WO2014097583A1 - 周波数オフセット補償装置および周波数オフセット補償方法 - Google Patents
周波数オフセット補償装置および周波数オフセット補償方法 Download PDFInfo
- Publication number
- WO2014097583A1 WO2014097583A1 PCT/JP2013/007289 JP2013007289W WO2014097583A1 WO 2014097583 A1 WO2014097583 A1 WO 2014097583A1 JP 2013007289 W JP2013007289 W JP 2013007289W WO 2014097583 A1 WO2014097583 A1 WO 2014097583A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- frequency offset
- output data
- fft
- offset compensation
- order
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/60—Receivers
- H04B10/61—Coherent receivers
- H04B10/616—Details of the electronic signal processing in coherent optical receivers
- H04B10/6164—Estimation or correction of the frequency offset between the received optical signal and the optical local oscillator
Definitions
- the present invention relates to a frequency offset compensation technique for an optical coherent receiver in optical communication.
- local light in which the frequency and phase of the carrier wave completely coincide with each other is required at the receiver.
- a non-zero intermediate frequency that is, a frequency offset
- the oscillation frequency error of a laser used for commercial use is ⁇ 2.5 GHz at the maximum, so that the frequency offset reaches a maximum of ⁇ 5 GHz.
- FIG. 2 is a configuration example of an optical digital coherent receiver by digital signal processing.
- the input light is mixed with the light from the local oscillation laser 201 by the optical frequency mixer 200, and proceeds to the PD (photodiode) 202 and the ADC (analog / digital converter) 203.
- PD photodiode
- ADC analog / digital converter
- the frequency offset estimator 207 calculates a frequency offset compensation amount, and the frequency offset compensator 204 performs the compensation.
- the compensation is performed by complex multiplication of phases having the same phase amount and opposite values in order to cancel the phase added as an offset.
- the speed of optical communication which is being researched and developed, is over 100 Gbps, while the digital signal processing unit 210 can be operated only at a few GHz at most.
- a plurality of the same circuits are arranged and compensated.
- the frequency offset compensator 204 a plurality of complex multipliers for multiplying the opposite phases are arranged. More specifically, 100 or more complex multipliers are prepared, and the scale thereof is several megagates.
- the digital signal processing unit 210 includes an equalizer 205 for performing chromatic dispersion compensation and polarization separation, an integrator 206, a phase estimator 208 for estimating the phase of the carrier wave, Various circuits such as data restoration 209 are incorporated.
- the chromatic dispersion compensation included in the equalizer 205 is performed in the frequency domain because the circuit scale is too large for a time domain FIR (Finite Impulse Response) filter. Therefore, the digital signal processing unit 210 also includes an FFT (Fast Fourier Transform) that converts time-axis data into a frequency, and an IFFT (Inverse Fast Fourier Transform) that performs the reverse process.
- FFT Fast Fourier Transform
- IFFT Inverse Fast Fourier Transform
- FIG. 3 is a diagram in which a 4096-point FFT / IFFT is decomposed into two 64-point FFTs 302 using the Prime Factor method.
- FFT block 301 in FIG. 3 is implemented as physical hardware.
- FFT block 300 instead of 128 parallels (FFT block 301).
- the 64-point FFT may be decomposed into 8 ⁇ 8 by the Cooley-Tukey method, other methods, or the Prime Factor method.
- FIG. 6 is an enlarged view of the Analog / Digital Converter (hereinafter referred to as ADC 203), the frequency offset compensator 204, and the equalizer 205 in FIG.
- the equalizer 205 includes an FFT 602 for performing chromatic dispersion compensation in the frequency domain, a filter operation (complex multiplier) 603, and an IFFT 604.
- FFT 602 for performing chromatic dispersion compensation in the frequency domain
- filter operation complex multiplier
- many functions such as polarization separation are implemented.
- time 1 represents the sampling interval of the ADC.
- 64 pieces of data output from the ADC 203 are arranged in one clock cycle as shown in FIG. 4 and input to the digital signal processing unit 210.
- the data to be input to the FFT is the data input in the first cycle x [0], x [64],..., X [63 ⁇ 64] and the data input in the second cycle.
- x represents an input signal to the FFT
- y represents an output signal of the FFT. Comparing FIG. 4 and FIG. 5, it can be seen that the output order from the ADC and the input order to the FFT are different. Therefore, it is necessary to rearrange the data, and the rearrangement memory (1) 601 and rearrangement memory (2) 605 are required as shown in FIG.
- a memory block as shown in FIG. 7 is prepared.
- 64 pieces of 1R1W memory 701 having a width of one data and a depth of 64 words are arranged side by side (the 702 in which 64 pieces of 1R1W memory are arranged), and two sets (from the rearrangement memory 303).
- the 1R1W memory 701 is a general 1R1W memory having a width of one data and a depth of 64 words (a memory capable of one read process and one write process in one cycle).
- a value is written at the position shown in the upper diagram of FIG. [A, B] of FIG. 8 shows the output of the FFT column on the left side of FIG. That is, the 64 values [0, 0], [0, 1],..., [0, 63] calculated in the first cycle are the positions on the diagonal line described in bold in the upper diagram of FIG. Is written to. Similarly, [1, 0], [1, 1],..., [1, 63] in the second cycle are sequentially written in the memory in the upper diagram of FIG. Such an operation is continued 64 times to fill the memory. Conversely, when reading, the position surrounded by the dotted line in the upper diagram of FIG. 8 is read.
- y [4095] is output at the 64th cycle, but in order to treat it as y [0] and to treat y [0] as y [1], the delay is at least 64 cycles or more. There is a need. In other words, a memory that can hold at least 4096 data is required even when shifting right by one, and such a memory requires several hundred kilogates.
- the problem with frequency offset compensation is that a complex multiplier is added when compensating in the time domain, and a memory for delaying when compensating in the frequency domain, in addition to various compensation circuits. It is to be done. As a result, the circuit scale of the digital signal processing unit increases, and as a result, problems such as an increase in power consumption and a decrease in chip yield occur.
- the present invention has been made in view of the above problems, and an object of the present invention is to provide a frequency offset compensation apparatus and a frequency offset compensation method for performing frequency offset compensation without requiring a new complex multiplier or a new memory. Is to realize.
- the present invention is an invention for compensating for a frequency offset, and is not for obtaining a compensation amount. That is, it is an invention of how to perform frequency offset compensation after a frequency offset compensation amount is given.
- the frequency offset compensator comprises: a first FFT unit that performs a discrete-time Fourier transform of an input signal; a second FFT unit that performs a discrete-time Fourier transform; and an output data order of the first FFT unit.
- An offset compensation unit comprising: an offset compensation unit that changes according to a frequency offset compensation amount and rearranges the output data in an order according to the frequency offset compensation amount and inputs the rearranged output data to the second FFT unit. It is.
- the order of the output data of the first FFT process for performing the discrete time Fourier transform of the input signal, the second FFT process for performing the discrete time Fourier transform, and the first FFT process is there.
- FIG. 1 It is a block diagram which shows the structure of the 1st and 2nd embodiment of this invention. It is a structural example of the optical digital coherent receiver by digital signal processing. This is a circuit that performs 4096 point FFT / IFFT using the Prime Factor method. It is an example of 64 parallel output data from ADC. This is a data string to be input to the FFT block. It is the block diagram which described the flow of the process from ADC. It is the figure which showed the write position to the memory for rearrangement in FFT / IFFT. It is the figure which showed the operation
- FIG. 10 is a diagram obtained by replacing FIG. 10 with the input to the second FFT instead of the FFT output order. It is the figure which described the output order of FFT at the time of performing frequency offset compensation with respect to left 64 shift and left 65 shift. It is the figure which replaced FIG. 12 with the input to 2nd FFT instead of the output order of FFT. It is the figure which showed the value of the 1st (1st clock cycle) rearrangement memory at the time of 1 shift left. It is the figure which showed the value of the 2nd (2nd clock cycle) rearrangement memory at the time of 1 shift left.
- the frequency offset compensator of this embodiment includes a first FFT unit 101 that performs a discrete-time Fourier transform of an input signal and a second FFT unit 105 that performs a discrete-time Fourier transform. Further, the order of the output data of the first FFT means 101 is changed according to the frequency offset compensation amount 109, and the output data in the order according to the frequency offset compensation amount 109 is rearranged and input to the second FFT means 105.
- the offset compensation means 110 is provided.
- FIG. 1 is a block diagram showing a configuration of a frequency offset compensator according to a second embodiment of the present invention.
- the FFT block 100 serving as the frequency offset compensation device of this embodiment includes a first FFT unit 101 that performs a discrete-time Fourier transform of an input signal and a second FFT unit 105 that performs a discrete-time Fourier transform.
- the order of the output data of the first FFT unit 101 is changed according to the frequency offset compensation amount 109, the output data in the order corresponding to the frequency offset compensation amount 109 is rearranged and input to the second FFT unit 105.
- Offset compensation means 110 Offset compensation means 110.
- the offset compensation unit 110 includes a counter 108 that counts the order of input data. Further, a rearrangement memory 102 for appropriately inputting the output result of the first FFT means 101 to the second FFT means 105 is provided. Further, a write position determination unit 107 that determines a write position of the rearrangement memory 102 based on the value of the counter 108, and a read position that changes the read order to the rearrangement memory 102 based on the frequency offset compensation amount 109. Determination means 106. Furthermore, a first rearrangement unit 103 that rearranges the output of the rearrangement memory 102 based on the value of the counter 108 and the frequency offset compensation amount 109 is provided. Furthermore, a second rearrangement unit 104 that rearranges the results read from the first rearrangement unit 103 based on the value of the counter 108 and generates data to be input to the second FFT unit 105 is provided.
- the offset compensation unit 110 includes a rearrangement memory 102, a counter 108, a write position determination unit 107, a read position determination unit 106, a first rearrangement unit 103, and a second rearrangement unit 104.
- the frequency offset compensator includes a first FFT unit 101, a second FFT unit 105, and the offset compensation unit 110 between the first FFT unit 101 and the second FFT unit 105.
- the frequency offset compensation is performed by shifting the signal (before offset compensation) 901 to the left or right by the amount to be offset as shown in FIG. It is realized with. Therefore, if the signal y is the result of FFT, when p-shifting to the left, y [(np)% 4096] (% indicates the remainder when np is divided by 4096) and pshifting to the right If so, y [(n + p)% 4096] may be calculated. In other words, when shifting 1 to the right, the FFT result y [4095] may be handled as y [0], y [0] as y [1], y [1] as y [2], and the like.
- y [64],..., Y [4032], y [0] are output in the 64th cycle. Accordingly, the subsequent processing block processes y [1], y [65], ..., y [4033] as y [0], y [64], ..., y [4032]. Therefore, it is possible to perform the same operation as when one shift is made to the left.
- FIG. 10 is a table focusing on the output of the FFT.
- FIG. 11 shows a table obtained by converting the output to the FFT on the right side of FIG.
- frequency offset compensation is realized by making the order of input to the right FFT of FIG. 3 as shown in FIG.
- the first FFT means 101 and the second FFT means 105 are means for individually performing FFT on input data and outputting the result.
- the FFT implementation method may be a general Cookie-Tukey (Non-Patent Document 2) method or a method of theoretically calculating a discrete-time Fourier transform.
- the first FFT means 101 and the second FFT means 105 correspond to FIG. 3, the first FFT means 101 points to one 64-point FFT 302 on the left side, and the second FFT means points to one 64-point FFT 302 on the right side. Point to.
- the rearrangement memory 102 is a memory (rearrangement memory 303) for performing appropriate data exchange between the FFT group on the left side and the FFT group on the right side in the FFT block 300 of FIG. Details of the operation are as described with reference to FIGS. As described with reference to FIG. 7, the operation of the write position determining means 107 also determines the write position of the data output from the first FFT means 101 as shown in FIG. 8 according to the value of the counter 108. It is. In FIG. 8, the upper diagram is a table focusing on the output of the left FFT in FIG. 3 (for example, corresponding to FIG. 10), and the lower diagram is a table converted to the input to the right FFT in FIG. (Corresponding to FIG. 11).
- the second rearrangement unit 104 rearranges values read from the rearrangement memory 102.
- a novel feature of the present embodiment is that a reading position determination unit 106 and a first rearrangement unit 103 are added. Hereinafter, the operation of these two means will be described in detail.
- the reading position determination means 106 changes the reading order of the rearrangement memory 102 based on the frequency offset compensation amount 109.
- frequency offset compensation 64 cycles are read in order from the top of the address as shown in FIG.
- the input data [0, 0] for y [0], y [64],. ], [1, 0],..., [63, 0] should be supplied to the second FFT means 105. Changes it to [0, 1], [1, 1],..., [63, 1]. That is, it operates so that the read address starts from 1 instead of 0. In the last 64th cycle, an operation is performed so as to read out address 0.
- the data to be input to the second FFT means 105 is in the state of rotating 1 left from [1, 0], [2, 0],..., [63, 0], [0, 0] from FIG. is there.
- the first rearrangement means 103 performs the left 2 rotation from 1 to 62 cycles, and the 63 and 64 cycles perform the left 3 rotation.
- FIG. 12 shows the output signal order of FFT.
- FIG. 13 shows a table in which this is converted into the input to the FFT on the right side of FIG. 3 in the same manner as FIG.
- the reading position determination unit 106 reads out the rearrangement memory 102 from the address 0 as in the case where frequency offset compensation is not performed, but the first rearrangement unit 103 always performs left 1 rotation.
- the reading position determining means 106 performs 2 left rotations from the 1st to 63rd cycles and 3 left rotations in the 64th cycle.
- the reading position determining means 106 reads in the order of addresses 62, 63, 0, 1,... 61, and the first rearranging means 103 rotates the right 3 in the first and second cycles. Rotate right 2 for 3 to 64 cycles.
- the reading position determination means 106 reads in the order of addresses 0, 1,... 62, 63, and the first rearrangement means 103 performs right 1 rotation in every cycle. In the case of 65 shift to the right, the reading position determination means 106 performs reading in the order of addresses 63, 0, 1,..., 62, and rotates the right 3 rotations in the first cycle and the right 2 rotations in the second to 64th cycles. do it.
- the first rearrangement unit 103 and the second rearrangement unit 104 are described separately for easy understanding of the effect.
- the rearrangement memory 102 is based on the counter value and the frequency offset amount. May be integrated as one rearrangement means that operates to rearrange the outputs.
- the rearrangement memory 102 changes as shown in FIG.
- the FFT result is written at the position on the diagonal line written in bold, and at the same time, the address 1 of the memory on the read side is read.
- the data are rearranged by the first rearranging means 103 and then supplied to the second FFT means 105.
- the FFT result of the second FFT unit 105 is y [0], y [64],..., Y [63 ⁇ 64] as shown in FIG.
- the second rearranging means 104 does nothing, and the first rearranging means 103 performs left one rotation.
- the rearrangement memory 102 changes as shown in FIG.
- the FFT result is written at the position indicated in bold, and at the same time, the address 2 of the memory on the reading side is read.
- the second rearrangement unit 104 rotates left one, and the first rearrangement unit 103 also performs left one rotation and is supplied to the second FFT unit 105.
- the result of the second FFT means 105 is y [1], y [65],..., Y [63 ⁇ 64 + 1] as shown in FIG.
- FIG. 16 shows the operation in the 63rd cycle.
- the rearrangement memory changes as shown in FIG.
- the FFT result is written at the position indicated in bold, and at the same time, the address 0 of the memory on the reading side is read.
- the first rearrangement means 103 rotates left 2 and the second rearrangement means rotates 63 left (rotate left 1 together) and supplies it to the second FFT means 105.
- the result of the second FFT is y [63], y [64],..., Y [63 ⁇ 64 + 63] as shown in FIG.
- the rearrangement memory changes as shown in FIG.
- the positions of the memory to be written and the memory to be read are reversed, but other operations are the same as those in the first cycle.
- the address written to the memory and the address read from the memory described in the present embodiment are examples, and the present invention is not limited thereto. Similarly, the number of right / left rotations is not limited thereto.
- the reading order of the rearrangement memory 102 provided in the FFT block 100 is changed according to the frequency offset compensation amount 109. Furthermore, a first rearrangement unit 103 that appropriately rearranges data read from the rearrangement memory according to the frequency offset compensation amount 109 is provided. Thereby, it is not necessary to provide a new memory for frequency offset compensation in the frequency domain, and frequency offset compensation can be performed. On the other hand, in the present embodiment, the reading position determining means 106 and the first rearranging means 103 are required, but providing them is overwhelmingly less than providing a new memory. (Third embodiment) A third embodiment of the present invention will be described with reference to the drawings. In the second embodiment, the frequency offset compensation is realized by rearranging necessary data by changing the order of reading from the memory according to the compensation amount. However, in the present embodiment, the order of writing this is changed. Realize with.
- FIG. 19 is a block diagram showing the configuration of the frequency offset compensation apparatus of the present embodiment.
- a first FFT unit 101 and a second FFT unit 105 that perform a small FFT and a counter 108 that counts the order of input data are included in an FFT block 100 that performs a discrete-time Fourier transform serving as a frequency offset compensator.
- it is provided between the first FFT unit 101 and the second FFT unit 105, and includes a rearrangement memory 102 for appropriately inputting the output result of the first FFT unit 101 to the second FFT unit 105.
- a read position determining means 1901 is provided for determining a read position to the rearrangement memory 102 based on the value of the counter 108. Furthermore, a writing position determining means 1902 for determining a writing position to the rearrangement memory 102 based on the value of the counter 108 and the frequency offset amount is provided. Furthermore, a first rearrangement unit 1903 that rearranges the output of the rearrangement memory 102 based on the counter 108 and the frequency offset compensation amount 109 is provided. Furthermore, the second rearrangement unit 104 rearranges the data output from the first rearrangement unit 1903 based on the value of the counter 108.
- the offset compensation unit 111 includes a rearrangement memory 102, a counter 108, a write position determination unit 1902, a read position determination unit 1901, a first rearrangement unit 1903, and a second rearrangement unit 104.
- the frequency offset compensator includes a first FFT unit 101, a second FFT unit 105, and the offset compensation unit 111 between the first FFT unit 101 and the second FFT unit 105.
- the first FFT unit 101, the second FFT unit 105, the counter 108, the rearrangement memory 102, and the second rearrangement unit 104 operate in the same manner as in the second embodiment.
- a novel feature of the present embodiment is that a write position determination unit that determines a position at which the result of the first FFT unit 101 is written in the rearrangement memory 102 based on the frequency offset compensation amount 109. 1902 and first rearranging means 1903.
- FIG. 20 shows the positions where the write position determination means 1902 writes the result of the first FFT means 101 in the rearrangement memory 102, taking the left 1 shift, the left 2 shift, the right 1 shift, and the right 2 shift as an example. If the writing position is left X shift, the result of the first FFT means is rotated to the right (X% 64), and if it is right X shift, (X% 64) left is rotated and written to the memory.
- the first rearranging means 1903 rearranges so that the result is the same as that in FIG. That is, in the case of 1 shift to the left, nothing is rearranged in the 1st to 63rd cycles, and the left is rotated only in the 64th cycle. In the case of 2 shifts to the left, nothing is rearranged in the 1st to 62nd cycles, and the left 1 rotation is performed only in the 63th and 64th cycles. Further, in the case of 1 shift to the right, the right 1 is rotated only in the case of 1 cycle, and nothing is rearranged in the case of 2 to 64 cycles. In the case of 2 shifts to the right, the rotation is rotated 1 right only in the 1st and 2nd cycles, and nothing is rearranged in the 3rd to 64th cycles.
- FIG. 21 shows a method in which the reading position determination unit 106 and the first rearrangement unit 103 are placed between the first FFT unit 101 and the second FFT unit 105.
- FIG. 22 shows a method in which the reading position determining means 106 and the first rearranging means 103 are placed between the second FFT means 105 and the third FFT means 2101.
- FIG. 21 and FIG. 22 are the same as those in the second and third embodiments. Further, even when the FFT block 100 is divided into four or more FFTs, the frequency offset compensation can be performed by providing the read position determining unit and the first rearrangement unit between any of the FFTs. .
- the second embodiment is applied to the case where the FFT block is decomposed into three or more.
- the third embodiment can be similarly applied.
- FIG. 23 is a block diagram showing a configuration of the present embodiment. In this embodiment, a forced zero means 2301 is added to the second embodiment. Means other than the forced zero means 2301 are the same as those in the second embodiment.
- the forced zero means 2301 forcibly sets some of the data output from the second FFT means 105 to zero and outputs the other values as they are.
- the location and amount forcing to zero are determined based on the frequency offset compensation amount 109.
- the number of FFT points is 4096, for example, only 2048 is set to 0 in the case of 1 left shift, and only 2047 and 2048 is set to 0 in the case of 2 left shift.
- the case of right 1 shift only 2049 is set to 0, and in the case of right 2 shift, only 2049 and 2050 are set to zero.
- Data to be zeroed by the forced zero means is calculated based on the frequency offset compensation amount 109 and the value of the counter 108.
- the forced zero means 2301 makes the output value zero, as shown in FIG. 24, by the shift operation for the frequency offset, it cannot be expressed before the shift, but can be newly expressed after the shift. This is a frequency domain portion ((1) in FIG. 24).
- the part (2) in FIG. 24 was visible in (1), but theoretically, it has already been cut (cut by the bandpass filter before the ADC). Therefore, when the forced zero means 2301 sets this portion to zero, the operation is more ideal.
- FIG. 25 is a block diagram showing a configuration of the present embodiment.
- frequency offset compensation is performed inside IFFT 2502 instead of FFT 602.
- the operation inside the IFFT 2501 is the same as that of the FFT 101 of the second embodiment.
- the filter operation 2501 it is necessary to change the filter operation 2501 as shown in FIG.
- the dotted line signal (after frequency offset compensation) 900 is multiplied by the dotted line filter characteristic (no offset) in FIG. Just add.
- the frequency offset is performed by IFFT
- a solid line signal (before frequency offset compensation) 901 that is not offset-compensated is input to the filter operation 2501. Therefore, it is necessary that the filter characteristic is also a solid line filter characteristic (with offset) shifted in reverse by the frequency offset.
- the filter calculation 2501 the solid line filter characteristic (with an offset) obtained by shifting the filter characteristic based on the frequency offset compensation amount 109 is multiplied by the result of the FFT 602.
- the effect of frequency offset compensation in IFFT 2502 can be made the same as in the second embodiment.
- This embodiment can be applied to uses such as optical digital coherent communication. Further, it can be applied to a system having a frequency offset such as wireless communication.
- Appendix 1 The first FFT means for performing discrete time Fourier transform of the input signal, the second FFT means for performing discrete time Fourier transform, and the order of the output data of the first FFT means are changed according to the frequency offset compensation amount,
- An offset compensation unit comprising: an offset compensation unit that rearranges the output data in an order corresponding to the frequency offset compensation amount and inputs the rearranged output data to the second FFT unit.
- the offset compensation means writes the output data of the first FFT means and performs rearrangement for input to the second FFT means, and the output to the rearrangement memory means.
- a write position determining means for determining a position to write data; a read position determining means for determining a read order of the output data written in the rearranging memory means based on the frequency offset compensation amount; Based on the frequency offset compensation amount, the output data is rearranged by the first rearrangement unit, the first rearrangement unit rearranging the output data corresponding to the order read in the read order.
- the frequency offset according to claim 1, further comprising: a second rearrangement unit that rearranges data and inputs the data to the second FFT unit. Amortization apparatus.
- the offset compensation means includes counter means for counting the order of the input signals, and the write position determination means determines a position for writing the output data to the rearrangement memory means based on the count value of the counter means. And the reading position determining means determines the reading order of the output data written in the rearranging memory means based on the frequency offset compensation amount, and the first rearranging means includes the counter Based on the count value of the means and the frequency offset compensation amount, the output data is rearranged corresponding to the order read in the read order, and the second rearrangement means sets the count value of the counter means to Based on the second FFT operation, the output data rearranged by the first rearranging unit is rearranged based on the second FFT operation.
- the offset compensation means writes the output data of the first FFT means and performs rearrangement for input to the second FFT means, A write position determining means for determining a position for writing the output data to the rearrangement memory means; a read position determining means for determining a read order of the output data written in the rearrangement memory means; Based on the frequency offset compensation amount, the output data is rearranged by the first rearrangement unit, the first rearrangement unit rearranging the output data corresponding to the order read in the read order.
- the offset compensation means includes a counter means for counting the order of the input signals, and the write position determination means determines a position for writing the output data to the rearrangement memory means based on the frequency offset compensation amount.
- the reading position determining means determines the reading order of the output data written in the rearranging memory means based on the count value of the counter means, and the first rearranging means includes the counter Based on the count value of the means and the frequency offset compensation amount, the output data is rearranged corresponding to the order read in the read order, and the second rearrangement means sets the count value of the counter means to Based on the second FFT operation, the output data rearranged by the first rearranging unit is rearranged based on the second FFT operation.
- a first discrete-time Fourier transform is performed on the input signal, the order of the output data obtained by the first discrete-time Fourier transform is changed according to the frequency offset compensation amount, and the output data in the order corresponding to the frequency offset compensation amount is
- a frequency offset compensation method comprising rearranging and performing a second discrete time Fourier transform on the rearranged output data.
- (Appendix 10) Determining a write position to the memory of the output data subjected to the first discrete time Fourier transform, writing the output data to the write position, determining a reading order of the output data written to the write position; The output data read in the reading order is first rearranged, the output data after the first rearrangement is secondly rearranged, and the second discrete time Fourier transform is performed. 10.
- the frequency offset compensation method according to 9. (Appendix 11) The frequency offset compensation method according to appendix 10, wherein the reading order is determined or the writing position is determined based on the frequency offset compensation amount. (Appendix 12) 12.
- (Appendix 13) 13 The frequency offset compensation method according to one of appendices 9 to 12, wherein at least one of the first discrete-time Fourier transform and the second discrete-time Fourier transform performs an inverse Fourier transform.
- (Appendix 14) 14 The frequency offset compensation method according to one of appendices 9 to 13, wherein a predetermined output value of the second discrete time Fourier transform is set to zero after the second discrete time Fourier transform.
- (Appendix 15) The frequency offset compensation method according to appendix 14, wherein the output value is set to zero based on the frequency offset compensation amount.
- the present invention relates to a frequency offset compensation technique for an optical coherent receiver in optical communication, and can be used for an optical communication system.
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Complex Calculations (AREA)
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
- Optical Communication System (AREA)
Abstract
Description
(第1の実施形態)
本発明の第1の実施形態について図1を参照して説明する。本実施形態の周波数オフセット補償装置は、入力信号の離散時間フーリエ変換を行う第一FFT手段101と、離散時間フーリエ変換を行う第二FFT手段105とを備える。さらに、第一FFT手段101の出力データの順序を周波数オフセット補償量109に応じて変え、前記周波数オフセット補償量109に応じた順序とした前記出力データを並び替えして第二FFT手段105に入力する、オフセット補償手段110を備える。
(第2の実施形態)
図1は、本発明の第2の実施形態の周波数オフセット補償装置の構成を示すブロック図である。本実施形態の周波数オフセット補償装置となるFFTブロック100は、入力信号の離散時間フーリエ変換を行う第一FFT手段101と、離散時間フーリエ変換を行う第二FFT手段105とを備える。さらに、第一FFT手段101の出力データの順序を周波数オフセット補償量109に応じて変え、周波数オフセット補償量109に応じた順序とした出力データを並び替えして第二FFT手段105に入力する、オフセット補償手段110とを備える。
(第3の実施形態)
本発明の第3の実施形態について図面を参照して説明する。第2の実施形態では、周波数オフセット補償を、補償量に応じてメモリから読み出す順序を変えることで必要なデータを並び替えして実現したが、本実施形態では、これを書き込む順序を変更することで実現する。
(第4の実施形態)
本発明の第4の実施形態について図面を参照して説明する。本実施形態は、FFT手段が3つある場合である。図21および図22は、本実施形態の構成を示すブロック図である。図21は、読み出し位置決定手段106、第一並替手段103を第一FFT手段101と第二FFT手段105の間に入れる方法である。また、図22は、読み出し位置決定手段106、第一並替手段103を第二FFT手段105と第三FFT手段2101の間に入れる方法である。
(第5の実施形態)
本発明の第5の実施形態について図面を参照して説明する。図23は、本実施形態の構成を示すブロック図である。本実施形態は、第2の実施形態に強制ゼロ手段2301を付加したものである。強制ゼロ手段2301以外の手段については、第2の実施形態と同じである。
(第6の実施形態)
本発明の第6の実施形態について図面を参照して説明する。図25は、本実施形態の構成を示すブロック図である。本実施形態は、周波数オフセット補償をFFT602ではなくIFFT2502の内部で行う方法である。IFFT2501内部での動作は、第2の実施形態のFFT101の場合と同じである。
(付記1)
入力信号の離散時間フーリエ変換を行う第1のFFT手段と、離散時間フーリエ変換を行う第2のFFT手段と、前記第1のFFT手段の出力データの順序を周波数オフセット補償量に応じて変え、前記周波数オフセット補償量に応じた順序とした前記出力データを並び替えして前記第2のFFT手段に入力する、オフセット補償手段と、を備えた周波数オフセット補償装置。
(付記2)
前記オフセット補償手段が、前記第1のFFT手段の出力データを書き込み、前記第2のFFT手段に入力するための並び替えを行う、並び替え用メモリ手段と、前記並び替え用メモリ手段へ前記出力データを書き込む位置を決定する、書き込み位置決定手段と、前記周波数オフセット補償量に基づいて、前記並び替え用メモリ手段に書き込まれた前記出力データの読み出し順序を決定する、読み出し位置決定手段と、前記周波数オフセット補償量に基づいて、前記読み出し順序で読み出された順序に対応して前記出力データを並び替える、第1の並び替え手段と、前記第1の並び替え手段で並び替えられた前記出力データを並び替えして前記第2のFFT手段へ入力する、第2の並び替え手段と、を備えた、付記1記載の周波数オフセット補償装置。
(付記3)
前記オフセット補償手段が、前記入力信号の順序をカウントするカウンタ手段を備え、前記書き込み位置決定手段が、前記カウンタ手段のカウント値に基づいて、前記並び替え用メモリ手段へ前記出力データを書き込む位置を決定し、前記読み出し位置決定手段が、前記周波数オフセット補償量に基づいて、前記並び替え用メモリ手段に書き込まれた前記出力データの読み出し順序を決定し、前記第1の並び替え手段が、前記カウンタ手段のカウント値と前記周波数オフセット補償量に基づいて、前記読み出し順序で読み出された順序に対応して前記出力データを並び替え、前記第2の並び替え手段が、前記カウンタ手段のカウント値に基づいて、前記第1の並び替え手段で並び替えられた前記出力データを並び替えして前記第2のFFT手段へ入力する、付記2記載の周波数オフセット補償装置。
(付記4)
前記オフセット補償手段が、前記第1のFFT手段の出力データを書き込み、前記第2のFFT手段に入力するための並び替えを行う、並び替え用メモリ手段と、前記周波数オフセット補償量に基づいて、前記並び替え用メモリ手段へ前記出力データを書き込む位置を決定する、書き込み位置決定手段と、前記並び替え用メモリ手段に書き込まれた前記出力データの読み出し順序を決定する、読み出し位置決定手段と、前記周波数オフセット補償量に基づいて、前記読み出し順序で読み出された順序に対応して前記出力データを並び替える、第1の並び替え手段と、前記第1の並び替え手段で並び替えられた前記出力データを、並び替えして前記第2のFFT手段へ入力する、第2の並び替え手段と、を備えた、付記1記載の周波数オフセット補償装置。
(付記5)
前記オフセット補償手段が、前記入力信号の順序をカウントするカウンタ手段を備え、前記書き込み位置決定手段が、前記周波数オフセット補償量に基づいて、前記並び替え用メモリ手段へ前記出力データを書き込む位置を決定し、前記読み出し位置決定手段が、前記カウンタ手段のカウント値に基づいて、前記並び替え用メモリ手段に書き込まれた前記出力データの読み出し順序を決定し、前記第1の並び替え手段が、前記カウンタ手段のカウント値と前記周波数オフセット補償量に基づいて、前記読み出し順序で読み出された順序に対応して前記出力データを並び替え、前記第2の並び替え手段が、前記カウンタ手段のカウント値に基づいて、前記第1の並び替え手段で並び替えられた前記出力データを並び替えして前記第2のFFT手段へ入力する、付記4記載の周波数オフセット補償装置。
(付記6)
前記第1のFFT手段と前記第2のFFT手段の少なくとも一方が、逆離散時間フーリエ変換である、付記1から5の内の1項記載の周波数オフセット補償装置。
(付記7)
前記第2のFFT手段の後に、前記第2のFFT手段の所定の出力値をゼロにする強制ゼロ手段を備えた、付記1から6の内の1項記載の周波数オフセット補償装置。
(付記8)
前記強制ゼロ手段が、前記周波数オフセット補償量に基づいて前記出力値をゼロにする、付記7に記載の周波数オフセット補償装置。
(付記9)
入力信号を第1の離散時間フーリエ変換し、前記第1の離散時間フーリエ変換した出力データの順序を周波数オフセット補償量に応じて変え、前記周波数オフセット補償量に応じた順序とした前記出力データを並び替え、前記並び替えした前記出力データを第2の離散時間フーリエ変換する、周波数オフセット補償方法。
(付記10)
前記第1の離散時間フーリエ変換した前記出力データのメモリへの書き込み位置を決定し、前記書き込み位置に前記出力データを書き込み、前記書き込み位置に書き込まれた前記出力データの読み出し順序を決定し、前記読み出し順序で読み出された前記出力データを第1の並び替えをし、前記第1の並び替えをした前記出力データを第2の並び替えをして前記第2の離散時間フーリエ変換する、付記9に記載の周波数オフセット補償方法。
(付記11)
前記周波数オフセット補償量に基づいて、前記読み出し順序を決定する、または、前記書き込み位置を決定する、付記10に記載の周波数オフセット補償方法。
(付記12)
前記第1の並び替えは、前記周波数オフセット補償量に基づいて前記出力データを並び替える、付記10または11に記載の周波数オフセット補償方法。
(付記13)
前記第1の離散時間フーリエ変換と前記第2の離散時間フーリエ変換の少なくとも一方が、逆フーリエ変換を行う、付記9から12の内の1項記載の周波数オフセット補償方法。
(付記14)
前記第2の離散時間フーリエ変換の後に、前記第2の離散時間フーリエ変換の所定の出力値をゼロにする、付記9から13の内の1項記載の周波数オフセット補償方法。
(付記15)
前記ゼロにするが、前記周波数オフセット補償量に基づいて前記出力値をゼロにする、付記14に記載の周波数オフセット補償方法。
101 第一FFT手段
102 並び替え用メモリ
103 第一並替手段
104 第二並替手段
105 第二FFT手段
106 読み出し位置決定手段
107 書き込み位置決定手段
108 カウンタ
109 周波数オフセット補償量
110、111 オフセット補償手段
200 光周波数混合器
201 ローカル発振レーザー
202 PD(フォトダイオード)
203 ADC
204 周波数オフセット補償器
205 等化器
206 積分器
207 周波数オフセット推定器
208 位相推定器
209 データ復元
210 デジタル信号処理部
300 FFTブロック
301 FFTブロック
302 64ポイントFFT
303 並び替え用メモリ
601 並び替え用メモリ(1)
602 FFT
603 フィルタ演算(複素乗算器)
604 IFFT
605 並び替え用メモリ(2)
701 1R1Wメモリ
702 64個の1R1Wメモリを並べたもの
900 信号(周波数オフセット補償後)
901 信号(周波数オフセット補償前)
1901 読み出し位置決定手段
1902 書き込み位置決定手段
1903 第一並替手段
2101 第三FFT手段
2301 強制ゼロ手段
2501 フィルタ演算(複素乗算器)
2502 IFFT
Claims (10)
- 入力信号の離散時間フーリエ変換を行う第1のFFT手段と、離散時間フーリエ変換を行う第2のFFT手段と、
前記第1のFFT手段の出力データの順序を周波数オフセット補償量に応じて変え、前記周波数オフセット補償量に応じた順序とした前記出力データを並び替えして前記第2のFFT手段に入力する、オフセット補償手段と、
を備えた周波数オフセット補償装置。 - 前記オフセット補償手段が、
前記第1のFFT手段の出力データを書き込み、前記第2のFFT手段に入力するための並び替えを行う、並び替え用メモリ手段と、
前記並び替え用メモリ手段へ前記出力データを書き込む位置を決定する、書き込み位置決定手段と、
前記周波数オフセット補償量に基づいて、前記並び替え用メモリ手段に書き込まれた前記出力データの読み出し順序を決定する、読み出し位置決定手段と、
前記周波数オフセット補償量に基づいて、前記読み出し順序で読み出された順序に対応して前記出力データを並び替える、第1の並び替え手段と、
前記第1の並び替え手段で並び替えられた前記出力データを並び替えして前記第2のFFT手段へ入力する、第2の並び替え手段と、
を備えた、請求項1記載の周波数オフセット補償装置。 - 前記オフセット補償手段が、前記入力信号の順序をカウントするカウンタ手段を備え、
前記書き込み位置決定手段が、前記カウンタ手段のカウント値に基づいて、前記並び替え用メモリ手段へ前記出力データを書き込む位置を決定し、
前記読み出し位置決定手段が、前記周波数オフセット補償量に基づいて、前記並び替え用メモリ手段に書き込まれた前記出力データの読み出し順序を決定し、
前記第1の並び替え手段が、前記カウンタ手段のカウント値と前記周波数オフセット補償量に基づいて、前記読み出し順序で読み出された順序に対応して前記出力データを並び替え、
前記第2の並び替え手段が、前記カウンタ手段のカウント値に基づいて、前記第1の並び替え手段で並び替えられた前記出力データを並び替えして前記第2のFFT手段へ入力する、
請求項2記載の周波数オフセット補償装置。 - 前記オフセット補償手段が、
前記第1のFFT手段の出力データを書き込み、前記第2のFFT手段に入力するための並び替えを行う、並び替え用メモリ手段と、
前記周波数オフセット補償量に基づいて、前記並び替え用メモリ手段へ前記出力データを書き込む位置を決定する、書き込み位置決定手段と、
前記並び替え用メモリ手段に書き込まれた前記出力データの読み出し順序を決定する、読み出し位置決定手段と、
前記周波数オフセット補償量に基づいて、前記読み出し順序で読み出された順序に対応して前記出力データを並び替える、第1の並び替え手段と、
前記第1の並び替え手段で並び替えられた前記出力データを、並び替えして前記第2のFFT手段へ入力する、第2の並び替え手段と、
を備えた、請求項1記載の周波数オフセット補償装置。 - 前記オフセット補償手段が、前記入力信号の順序をカウントするカウンタ手段を備え、
前記書き込み位置決定手段が、前記周波数オフセット補償量に基づいて、前記並び替え用メモリ手段へ前記出力データを書き込む位置を決定し、
前記読み出し位置決定手段が、前記カウンタ手段のカウント値に基づいて、前記並び替え用メモリ手段に書き込まれた前記出力データの読み出し順序を決定し、
前記第1の並び替え手段が、前記カウンタ手段のカウント値と前記周波数オフセット補償量に基づいて、前記読み出し順序で読み出された順序に対応して前記出力データを並び替え、
前記第2の並び替え手段が、前記カウンタ手段のカウント値に基づいて、前記第1の並び替え手段で並び替えられた前記出力データを並び替えして前記第2のFFT手段へ入力する、
請求項4記載の周波数オフセット補償装置。 - 前記第1のFFT手段と前記第2のFFT手段の少なくとも一方が、逆離散時間フーリエ変換である、請求項1から5の内の1項記載の周波数オフセット補償装置。
- 前記第2のFFT手段の後に、前記第2のFFT手段の所定の出力値をゼロにする強制ゼロ手段を備えた、請求項1から6の内の1項記載の周波数オフセット補償装置。
- 入力信号を第1の離散時間フーリエ変換し、
前記第1の離散時間フーリエ変換した出力データの順序を周波数オフセット補償量に応じて変え、
前記周波数オフセット補償量に応じた順序とした前記出力データを並び替え、
前記並び替えした前記出力データを第2の離散時間フーリエ変換する、
周波数オフセット補償方法。 - 前記第1の離散時間フーリエ変換した前記出力データのメモリへの書き込み位置を決定し、
前記書き込み位置に前記出力データを書き込み、
前記書き込み位置に書き込まれた前記出力データの読み出し順序を決定し、
前記読み出し順序で読み出された前記出力データを第1の並び替えをし、
前記第1の並び替えをした前記出力データを第2の並び替えをして前記第2の離散時間フーリエ変換する、
請求項8に記載の周波数オフセット補償方法。 - 前記周波数オフセット補償量に基づいて、前記読み出し順序を決定する、または、前記書き込み位置を決定する、請求項9に記載の周波数オフセット補償方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/651,037 US9621278B2 (en) | 2012-12-20 | 2013-12-11 | Frequency offset compensation apparatus and frequency offset compensation method |
JP2014552918A JP6233318B2 (ja) | 2012-12-20 | 2013-12-11 | 周波数オフセット補償装置および周波数オフセット補償方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2012277818 | 2012-12-20 | ||
JP2012-277818 | 2012-12-20 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2014097583A1 true WO2014097583A1 (ja) | 2014-06-26 |
Family
ID=50977946
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/007289 WO2014097583A1 (ja) | 2012-12-20 | 2013-12-11 | 周波数オフセット補償装置および周波数オフセット補償方法 |
Country Status (3)
Country | Link |
---|---|
US (1) | US9621278B2 (ja) |
JP (1) | JP6233318B2 (ja) |
WO (1) | WO2014097583A1 (ja) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107409000B (zh) * | 2015-04-10 | 2019-05-28 | 华为技术有限公司 | 一种相干光源频偏估计和补偿的相干接收机、方法和系统 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010268210A (ja) * | 2009-05-14 | 2010-11-25 | Nippon Telegr & Teleph Corp <Ntt> | 光受信装置及び光受信方法 |
WO2011007803A1 (ja) * | 2009-07-17 | 2011-01-20 | 日本電信電話株式会社 | 波長分散量算出装置、光信号受信装置、光信号送信装置及び波長分散量算出方法 |
WO2011158932A1 (ja) * | 2010-06-17 | 2011-12-22 | 日本電信電話株式会社 | 周波数オフセット推定装置、受信装置、周波数オフセット推定方法、および受信方法 |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101453269B (zh) | 2007-11-30 | 2012-01-04 | 富士通株式会社 | 频差补偿装置和方法、光相干接收机 |
WO2010145075A1 (en) * | 2009-06-17 | 2010-12-23 | Huawei Technologies Co., Ltd. | Method for carrier frequency recovery and optical intradyne coherent receiver |
JP5651990B2 (ja) * | 2010-04-15 | 2015-01-14 | 富士通株式会社 | デジタルコヒーレント受信器および受信方法 |
-
2013
- 2013-12-11 WO PCT/JP2013/007289 patent/WO2014097583A1/ja active Application Filing
- 2013-12-11 US US14/651,037 patent/US9621278B2/en active Active
- 2013-12-11 JP JP2014552918A patent/JP6233318B2/ja active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010268210A (ja) * | 2009-05-14 | 2010-11-25 | Nippon Telegr & Teleph Corp <Ntt> | 光受信装置及び光受信方法 |
WO2011007803A1 (ja) * | 2009-07-17 | 2011-01-20 | 日本電信電話株式会社 | 波長分散量算出装置、光信号受信装置、光信号送信装置及び波長分散量算出方法 |
WO2011158932A1 (ja) * | 2010-06-17 | 2011-12-22 | 日本電信電話株式会社 | 周波数オフセット推定装置、受信装置、周波数オフセット推定方法、および受信方法 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2014097583A1 (ja) | 2017-01-12 |
US9621278B2 (en) | 2017-04-11 |
JP6233318B2 (ja) | 2017-11-22 |
US20150311983A1 (en) | 2015-10-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106165366B (zh) | 利用符号定时恢复对调制信号进行频率与相位偏移补偿 | |
US8514979B2 (en) | Integrated demodulator, filter and decimator (DFD) for a radio receiver | |
Chen et al. | Non-maximally decimated analysis/synthesis filter banks: Applications in wideband digital filtering | |
EP3512125B1 (en) | Optical receiver, optical reception method and optical communication system | |
CN104969523A (zh) | 用于处理i/q下变频信号及双通道ti-adc的通道失配的方法和设备 | |
CN104780037A (zh) | 一种时钟恢复方法、装置及系统 | |
JP2018093280A (ja) | 波長分散補償装置、波長分散補償方法及び通信装置 | |
CN107395287B (zh) | 一种频偏估计方法和装置 | |
JP6233318B2 (ja) | 周波数オフセット補償装置および周波数オフセット補償方法 | |
CN106972833B (zh) | 一种可任意倍数重采样的数字上变频器 | |
WO2014063539A1 (zh) | 一种多载波基带消峰装置及方法 | |
CN111262634B (zh) | 色散估计方法、装置、接收机及存储介质 | |
JP5876849B2 (ja) | サンプリングレート変換システム、及びサンプリングレート変換方法 | |
Ramon et al. | Efficient parallelization of polyphase arbitrary resampling FIR filters for high-speed applications | |
JP6454725B2 (ja) | 光信号の受信機用の装置、方法およびコンピュータプログラム | |
JP6009953B2 (ja) | 光受信装置 | |
EP2976852B1 (en) | Timing recovery apparatus | |
Zhang et al. | Digital Modulation and FPGA Implementation of MSK Based on SDR | |
US20230129067A1 (en) | Digital Filter Arrangement for Compensating Group Velocity Dispersion in an Optical Transmission System | |
KR20050042753A (ko) | 타이밍추출장치 및 방법 그리고 그 타이밍추출장치를구비한 복조장치 | |
Sousa et al. | Parallel polyphase filtering for pulse shaping on high-speed optical communication systems | |
JP5856580B2 (ja) | 信号処理装置、及び信号処理方法 | |
KR20120129003A (ko) | 2 이상의 샘플링 채널을 사용하여 신호의 복소 샘플링을 수행하고 이들 채널들 간에 시간지연을 계산하기 위한 방법 및 시스템 | |
Huang et al. | Parallel Symbol Timing Recovery Using FPGA for 600 Msps QPSK | |
Sahukar et al. | Area Efficient Fractional Sample Rate Conversion Architecture For Software Defined Radios |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 13865537 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2014552918 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14651037 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 13865537 Country of ref document: EP Kind code of ref document: A1 |