically adjusting the threshold value by taking a time average. Further, as in Non-Patent Document 2, there is a type that uses a dispersion compensation circuit that calculates an error signal from an identification result and updates a tap coefficient of an adaptive filter.
[0006]
[Patent Document 1]
Japanese Patent Application Laid-Open No. 60-51042
[Patent Document 2]
JP-A-63-111591
[Non-Patent Document 1]
Matsumoto. Y; Kuriyama. T, Inami. D. Ohta. M, Shiraiwa. M.M. "An adaptive decision threshold control of the optical receiver for multi-gibitit terrestrial DWDM transfibration systems 200." OFC 2001, Volume: 2, 2001 Page (s): TuR2-T1-3 vol. 2 (fig1)
[0009]
[Non-Patent Document 2]
Kamran. Azadet, Erich. F, “Equalization and Fec Techniques for Optical Transceivers”, IEEE JOURNAL OF SOLID-STATECIRCUT, VOL37, NO3, MARCH 2002 (FIG. 10)
[0010]
[Problems to be solved by the invention]
FIG. 8 shows an eye pattern at the input of the code discriminator when a 10 Gbps NRZ random pattern is transmitted to an optical fiber having dispersion. As shown in FIG. 8, when the waveform deteriorates due to the chromatic dispersion of the optical fiber and intersymbol interference occurs, the received signal level when the code “0” is received and the code “1” are received depending on the reception pattern sequence. In this case, the received signal level cannot be identified by a single threshold value.
[0011]
The conventional code identification circuit has a fixed threshold value as in Patent Documents 1 and 2, or a non-patent document 1 that takes a time average and adjusts the threshold value gently. There is a problem that a code error occurs due to intersymbol interference depending on the sequence.
[0012]
For example, in a transmission system with a bit rate of 10 Gbps, a modulation scheme NRZ, and a transmission fiber SMF (single mode fiber), an error occurs at a residual dispersion of 2200 ps / nm or more in a conventional code identification circuit with a fixed threshold value. However, the received waveform changes depending on the transmission chirp, transmission waveform, fiber input power, and the like. Also, noise generated in the optical region (for example, an optical amplifier) and an electric circuit is not taken into consideration.
[0013]
Further, when a dispersion compensation circuit that calculates an error signal from the identification result and updates the tap coefficient of the adaptive filter as in Non-Patent Document 2 is used, training cannot be performed successfully due to this code error, and a desired filter characteristic can be obtained. There may not be.
[0014]
In view of such a situation, the problem to be solved by the present invention is a code identification method and a code identification that are less likely to cause a code error even if the waveform deteriorates due to chromatic dispersion of an optical fiber and intersymbol interference occurs. To provide a circuit.
[0015]
[Means for Solving the Problems]
In order to solve the above-mentioned problems, the present invention provides a method having the following characteristics in a method for identifying a code of a signal in a receiving circuit of an optical transmission system. In other words, the code identification method of the present invention is the result of comparing the identified signal with the central threshold value in Step 1 for determining the upper threshold value, the central threshold value and the lower threshold value in descending order. Step 2 for obtaining the comparison result, Step 3 for obtaining the second comparison result obtained by comparing the identified signal with one of the upper threshold value and the lower threshold value, and the magnitude of the intersymbol interference And step 4 of selecting one of the first and second comparison results as the identification result of the code of the identified signal in accordance with the switching signal based on the length.
[0016]
Stage 1 may include placing the signal amplitude within a predetermined range prior to code identification and defining an upper threshold, a center threshold, and a lower threshold according to the range.
[0017]
Alternatively, in step 1, a plurality of different threshold values may be generated according to the identification target signal input immediately before.
[0018]
Alternatively, in the step 1, a plurality of different threshold values may be generated for each identification result based on the average of the signal amplitude of the identified signal.
[0019]
Alternatively, in stage 1, the identified signal is sampled at the normal phase of the operation clock and the average of the power is obtained; and the identified signal is sampled at the opposite phase of the operation clock and the average of the power is obtained. And a step of generating a switching signal based on a comparison between the average A and the average B.
[0020]
The present invention also provides a circuit having the following characteristics in a circuit for identifying the sign of a signal in a receiving circuit of an optical transmission system. That is, the code identification circuit according to the present invention is a result of comparing means 1 for determining the upper threshold value, the central threshold value, and the lower threshold value in descending order with the signal to be identified and the central threshold value. Means 2 for obtaining the comparison result, means 3 for obtaining the second comparison result obtained by comparing the identified signal with either the upper threshold value or the lower threshold value, and the magnitude of the intersymbol interference. And means 4 for selecting one of the first and second comparison results as the identification result of the code of the identified signal in accordance with the switching signal based on the length.
[0021]
The means 1 may include means for keeping the signal amplitude within a predetermined range prior to code identification, and means for determining an upper threshold value, a central threshold value, and a lower threshold value according to the range. .
[0022]
Further, the means 1 may generate a plurality of different threshold values according to the identification signal inputted immediately before.
[0023]
In addition, the means 1 may generate a plurality of different threshold values based on the average signal amplitude of the identified signal for each identification result.
[0024]
Means 1 samples the signal to be identified at the normal phase of the operation clock to determine the average power of the signal, and means to sample the signal to be identified at the reverse phase of the operation clock to determine the average power of the signal. And means for generating a switching signal based on a comparison between the average A and the average B.
[0025]
Furthermore, the present invention provides a receiving circuit for an optical transmission system comprising the above-described code identification circuit.
[0026]
DETAILED DESCRIPTION OF THE INVENTION
A first embodiment of the present invention will be described. Referring to FIG. 1, a code identification circuit 101 according to the present invention includes a comparator 107 that receives a signal to be identified 102 and a threshold 104, a comparator 108 that receives a signal 102 and a threshold 105, and The comparator 109 that receives the identification signal 102 and the threshold 106, the selector 110 that receives the output of the comparator 108, the output of the comparator 109, and the output of the pattern identification circuit 111, and the output of the selector 110 are input. A pattern identification circuit 111, a comparator 107, and a selector 112 that receives the outputs of the selector 110 are included.
[0027]
Next, the operation of the embodiment of the present invention will be described. The comparators 107, 108, and 109 compare the identified signal 102 with each threshold value, and output “1” when the signal is larger than the threshold value and “0” when the signal is smaller. The threshold values 104, 105, and 106 are different from each other and have the following relationship.
Threshold 105> threshold 104> threshold 106
[0028]
The selector 110 selects the output of the comparator 108 when the output of the pattern identification circuit 111 is “1”, and selects the output of the comparator 109 when the output of the pattern identification circuit 111 is “0”. The pattern identification circuit 111 outputs the output of the selector 110 one sample before to the selector 110. The selector 112 selects the output of the selector 110 when the switching signal 113 is “1”, and selects the output of the comparator 107 when it is “0”. The switching signal 113 is set to “1” when the intersymbol interference is large, and is set to “0” when the intersymbol interference is small. When the intersymbol interference is small, the center signal is used to identify the code. Do. Table 1 below shows the relationship between the switching signal 113, the identification result one sample before, and the identification threshold value used.
[0029]
[Table 1]

[0030]
As shown in Table 1, when the intersymbol interference is small, the code of the signal to be identified is identified based on the center threshold value regardless of the identification result one sample before. On the other hand, when the intersymbol interference is large, that is, when the switching signal is 1, the comparison result with the upper threshold (output of the comparator 108) and the lower The comparison result (output of the comparator 109) with the threshold value is selected.
[0031]
When there is intersymbol interference, the amplitude drops and the waveform widens. For this reason, the amplitude of the code “1” following the code “0” tends to be low. On the contrary, the amplitude of the code “0” after the code “1” tends to increase. By switching the identification threshold as described above, code errors can be reduced.
[0032]
FIG. 2 shows the signal to be identified, the input threshold value of the selected comparator, and the time waveform of the identification result (the identification result is “0” or “1”, but is adjusted and displayed for easy understanding in the figure) is doing). In contrast to the case where amplitude 0 is selected as the threshold value, the code error is eliminated by using the threshold value switching of the present invention.
[0033]
Next, a method for adjusting the threshold value will be described.
[0034]
When the signal amplitude becomes substantially constant before the code identification circuit using an automatic gain setting circuit or the like, each threshold value may be fixed based on the result of evaluation in advance.
[0035]
Further, a threshold adjustment circuit as shown below may be used. FIG. 3 is a block diagram showing a threshold adjustment circuit 301 as an example. This circuit is also described in Japanese Patent Application No. 2001-377789 “Reception Quality Monitoring Circuit”.
[0036]
The comparator 306 is used in conjunction with the threshold control circuit 309 to obtain the signal reference of the code “1”. The comparator 307 and the threshold control circuit 310 are used to obtain an average value of the signal level. The comparator 308 and the threshold value control circuit 311 are used to obtain the reference level of the code “0”.
[0037]
The comparators 306, 307, and 308 output “1” when the identified signal 302 is larger than the output of the threshold control circuits 309, 310, and 311, and output “0” when it is smaller. The threshold control circuits 309, 310, and 311 increase the threshold when the input is “1”, and decrease the threshold when the input is “0”. A desired level can be obtained by changing the control width for increasing the threshold value and the control width for decreasing the threshold value. When the control width for raising the threshold value is large, the threshold value approaches the level of “1”, and when the control width to be lowered is large, the threshold value approaches “0”. For example, the control width ratio may be selected as shown in Table 2 below.
[0038]
[Table 2]

[0039]
The threshold calculation circuit 312 uses the intermediate value between the threshold control circuit 309 and the threshold control circuit 310 as the threshold 303 and the output of the threshold control circuit 310 as the threshold 304. An intermediate value between the circuit 310 and the threshold control circuit 311 is output as the threshold 305.
[0040]
FIG. 4 is a block diagram showing a threshold adjustment circuit 401 as a second example of the threshold adjustment circuit. The average of the signal amplitude is taken for each identification result, and the upper threshold 404 and the lower threshold 405 are generated by multiplying each average by a gain. The center threshold value may be a fixed value (which may be 0 if capacitive coupling is performed and DC is cut off), or may be an intermediate value between the upper threshold value and the lower threshold value.
[0041]
Next, a method for calculating the switching signal will be described. The switching signal is used to switch the threshold selection method when the intersymbol interference becomes large.
[0042]
FIG. 5 is a block diagram illustrating an example of a circuit that estimates waveform degradation. A waveform degradation estimation circuit 501 samples a sampler 506 that samples the identified signal 502 with an operation clock (normal phase) 504, a sampler 507 that samples the identified signal 502 with an operation clock (reverse phase) 503, and outputs of the sampler 506. The power averaging circuit 508 that calculates the average power as an input, the power averaging circuit 509 that calculates the average power using the output of the sampler 507 as an input, and the outputs of the power averaging circuit 508 and the power averaging circuit 509 are input. Comparator 510.
[0043]
When the waveform is not degraded, the average value of the signal power sampled at the optimum discrimination phase is larger than that sampled at the opposite phase. When the waveform deteriorates and the intersymbol interference increases, the signal power sampled in the opposite phase becomes larger than the average of the signal power sampled in the optimum discrimination phase, so this is detected by the comparator 510 and the detection result is obtained. A waveform deterioration presence / absence signal, that is, a switching signal is output.
[0044]
The present invention has been described above based on the embodiments. However, the present invention is not limited to this, and it is needless to say that modifications and improvements can be made within the ordinary knowledge of those skilled in the art. .
[0045]
【The invention's effect】
It was confirmed by simulation that the error can be suppressed to 2600 ps / nm by using the code identification circuit of the present invention.
[0046]
This is an improvement effect of the fiber length of 20 km when converted from the dispersion value per km of the SMF fiber of about 20 ps / nm / km (only an example. Depending on the transmission chirp, transmission waveform, fiber input power, etc. Since the received waveform changes, the improvement effect varies, and noise generated in the optical region (for example, an optical amplifier) and an electric circuit is not considered).
[Brief description of the drawings]
FIG. 1 is a functional block diagram of a code identification circuit 101 according to an embodiment of the present invention.
FIG. 2 is a graph for explaining a code identification result by a code identification circuit 101;
3 is a functional block diagram of a threshold adjustment circuit 301. FIG.
4 is a functional block diagram of a threshold adjustment circuit 401. FIG.
5 is a functional block diagram of a waveform deterioration estimation circuit 501. FIG.
6 is a functional block diagram of a receiving circuit 601 for an optical transmission system. FIG.
7 is a functional block diagram of a conventional code identification circuit 701. FIG.
FIG. 8 is a graph for explaining an eye pattern at the input of a code discriminator when an NRZ random pattern is transmitted to an optical fiber having dispersion.
[Explanation of symbols]
101 Code identification circuit 107, 108, 109, 306, 307, 308, 510 Comparator 110, 112, 406 Selector 111 Pattern identification circuit 301, 401 Threshold adjustment circuit 309, 310, 311 Threshold control circuit 312 Threshold Value calculation circuit 407, 408 Amplitude averaging circuit 409, 410 Gain circuit 501 Waveform deterioration estimation circuit 506, 507 Sampler 508, 509 Power averaging circuit

Claims (11)

In a method for identifying a sign of a signal in a receiving circuit of an optical transmission system,
Stage 1 defining the upper threshold, the central threshold and the lower threshold in descending order;
Obtaining a first comparison result which is a result of comparing the signal to be identified and the central threshold;
Obtaining a second comparison result that is a result of comparing the identified signal with either the upper threshold value or the lower threshold value; and
And a step 4 of selecting one of the first and second comparison results as the identification result of the code of the identified signal according to the switching signal based on the magnitude of the intersymbol interference. Identification method. The code identification method according to claim 1, wherein the step 1 includes:
Placing the signal amplitude within a predetermined range prior to code identification;
And determining the upper threshold value, the central threshold value, and the lower threshold value according to the range. 2. A code identification method according to claim 1, wherein said step 1 generates a plurality of different threshold values in accordance with identification signals inputted simultaneously or in the past . 2. The code identification method according to claim 1, wherein the step 1 generates a plurality of different threshold values based on an average of signal amplitudes of the identified signals for each identification result. 2. The code identification method according to claim 1, wherein the step 4 includes:
Sampling the identified signal at the positive phase of the operation clock and obtaining the average of its power,
Sampling the identified signal with the opposite phase of the operating clock and obtaining the average power of the power,
And a step of generating a switching signal based on a comparison between the average A and the average B. In the circuit for identifying the sign of the signal in the receiving circuit of the optical transmission system,
Means 1 for determining the upper threshold value, the central threshold value and the lower threshold value in descending order;
Means 2 for obtaining a first comparison result which is a result of comparing the signal to be identified and the center threshold value;
Means 3 for obtaining a second comparison result which is a result of comparing the identified signal with either the upper threshold value or the lower threshold value;
A code comprising: means for selecting one of the first and second comparison results as the identification result of the code of the identified signal in accordance with the switching signal based on the magnitude of the intersymbol interference. Identification circuit. 7. The code identification circuit according to claim 6, wherein the means 1 includes:
Means for keeping the signal amplitude within a predetermined range prior to code identification;
And a means for determining the upper threshold value, the central threshold value, and the lower threshold value in accordance with the range. 7. The code identification circuit according to claim 6, wherein said means 1 generates a plurality of different threshold values according to identification signals inputted simultaneously or in the past . 7. The code identification circuit according to claim 6, wherein the means 1 generates a plurality of different threshold values based on an average of signal amplitudes of the signals to be identified for each identification result. 7. The code identification circuit according to claim 6, wherein said means 4 includes:
Means for sampling the signal to be identified at the positive phase of the operation clock and obtaining the average of its power;
Sampling the identified signal with the opposite phase of the operating clock and determining the average power of the power;
A code identification circuit comprising: means for generating a switching signal based on a comparison between the average A and the average B. A receiving circuit for an optical transmission system, comprising the code identification circuit according to claim 6.

Images