Granted Patent B2

US 12,362,837 · App. 18/017,890 · Granted Jul 15, 2025

Inventors: Shuto Yamamoto (Musashino, JP); Masanori Nakamura (Musashino, JP); Hiroki Taniguchi (Musashino, JP); Yoshiaki Kisaka (Musashino, JP)

Assignee: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

H04B10/69

Abstract

Claims (161)

1. An optical signal demodulator comprising:

an optical receiver configured to receive an optical signal transmitted from a transmitter via an optical fiber transmission line and convert the optical signal into an analog signal;

an analog to digital (AD) converter configured to convert the analog signal which has been converted by the optical receiver into a digital signal;

a digital signal processor configured to perform waveform shaping of the digital signal which has been converted by the AD converter; and

a signal determinator configured to perform bit determination of the digital signal on which waveform shaping has been performed by the digital signal processor;

wherein the signal determinator comprises:

a spectrum shaping filter processor configured to receive a reception signal sequence of the digital signal from the digital signal processor and output a value obtained by performing a convolution operation on the reception signal sequence which has been received in case that the transmitter transmits an encoded m-value symbol v n generated based on the following Equations (1) and (2) in case that

n is a parameter representing a timing,

u n ∈{0, 1, 2, . . . , m−1} is an m-value symbol before encoding, and

α is any real number satisfying 0<α<1;

a transmission line simulation filter processor configured to output a value obtained by performing a convolution operation on a candidate signal sequence which has been determined based on a symbol transition of the reception signal;

an error information acquisitor configured to acquire error information between the value output from the spectrum shaping filter processor and the value output from the transmission line simulation filter processor;

a symbol likelihood calculator configured to calculate a symbol likelihood indicating a likelihood of the symbol before encoding, the calculation of the symbol likelihood being performed based on the error information acquired by the error information acquisitor; and

a bit determinator configured to determine a value of a bit included in the symbol before encoding based on the symbol likelihood calculated by the symbol likelihood calculator,

wherein the candidate signal sequence is a sequence (r n−1 , r n ) indicating a transition of the encoded symbol,

wherein the symbol before encoding is uniquely determined corresponding to the candidate signal sequence,

wherein the error information acquisitor is configured to calculate a value to which a factor mr n−1 is added representing an influence from a previous symbol sequence represented by the following Equations (5) to (7) for a square error of the value output from the spectrum shaping filter processor and the value output from the transmission line simulation filter processor in case that the error information is lr n−1 r n in case that in case that certain timings n−1 and n are focused,

wherein the symbol likelihood calculator is configured to calculate a likelihood of each symbol before encoding by a sum of likelihoods calculated from the error information acquired by the error information acquisitor, the error information being acquired from a value obtained by performing a convolution operation on the candidate signal sequence including same symbol before encoding and the value output from the spectrum shaping filter processor, and

wherein the Equations (1), (2), (5), (6) and (7) are as follows:

v n =ũ n +└αv n−1 ┘  (1)

ũ n =mod( u n −└αv n−1 ┘, m )  (2)

m

~

r

n

-

1

=

min

r

n

-

2

l

r

n

-

2

⁢

r

n

-

1

(

5

)

m

r

n

-

1

=

m

~

r

n

-

1

-

min

r

n

-

1

⁢

m

~

r

n

-

1

(

6

)

m r 0 =0  (7).

2. The optical signal demodulator according to claim 1 ,

wherein, in case that a multi-value number of the symbol before encoding is 4, and in case that a multi-value number of the symbol after encoding is 6, the symbol likelihood calculator is configured to calculate a likelihood of each symbol before encoding based on following Equations of U u in case that a group of the candidate signal sequence (r n−1 , r n ) including same symbol u n before encoding is the U u , and

the Equations of U u are as follows:

U 0 ={(0,0), (1,0), (2,4), (3,4), (4,4), (5,4)}

U 1 ={(0,1), (1,1), (2,1), (3,1), (4,5), (5,5)}

U 2 ={(0,2), (1,2), (2,2), (3,2), (4,2), (5,2)}

U 3 ={(0,3), (1,3), (2,3), (3,3), (4,3), (5,3)}.

3. The optical signal demodulator according to claim 1 , wherein:

the signal determinator further includes a bit likelihood calculator configured to calculate a bit likelihood indicating a likelihood of a value of a bit included in the symbol before encoding from the symbol likelihood calculated by the symbol likelihood calculator, and

the bit likelihood calculator is configured to calculate a log likelihood ratio of a value of a bit for each bit included in the symbol before encoding.

4. The optical signal demodulator according to claim 3 , wherein

a soft-determination error correction code is applied to the optical signal transmitted by the transmitter, and

the signal determinator further includes an error correction decoder configured to perform error correction decoding using the bit likelihood that has been calculated by the bit likelihood calculator.

5. A control method of an optical signal demodulator, the control method comprising

receiving an optical signal transmitted from a transmitter via an optical fiber transmission line and converting the optical signal into an analog signal;

converting the analog signal which has been converted into a digital signal;

performing waveform shaping of the digital signal which has been converted;

performing bit determination of the digital signal on which waveform shaping has been performed;

receiving a reception signal sequence of the digital signal from the digital signal processor and outputting a value obtained by performing a convolution operation on the reception signal sequence which has been received in case that the transmitter transmits an encoded m-value symbol v n generated based on the following Equations (1) and (2) in case that

n is a parameter representing a timing,

u n ∈{0, 1, 2, . . . , m−1} is an m-value symbol before encoding, and

α is any real number satisfying 0<α<1;

outputting a value obtained by performing a convolution operation on a candidate signal sequence which has been determined based on a symbol transition of the reception signal;

acquiring error information between the value output from the step of receiving a reception signal and the value output from the step of outputting a value;

calculating a symbol likelihood indicating a likelihood of the symbol before encoding, the calculation of the symbol likelihood being performed based on the error information acquired; and

determining a value of a bit included in the symbol before encoding based on the symbol likelihood calculated,

wherein the candidate signal sequence is a sequence (r n−1 , r n ) indicating a transition of the encoded symbol,

wherein the symbol before encoding is uniquely determined corresponding to the candidate signal sequence,

wherein a value to which a factor mr n−1 is added representing an influence from a previous symbol sequence represented by the following Equations (5) to (7) is calculated for a square error of the value obtained by performing the convolution operation on the reception signal sequence and the value obtained by performing a convolution operation on a candidate signal sequence in case that the error information is lrn−1rn in case that in case that certain timings n−1 and n are focused,

wherein a likelihood of each symbol before encoding is calculated by a sum of likelihoods calculated from the error information acquired, the error information being acquired from a value obtained by performing a convolution operation on the candidate signal sequence including same symbol before encoding and the value obtained by performing a convolution operation on the reception signal sequence, and

wherein the equations (1), (2), (5), (6) and (7) are as follows:

v n =ũ n +└αv n−1 ┘  (1)

ũ n =mod( u n −└αv n−1 ┘, m )  (2)

m

~

r

n

-

1

=

min

r

n

-

2

l

r

n

-

2

⁢

r

n

-

1

(

5

)

m

r

-

1

=

m

˜

r

n

-

1

-

min

r

n

-

1

m

˜

r

n

-

1

(

6

)

m r 0 =0  (7).

6. A non transitory computer readable recording media which stores a program causing a computer to execute the control method according to claim 5 .

Assignments (2)

Continuity (1)

References Cited (13)