JP4351508B2 - Duobinary pulse shaping for optical transmission systems using pulse amplitude modulation techniques - Google Patents

Description

  The present invention relates generally to optical communication systems, and more particularly to optical communication systems that use duobinary multi-level pulse amplitude modulation techniques.

  This application claims the benefit of US Provisional Application No. 60 / 412,870, filed September 23, 2002.

  The explosive growth of digital communication technology has further increased the demand for bandwidth to carry digital information such as data, voice, and / or video information. To keep pace with the increased demand for bandwidth, new or improved network components and technologies that function effectively at increasing data rates must be constantly developed. However, in optical communication systems, the cost of deploying improved optical components is prohibitively high at such higher data rates. For example, the cost of deploying a 40 Gbps optical communication system is estimated to exceed ten times the cost of an existing 10 Gbps optical communication system. On the other hand, the achievable throughput increases only four times.

  Thus, much research in the field of optical communications has attempted to obtain higher throughput from existing optical technologies. In conventional optical transmission systems, dispersion compensation usually requires an expensive optical compensator. To alleviate the need for a dispersion compensation module (DCM), a duobinary pulse shaping technique was considered for the optical channel. The duobinary pulse shaping technique reduced the required signal bandwidth by a factor of two by mapping the binary data signal to a three level signal.

Typically, a duobinary modulation format uses two signal levels (+1 and -1) to represent a binary value of 1 and one signal level (0) to represent a binary value of zero. In order to minimize dispersion sensitivity and enable high-density wavelength division multiplexing techniques, narrow spectral width duobinary signals have been developed compared to conventional non-zero return (NRZ) binary signals. . In contrast to standard non-zero return (MRZ) signals, it has been observed that modulating the optical signal with a duobinary electrical signal reduces the optical spectrum by a factor of two.
US Provisional Application No. 60 / 412,870 Kazushige Yonenaga and Shigeru Kuwano, “Dispersion-Tolerant Optical Transmission System Usage Duobinary Transmitter and Binary Receiver.” of Lightwave Technology, vol. 15, no. 8, pp. 1530-1537 (August 1997)

  To further improve spectral efficiency, an optical communication system that uses duobinary pulse shaping techniques is needed. Among the benefits, improved spectral efficiency allows for wider tolerance for dispersion and the use of comprehensive and available optical technologies. For example, by improving spectral efficiency by a factor of 4, a 40 Gbps optical communication system is provided using an existing 10 Gbps optical communication system.

  In general, a duobinary optical communication system is disclosed that further improves spectral efficiency using pulse amplitude modulation (PAM) techniques. The disclosed PAM duobinary optical transmitter uses PAM techniques to convert multiple input bits into an N level signal and add the current N level signal to the preceding N level signal to generate a 2N-1 level signal. The 2N-1 level signal is converted to an optical signal for transmission to the receiver. Multiple input bits can be converted to an N-level signal by adding the current multiple input bits to the preceding multiple output bits.

  Similarly, the disclosed PAM duobinary optical receiver is capable of powering the received optical signal (encoded using pulse amplitude modulation and duobinary encoding techniques to encode multiple bits). The level is detected, and the detected power level is mapped to a plurality of bits and returned to the transmission information. By detecting the power level of the received optical signal, the optical signal can be converted into an electrical signal. A square operation can be performed to convert the optical signal into an electrical signal. The received 2N-1 level signal is automatically converted to an N level signal by the square operation.

In one exemplary implementation, the PAM-4 modulation technique is combined with a duobinary pulse shaping technique to improve the spectral efficiency by a factor of 4 by reducing the bandwidth of the optical signal. In this system, within the same bandwidth, for example, a 2.5 Gbps optical communication system can be improved to provide a 10 Gbps optical communication system, and a 10 Gbps optical communication system can be improved to provide a 40 Gbps optical signal system. Can be provided.
A more complete understanding of the present invention, as well as other features and advantages of the present invention, will be obtained by reference to the following detailed description and drawings.

  FIG. 1 is a schematic block diagram of an exemplary conventional duobinary optical communication system 100 that encodes each bit using three levels. As shown in FIG. 1, the input bit is applied to one terminal of an exclusive OR (XOR) gate 110 and the preceding bit is applied to the second terminal of the XOR gate 110. The leading bit is obtained by a feedback loop that includes a delay element 150. XOR gate 110 and delay element 150 include a precoder 115, which introduces correlation into the transmitted signal by adding the preceding output bits to the current input bits. Therefore, the precoder 115 is said to perform a 1 / (1 + D) {mod2} operation. D is a delay operation that delays a sequence of bits by one bit interval.

  The output of the precoder 115 is a 2-level value (0 or 1), which is applied to the encoder 120, which maps the 2-level value to a 3-level value. Specifically, for a given bit interval, encoder 120 adds both the previous and current bit levels according to the following table to generate a three level value (0, 1, 2).

  The tri-level value generated by encoder 120 is applied to Mach-Zehnder modulator 130. This Mach-Zehnder modulator maps each of the three levels (0, 1, 2) to the corresponding voltage level (-V, 0, + V) in a known manner. The generated voltage level is then converted to an optical signal (−e, 0, + e) and applied to the optical fiber 135.

As shown in FIG. 1, the receiver includes a power detector 140 that decodes the received optical signal. Normally, the power detector 140 performs photoelectric conversion by performing a square operation on the received signal. Since the received signal can have one of three optical values (−e, 0, + e), the square operation returns the signal to + e ² or 0. The two levels generated by power detector 140 are then mapped to the corresponding 1-bit binary value. In this scheme, the Mod2 operation at the receiver is obtained automatically by going through the mapping table and from the optical domain to the electrical domain (-V is equivalent to + V when measuring intensity). Accordingly, the bits at the output of the duobinary optical communication system 100 are equal to the input bits.

Duobinary optical communication system 100 reduces the bandwidth of the optical spectrum by a factor of two, thus improving performance in the presence of dispersion. The overall conversion function of the duobinary optical communication system 100 is expressed as follows.
1 / (1 + D) × (1 + D) {mod2} = 1 {mod2}

  For a more detailed discussion of a suitable conventional duobinary optical communication system 100, see, for example, Kazushige Yonenaga and Shigeru Kuwano, “Dispersion-Tolerant Optical Trans-Banding System Transmission Bistimating System. "J. of Lightwave Technology, vol. 15, no. 8, pages 1530-1537 (August 1997).

  According to one aspect of the invention, the duobinary pulse shaping technique is used in conjunction with a pulse amplitude modulation (PAM) technique. Thus, the present invention recognizes that the improvement in spectral efficiency obtained using duobinary pulse shaping techniques can be further improved by utilizing PAM modulation. In one exemplary implementation, the PAM-4 modulation technique is combined with a duobinary pulse shaping technique to improve the spectral efficiency by a factor of 4 by reducing the bandwidth of the optical signal. In this system, for example, within the same bandwidth, the 2.5 Gbps optical communication system can be improved to provide a 10 Gbps optical communication system, and the 10 Gbps optical communication system can be improved to provide a 40 Gbps optical communication system. Can be provided.

  FIG. 2 is a schematic block diagram of an exemplary PAM-4 duobinary optical communication system 200 incorporating features of the present invention. As shown in FIG. 2, in the exemplary PAM-4 implementation, two input bits comprising four levels are applied to one terminal of a modulo 4 adder 210, and the preceding 4-level value is added to the adder 210. To the second terminal. The leading four level value is obtained by a feedback loop that includes a delay element 250. Adder 210 and delay element 250 include a precoder 215 that introduces correlation into the transmitted signal by adding the preceding 4-level output value to the current 4-level input value. Thus, the precoder 215 is said to perform a 1 / (1 + D) {mod4} operation, and the output is a 4-level value.

  The 4-level output of precoder 215 is applied to encoder 220, which maps the 4 levels to 7-level values (0, 1, 2, 3, 4, 5, 6). Specifically, for a given bit interval, encoder 220 adds the previous and current 4-level values together according to the following table to generate a 7-level value.

  The 7 level value generated by encoder 220 is applied to a digital to analog converter (DAC) 225, which maps the 7 level value to a corresponding voltage. For example, the digital to analog converter 225 may use the following exemplary mapping:

(0, 1, 2, 3, 6, 5, 4) → (-3V, -2V, -V, 0, V, 2V, 3V)
Note that other mappings by the digital to analog converter 225 are possible. Next, the Mach-Zehnder modulator 230 outputs optical signals (-3e, -2e, -e, 0, 0) corresponding to the seven voltage levels (-3V, -2V, -V, 0, V, 2V, 3V), respectively. e, 2e, 3e) and the optical signal is applied to the optical fiber 235.

As shown in FIG. 2, the receiver includes a power detector 240 that decodes the received optical signal. In general, the power detector 240 performs photoelectric conversion by performing a square operation on the received signal. Since the received signal can have one of seven optical values (−3e, −2e, −e, 0, e, 2e, 3e), by square operation, the signal is + 9e ² , + 4e ² , + e Return to ² or 0. Slicer 260 then maps the four levels generated by power detector 240 to the corresponding 2-bit binary values. In this scheme, the Mod4 operation at the receiver is automatically acquired from the optical domain to the electrical domain by advancing the slicer 260 mapping table (when measuring the intensity, -V is equivalent to + V. ). Thus, two bits at the output of the PAM-4 duobinary optical communication system 220 are equal to the input bits.

The PAM-4 duobinary optical communication system 200 reduces the bandwidth of the optical spectrum by a factor of four, thus improving performance when dispersion is present. The overall conversion function of the duobinary optical communication system 200 is expressed as follows.
1 / (1 + D) × (1 + D) {mod4} = 1 {mod4}

  The embodiments and variations shown and described herein are merely illustrative of the principles of the present invention and various modifications may be made by those skilled in the art without departing from the scope and spirit of the invention. Please understand that is possible.

1 is a schematic block diagram of an exemplary conventional duobinary optical communication system. 1 is a schematic block diagram of an exemplary PAM-4 duobinary optical communication system incorporating features of the present invention. FIG.

Claims (10)

A method for transmitting information,
Converting a plurality of input bits into an N-level signal;
Adding the current N-level signal to the preceding N-level signal to generate a 2N-1 level signal , so that the signal received after the squaring operation at the receiver is represented as a modulo-N operation Generating the 2N-1 level signal mapped to a corresponding voltage value using the reordered set of 2N-1 levels ;
For transmission to the receiver, the method comprising the step of converting said 2N-1 level signal to an optical signal. Wherein said step of converting a plurality of input bits into N level signal, by adding the current plurality of input bits to the preceding plurality of output bits, further comprising the step of generating an N-level signal, The method of claim 1 . The additional step is as follows:
7 level values are mapped to corresponding voltage values as (0, 1, 2, 3, 6, 5, 4) → (−3V, −2V, −V, 0, V, 2V, 3V), respectively. The method of claim 1, further comprising a step. A method of receiving information,
Receiving an optical signal encoded using a pulse amplitude modulation technique and a duobinary encoding technique to encode a plurality of bits;
Detecting a power level of the received optical signal;
Mapping the detected power level to a plurality of bits , the 2N-1 level signal being the 2N-1 so that the received signal is represented as a modulo N operation after performing a square operation Mapping using a reordered set of levels to map to a corresponding voltage value .   The method of claim 4, further comprising converting the optical signal to an electrical signal using a square operation on the optical signal. A system for transmitting information,
A precoder for converting a plurality of input bits into an N-level signal;
An adder that adds a current N-level signal to a preceding N-level signal to generate a 2N-1 level signal, the signal received after performing a squaring operation at the receiver is represented as a modulo-N operation Such that the 2N-1 level signal is mapped to a corresponding voltage value using the reordered set of 2N-1 levels ;
A system comprising a digital-to-analog converter for converting the 2N-1 level signal into an optical signal for transmission to the receiver.   The system of claim 6, wherein the precoder comprises an adder that adds the current plurality of input bits to the preceding plurality of output bits to generate the N-level signal. A system for receiving information,
An input port for receiving an optical signal encoded using pulse amplitude modulation and duobinary encoding techniques to encode a plurality of bits;
A power detector for detecting a power level of the received optical signal;
A slicer that maps the detected power level to a plurality of bits , the 2N-1 level signal being the 2N-1 so that the received signal is represented as a modulo N operation after performing a square operation And a slicer that is mapped to a corresponding voltage value using a sorted set of levels .   The system of claim 8, wherein the power detector converts the optical signal into an electrical signal.   The system of claim 9, wherein the photoelectric conversion performs a square operation on an optical signal.

Images