WO2011158932A1 - 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 - Google Patents
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 Download PDFInfo
- Publication number
- WO2011158932A1 WO2011158932A1 PCT/JP2011/063903 JP2011063903W WO2011158932A1 WO 2011158932 A1 WO2011158932 A1 WO 2011158932A1 JP 2011063903 W JP2011063903 W JP 2011063903W WO 2011158932 A1 WO2011158932 A1 WO 2011158932A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- frequency
- frequency offset
- received signal
- offset
- estimation
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/10—Frequency-modulated carrier systems, i.e. using frequency-shift keying
- H04L27/14—Demodulator circuits; Receiver circuits
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/10—Polarisation diversity; Directional diversity
-
- 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
-
- 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
- H04B10/65—Intradyne, i.e. coherent receivers with a free running local oscillator having a frequency close but not phase-locked to the carrier signal
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
- H04B17/318—Received signal strength
- H04B17/327—Received signal code power [RSCP]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2647—Arrangements specific to the receiver only
- H04L27/2655—Synchronisation arrangements
- H04L27/2657—Carrier synchronisation
- H04L27/2659—Coarse or integer frequency offset determination and synchronisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/0014—Carrier regulation
- H04L2027/0024—Carrier regulation at the receiver end
- H04L2027/0026—Correction of carrier offset
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/0014—Carrier regulation
- H04L2027/0024—Carrier regulation at the receiver end
- H04L2027/0026—Correction of carrier offset
- H04L2027/0032—Correction of carrier offset at baseband and passband
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/0014—Carrier regulation
- H04L2027/0044—Control loops for carrier regulation
- H04L2027/0046—Open loops
Definitions
- the present invention relates to a frequency offset estimation device, a reception device, a frequency offset estimation method, and a reception method in a digital coherent optical receiver and a wireless communication receiver.
- This application claims priority based on Japanese Patent Application No. 2010-138402 filed in Japan on June 17, 2010 and Japanese Patent Application No. 2010-251868 filed in Japan on November 10, 2010. The contents thereof are incorporated herein.
- a digital coherent communication system that combines digital signal processing with a synchronous detection method that dramatically improves frequency utilization efficiency has attracted attention.
- the digital coherent communication system Compared with a system constructed by direct detection, the digital coherent communication system not only improves reception sensitivity, but also receives chromatic dispersion and polarization received by optical fiber transmission by receiving the transmission signal as a digital signal. It is known that the waveform distortion of a transmission signal due to mode dispersion can be compensated, and its introduction as a next generation optical communication technology is being studied.
- the signal light received by the coherent receiver is multiplied by the local oscillation light and converted into a baseband signal.
- Laser oscillators that generate carrier waves of signal light and local oscillation light are difficult to stabilize with a phase-locked loop that is generally used in oscillators for wireless communication.
- the output frequency of the laser oscillator of the transmitter and the frequency of the receiver A large frequency offset occurs between the output frequency of the laser oscillator. In an actual optical communication system, the frequency offset reaches ⁇ 5 GHz.
- a coherent communication system since information is carried on the phase of a carrier wave, it is necessary to estimate and compensate for a frequency offset in a receiver.
- a frequency offset occurs due to an error in the oscillation frequency of a reference oscillator used in the transmitter and the receiver, and a Doppler shift accompanying movement of the transmitter and the receiver. Again, it is necessary to estimate and compensate for the frequency offset at the receiver.
- Conventional frequency offset estimation includes a method using known pilot symbols (see Non-Patent Document 1).
- this method has a drawback in that a transmission rate is reduced by adding a known pilot symbol that does not contribute to information transmission to a transmission signal, and a circuit or procedure for detecting a known pilot symbol is required.
- a frequency offset estimation method that does not require a known pilot symbol includes a phase increase algorithm that uses symbol phase change information in one symbol period (see Non-Patent Document 2), and a method that uses a frequency spectrum (Non-Patent Document 2). Document 3) is known.
- FIG. 17 is a block diagram showing a configuration example of a conventional frequency offset estimation apparatus that uses a phase increase algorithm for an M-PSK (M-Phase Shift Keying) modulated signal.
- the frequency offset estimation apparatus shown in FIG. 17 includes a 1-symbol delay unit 101, a complex conjugate unit 102, a multiplication unit 103, an M-th power unit 104, an addition unit 105, and a phase detection unit 106.
- the input signal I + jQ is a complex signal in which the received signal is sampled in advance at a predetermined sampling frequency.
- This input signal is branched into two, and one branched signal passes through the 1-symbol delay unit 101 and the complex conjugate unit 102 and is multiplied by the other branched signal by the multiplier 103 to change the phase between 1 symbol. It becomes a complex signal with information.
- the complex signal is multiplied by M (positive integer) in the M-th power unit 104 to remove a phase change caused by data modulation.
- the signal from which the phase change has been removed is added over N (positive integer) symbols by the adder 105, thereby averaging the phase and removing the instantaneous change.
- the phase is extracted from the signal after the addition by the phase detection unit 106, and the phase that is M times the phase change between 1 symbol by the M-th power calculation of the M-th power unit 104 is made 1 / M times.
- a phase change ⁇ between one symbol caused by the frequency offset is obtained.
- the frequency offset estimated value ⁇ f is calculated by the following equation.
- R S is a symbol rate.
- FIG. 18 is a block diagram showing a configuration example of a conventional frequency offset estimation apparatus using a frequency spectrum.
- the frequency offset estimation apparatus in FIG. 18 includes a multiplier 107, an FFT (fast Fourier transform) unit 108, a frequency error detector 109, and an NCO (numerically-controlled oscillator) 110.
- the input signal I + jQ is a complex signal in which the received signal is sampled in advance at a predetermined sampling frequency.
- This input signal is multiplied by the output signal of the NCO 110 by the multiplier 107, and the frequency is changed.
- the signal with the changed frequency is input to the FFT unit 108 and converted into a frequency spectrum in the frequency domain.
- the frequency error detector 109 measures the frequency spectrum and outputs a frequency error signal. Based on this frequency error signal, the NCO 110 changes the frequency of the output signal in a predetermined step. The above loop calculation is repeated until the frequency error signal becomes substantially zero, and the frequency offset estimation is completed when the frequency error signal converges to almost zero.
- Non-Patent Document 4 and Non-Patent Document 5 disclose frequency offset estimation methods that do not require a known pilot symbol for a signal modulated by QAM. Formulas representing the estimation methods described in these documents are as follows.
- y (p, t) is a received signal and is a function of polarization p and time t.
- N is the number of symbols used for estimation, and R S is the symbol rate.
- FIG. 19 to FIG. 22 are explanatory diagrams showing the operation of the frequency offset estimation method according to Non-Patent Document 4 and Non-Patent Document 5 described above.
- FIG. 19 shows a constellation in the case where a signal modulated by 64QAM has a frequency offset of 0 or an integral multiple of R S / 4 and a phase offset remains. The period of the signal point is the reciprocal 1 / R S of the symbol rate. If the frequency offset is R S / 4, the signal point on the constellation is just ⁇ / 2 rotated from the position where the frequency offset is 0. To position. That is, the same constellation arrangement is shown when the frequency offset is 0 and when the frequency offset is an integral multiple of R S / 4.
- the signal modulated by QAM has a phase symmetry of ⁇ / 2
- the phase of one signal point is phase ⁇
- the distance from the origin is equal to the distance between this signal point and the origin.
- the phase ⁇ of these four black dots k1, k2, k3, k4 is expressed by the following mathematical formula 4.
- the four black points k1, k2, k3, k4 converge to the same point on the complex plane when raised to the fourth power.
- every four points having the same distance from the origin and a phase difference of an integral multiple of ⁇ / 2 converge to the same point on the complex plane.
- FIG. 20 is a signal point arrangement diagram when the signal points of the constellation in FIG. 19 are raised to the fourth power.
- the 64 points in FIG. 19 converge to 16 points in FIG.
- These signal points are asymmetric with respect to the real axis (horizontal axis) and the imaginary axis (vertical axis), and the added value or average value of the signal points takes a non-zero value.
- FIG. 21 shows a constellation in the case where there is a frequency offset other than an integral multiple of R S / 4 with respect to a signal modulated by 64QAM. Signal points having the same distance from the origin are arranged on the circumference.
- FIG. 22 is a signal point arrangement diagram when the signal points of the constellation in FIG. 21 are raised to the fourth power. Since these signal points are symmetric with respect to the real axis (horizontal axis) and the imaginary axis (vertical axis), the addition value or average value of the signal points is zero.
- the above-described expression 2 has the following terms.
- this term is an evaluation function ⁇ c (f)
- this is obtained by taking the time average by applying the inverse rotation operator exp ( ⁇ j2 ⁇ ft) of the frequency f after the received signal y (p, t) is raised to the fourth power. It is an operation.
- the action of the inverse rotation operator exp ( ⁇ j2 ⁇ ft) converts the frequency of the original signal by ⁇ f.
- the received signal y (p, t) when the frequency offset is fo the evaluation function phi c evaluation function at the frequency f in the (f) is f ⁇ 4fo + kR S (k is an integer) phi c (f ) Is 0.
- the evaluation function ⁇ c (f) takes a non-zero value.
- the range of the frequency offset that can be estimated is limited by the phase uncertainty.
- the phase range that can be extracted by the calculation “arg (•)” for extracting the phase is [ ⁇ to ⁇ ].
- the phase range that can be detected by the phase detector 106 is [ ⁇ / M to ⁇ / M]. Therefore, the frequency range that can be estimated by the frequency offset estimation apparatus shown in FIG. 17 is limited to [ ⁇ R S / 2M to R S / 2M] by Equation 1.
- the frequency of the output signal of the NCO 110 is changed in predetermined steps, and the number of data when performing FFT of several hundred to several thousand at each frequency is changed.
- the number of data when performing FFT of several hundred to several thousand at each frequency is changed.
- it takes time until the estimation process converges because it is necessary to fetch input signals having the same number of samples into the frequency offset estimation apparatus.
- a frequency offset estimation device that uses a frequency spectrum, when the cut-off frequency of the band-pass filter in the transmission path or the low-pass filter in the receiver is small and the frequency offset is large, one side of the frequency spectrum is trimmed and asymmetric. As a result, there is a problem that the estimation accuracy of the frequency offset deteriorates.
- Non-Patent Document 4 and Non-Patent Document 5 for a signal modulated by QAM represented by Formula 2 and Formula 3, a frequency band in which the estimable frequency range is the fourth power [-R S / 2 to R S / 2] and the original frequency band before the fourth power is limited to [-R S / 8 to R S / 8]. That is, a frequency offset exceeding this range is a problem because it is erroneously detected. Furthermore, when there is phase noise, the signal point on the constellation moves along the circumference, and the state is the same as in FIG. These problems will be specifically described with reference to FIGS. FIG. 23 to FIG. 25 are explanatory diagrams showing examples of simulation results for explaining the conventional problems.
- the modulation is DP-64QAM (polarization multiplexing 64 quadrature amplitude modulation) with a symbol rate of 28 GBaud
- OSNR optical signal to noise ratio
- N 1028.
- 5 GHz is given as a frequency offset
- the frequency f and the evaluation function ⁇ c (f) expressed by the following formula 6 are plotted.
- FIG. 23 shows a simulation result when the frequency f is swept from ⁇ 64 GHz to 64 GHz in a wide band. In this simulation, no phase noise is added.
- FIG. 24 and FIG. 25 show the simulation results assuming that the laser line width is 10 MHz. In this simulation, phase noise is added. 24 and 25, 1028 symbol sequences used for estimation are different from each other. In FIG. 24, the largest value is shown at 20 GHz, which is four times the frequency offset, but many local peaks are also seen at other frequencies. In FIG. 25, there is a peak at 20 GHz, but a local peak different from 20 GHz shows the largest value in the whole. In other words, when f that maximizes the evaluation function ⁇ c (f) is obtained from FIG. 25, an incorrect value is estimated as the frequency offset. As described above, the frequency offset estimation methods described in Non-Patent Document 4 and Non-Patent Document 5 have a problem that an erroneous estimation result is often output depending on the state of phase noise or thermal noise.
- the present invention has been made in view of such circumstances, and its purpose is to estimate the frequency offset of the received signal when estimating the frequency offset, which is the difference between the carrier frequency of the received signal and the frequency of the output signal of the local oscillator.
- An object of the present invention is to provide a frequency offset estimation device, a reception device, a frequency offset estimation method, and a reception method that can appropriately estimate an offset.
- the present invention has been made to solve the above-described problems, and the present invention provides a frequency offset estimation device that estimates a frequency offset that is a difference between a carrier frequency of a received signal and a frequency of an output signal of a local oscillator.
- the frequency of the received signal sampled in advance at a predetermined sampling frequency is frequency-converted, and a frequency spectrum having N frequency components ordered from 1 to N (N is an arbitrary natural number) in order of frequency magnitude is output.
- N is an arbitrary natural number
- a negative frequency component having a frequency component number from 1 to N / 2 and a positive frequency component having a frequency component number from N / 2 + 1 to N in the frequency spectrum
- a frequency band limiting unit that limits each frequency component, and the positive of the frequency spectrum that is frequency band limited.
- Each power is calculated by adding the square of the wave number component and the negative frequency component to calculate each power, and the absolute value of the power difference calculated from the power of the positive frequency component and the power of the negative frequency component is calculated in advance. All frequency components of the frequency spectrum are circulated and moved on the frequency axis until the threshold value is less than or equal to the set threshold value, and the frequency offset is estimated based on the amount of movement that is moved until the threshold value is less than or equal to the threshold value.
- a frequency offset estimation control unit is used to calculate the square of the wave number component and the negative frequency component to calculate each power, and the absolute value of the power difference calculated from the power of the positive frequency component and the power of the negative frequency component is calculated in advance. All frequency components of the frequency spectrum are circulated and moved on the frequency axis until the threshold value is less than or equal to the set threshold value, and the frequency offset is estimated based on the amount of movement that is moved until the threshold value is less than or equal to the threshold value.
- a frequency offset estimation control unit is used to calculate the square of
- the frequency offset estimation control unit when moving all the frequency components of the frequency spectrum in a circulating manner on the frequency axis, the power of the positive frequency component is the negative power. If the frequency component number after the movement is less than 1, move all frequency components of the frequency spectrum by a predetermined amount in the negative direction. When the power of the positive frequency component is less than or equal to the power of the negative frequency component, all frequency components of the frequency spectrum are moved in a positive direction by a predetermined amount, and the frequency component after the movement If the number exceeds N, N may be subtracted from the frequency component number.
- the present invention provides a frequency offset estimation apparatus according to the present invention, and a first frequency offset that compensates for the frequency offset of the received signal based on a value of the frequency offset of the received signal estimated by the frequency offset estimation apparatus.
- a phase increase frequency offset estimation unit for estimating the frequency offset based on a phase increase algorithm for the received signal compensated by the compensation unit, and the first frequency offset compensation unit; and the phase increase frequency offset estimation
- a second frequency offset compensation unit that compensates the frequency offset based on the value of the frequency offset of the reception signal estimated by the unit.
- the present invention relates to a frequency offset estimation method used in a frequency offset estimation apparatus for estimating a frequency offset which is a difference between a carrier frequency of a received signal and a frequency of an output signal of a local oscillator, and is pre-sampled at a predetermined sampling frequency.
- a negative frequency component having a frequency component number from 1 to N / 2 and a positive frequency component having a frequency component number from N / 2 + 1 to N are respectively represented in frequency bands.
- Frequency band limiting procedure for limiting and the positive frequency of the frequency spectrum limited by the frequency band Minute and the negative frequency component are squarely added to calculate each power, and the absolute value of the power difference calculated from the positive frequency component power and the negative frequency component power is preset. All frequency components of the frequency spectrum are circulated and moved on the frequency axis until the threshold value is less than or equal to the threshold value, and the frequency offset is estimated based on the amount of movement that has been moved to the threshold value or less.
- the power of the positive frequency component is the negative power. If the frequency component number after the movement is less than 1, move all frequency components of the frequency spectrum by a predetermined amount in the negative direction. When the power of the positive frequency component is less than or equal to the power of the negative frequency component, all frequency components of the frequency spectrum are moved in a positive direction by a predetermined amount, and the frequency component after the movement When the number exceeds N, the frequency spectrum may be circulated and moved by subtracting N from the frequency component number.
- the present invention compensates the frequency offset of the received signal based on the procedure by the frequency offset estimation method of the present invention and the value of the frequency offset of the received signal estimated by the procedure by the frequency offset estimation method.
- 1 frequency offset compensation procedure a phase increase frequency offset estimation procedure for estimating the frequency offset based on a phase increase algorithm for the received signal compensated by the first frequency offset compensation procedure, and the phase
- a second frequency offset compensation procedure for compensating the frequency offset based on the value of the frequency offset of the received signal estimated by an increased frequency offset estimation procedure.
- the present invention relates to a frequency offset estimation device for estimating a difference between a carrier frequency of a received signal and a frequency of an output signal of a local oscillator, the frequency of the received signal comprising two polarized waves pre-sampled at a predetermined sampling frequency.
- a frequency offset rough estimation unit that estimates a frequency offset from a spectrum
- a sweep frequency range control unit that determines a sweep frequency range based on a rough estimation value of the frequency offset rough estimation unit
- a sweep frequency range control unit A frequency offset fine estimation unit that estimates a frequency offset of the received signal in the sweep frequency range, and the frequency offset fine estimation unit has a signal point on the constellation having no frequency offset of the received signal.
- a first computing unit that performs frequency conversion for subtracting the sweep frequency from the frequency of the received signal after the two polarized waves in the received signal are each raised to the W power; and the first computing unit A second computing unit that computes an absolute value or a power of an absolute value after averaging or adding N (N: positive integer) symbols to the computation result of each polarization in
- N positive integer
- U U: positive
- (Integer) a third arithmetic unit that adds or averages frames, and a sweep frequency at which the arithmetic result of the third arithmetic unit becomes a maximum value, and the frequency offset is calculated by multiplying the sweep frequency by 1 / W. 4th operation part which estimates And a frequency offset estimation apparatus.
- the present invention relates to a frequency offset estimation method for estimating a difference between a carrier frequency of a received signal and a frequency of an output signal of a local oscillator, the frequency of the received signal comprising two polarizations sampled in advance at a predetermined sampling frequency.
- a frequency offset rough estimation procedure for estimating a frequency offset from a spectrum, a sweep frequency range control procedure for determining a sweep frequency range based on the rough estimation value estimated by the frequency offset rough estimation procedure, and the sweep frequency range control procedure A frequency offset fine estimation procedure for estimating a frequency offset of the received signal in the determined range of the sweep frequency, and the frequency offset fine estimation procedure includes signal points on a constellation having no frequency offset of the received signal.
- the present invention relates to a frequency offset estimation device for estimating a difference between a carrier frequency of a received signal and a frequency of an output signal of a local oscillator, the frequency of the received signal comprising two polarized waves pre-sampled at a predetermined sampling frequency.
- a rough frequency offset estimation unit for estimating a frequency offset from a spectrum a fine frequency offset estimation unit having a periodic frequency offset estimation characteristic with respect to the received signal or a signal whose dispersion of the received signal is compensated, and the frequency offset
- a frequency uncertainty removal control unit that removes the frequency uncertainty of the frequency offset estimated by the frequency offset fine estimation unit and estimates the frequency offset Is a frequency offset estimation device.
- the frequency offset precise estimation unit may estimate the frequency offset based on a phase increase algorithm for the received signal or a signal with compensated dispersion of the received signal. Good.
- the frequency offset fine estimation unit defines the rotational symmetry of the signal point on the constellation having no frequency offset of the received signal as 2 ⁇ / W (W: positive integer).
- W positive integer
- the two polarizations in the received signal are each converted to a frequency spectrum after being raised to the power W, and an absolute value or a power of an absolute value is calculated for the conversion result, and the frequency spectrum of these two polarizations Or the frequency spectrum of the polarized wave having the larger peak value is selected, and addition or averaging of U (U: positive integer) frames is performed on the frequency spectrum of a frame composed of N (N: positive integer) symbols. And the frequency at which the calculation result becomes the maximum value may be detected.
- the frequency uncertainty removal control unit removes the frequency uncertainty and estimates the frequency offset, to the frequency offset estimated by the frequency offset precision estimation unit. Based on the frequency offset candidate including the frequency uncertainty and calculating a frequency midpoint between adjacent frequency offset candidates on a frequency axis as a boundary between adjacent frequency offset candidates. And selecting a region including the value estimated by the frequency offset rough estimation unit from regions based on the boundary on the frequency axis, and selecting the frequency offset candidates included in the selected region May be selected as the estimated value of the frequency offset.
- the frequency offset rough estimation unit performs frequency conversion on the received signal, and outputs N frequency components that are ordered from 1 to N (N is an arbitrary natural number) in order of frequency magnitude.
- a frequency conversion unit that outputs a frequency spectrum, a negative frequency component having a frequency component number from 1 to N / 2, and a frequency component number from N / 2 + 1 to N of the frequency spectrum.
- a frequency band limiting unit that limits each positive frequency component that is a frequency component, and a square addition of each of the positive frequency component and the negative frequency component of the frequency spectrum that is frequency band limited. Then, each power is calculated, and the absolute value of the power difference calculated from the power of the positive frequency component and the power of the negative frequency component is calculated in advance.
- All frequency components of the frequency spectrum are circulated and moved on the frequency axis until the threshold value is less than or equal to the set threshold value, and the frequency offset is estimated based on the amount of movement that is moved until the threshold value is less than or equal to the threshold value.
- a frequency offset estimation control unit that performs the operation.
- the present invention provides a frequency offset estimator according to the present invention, and a frequency offset compensator for compensating the frequency offset of the received signal based on the value of the frequency offset of the received signal estimated by the frequency offset estimator.
- a reception apparatus comprising: a phase compensation unit that compensates a phase of the reception signal compensated by the frequency offset compensation unit; and a determination unit that performs determination of a symbol of the reception signal compensated for the phase. is there.
- the present invention relates to a frequency offset estimation method for estimating a difference between a carrier frequency of a received signal and a frequency of an output signal of a local oscillator, the frequency of the received signal comprising two polarizations sampled in advance at a predetermined sampling frequency.
- a frequency offset estimation method comprising: .
- the frequency offset fine estimation procedure may be configured to estimate the frequency offset based on a phase increase algorithm for the received signal or a signal whose dispersion of the received signal is compensated. Good.
- the rotational symmetry of the signal point on the constellation having no frequency offset of the received signal is defined as 2 ⁇ / W (W: positive integer).
- W positive integer
- the two polarizations in the received signal are each converted to a frequency spectrum after being raised to the power W, and an absolute value or a power of an absolute value is calculated for the conversion result, and the frequency spectrum of these two polarizations Or the frequency spectrum of the polarized wave having the larger peak value is selected, and addition or averaging of U (U: positive integer) frames is performed on the frequency spectrum of a frame composed of N (N: positive integer) symbols. And the frequency at which the calculation result becomes the maximum value may be detected.
- the frequency offset estimated by the frequency offset precision estimation procedure is added to the frequency offset. Based on the frequency offset candidate including the frequency uncertainty and calculating a frequency midpoint between adjacent frequency offset candidates on a frequency axis as a boundary between adjacent frequency offset candidates. And selecting a region including the value estimated by the frequency offset rough estimation procedure from regions based on the boundary on the frequency axis, and selecting the frequency offset candidates included in the selected region May be selected as the estimated value of the frequency offset.
- the frequency offset rough estimation procedure performs frequency conversion on the received signal, and includes N frequency components ordered from 1 to N (N is an arbitrary natural number) in order of frequency magnitude.
- the present invention provides a frequency offset compensation for compensating the frequency offset of the received signal based on a procedure by the frequency offset estimating method of the present invention and a value of the frequency offset of the received signal estimated by the frequency offset estimating method.
- a reception step comprising: a phase compensation procedure for compensating a phase of the reception signal compensated by the frequency offset compensation procedure; and a determination procedure for judging a symbol of the reception signal compensated for the phase. Is the method.
- the frequency offset is estimated based on the movement amount until all frequency components are moved, so that the frequency offset can be estimated in a wide band as compared with the frequency offset estimation device using the phase increase algorithm.
- it is a method that does not require a known pilot symbol and does not require an NCO. Therefore, the frequency offset can be estimated with high speed and high accuracy compared to a frequency offset estimation device that uses another wide frequency spectrum. can do.
- the frequency offset of a received signal when estimating the frequency offset that is the difference between the carrier frequency of the received signal and the frequency of the output signal of the local oscillator, based on the roughly estimated frequency offset value, the frequency offset of a received signal can be estimated appropriately.
- FIG. 1 is a schematic diagram illustrating a configuration example of a frequency offset estimation apparatus 15 that estimates a frequency offset that is a difference between a carrier frequency of a received signal and a frequency of an output signal of a local oscillator on the receiving side according to the first embodiment of the present invention. It is a block diagram.
- the frequency offset estimation device 15 includes an FFT unit 1, an SPDT (Single Pole, Double Throw) switch 2, a bandpass filter 3, and a frequency offset estimation control unit 12.
- the frequency offset estimation control unit 12 includes a first square addition unit 4, a second square addition unit 5, a subtraction unit 6, a first determination unit 7, a second determination unit 8, and a frequency spectrum circular movement.
- an input signal I + jQ is a complex signal in which a received signal is sampled in advance at a predetermined sampling frequency.
- the FFT unit 1 converts this input signal into a frequency spectrum in the frequency domain. For example, if the number of data when the FFT unit 1 performs FFT is N (N is an arbitrary natural number), a frequency spectrum having N number frequency components ordered from 1 to N in order of frequency magnitude is output. .
- frequency components having frequency component numbers from 1 to N / 2 in this frequency spectrum are defined as negative frequency components
- frequency components having frequency component numbers from N / 2 + 1 to N in the frequency spectrum are defined as positive frequency components.
- the SPDT switch 2 switches a signal input to the band pass filter 3 under the control of the first determination unit 7 described later.
- the SPDT switch 2 inputs the output signal of the FFT unit 1 to the band-pass filter 3 when the determination of the first determination unit 7 is NO, and the output signal of the frequency spectrum cyclic shift unit 9 when the determination is YES. Switch to.
- the initial setting is NO for the first determination unit 7, the common terminal of the SPDT switch 2 is connected to the upper contact, and the SPDT switch 2 passes the output signal of the FFT unit 1 to the bandpass filter 3. input.
- the band pass filter 3 limits the frequency band of the signal input through the SPDT switch 2.
- the band-pass filter 3 has a center frequency of 1 to N frequency components with respect to the positive frequency component signal and the negative frequency component signal of the received signal converted into the frequency spectrum by the FFT unit 1.
- the frequency band is limited so that positive and negative are symmetric, a signal having a frequency component with a positive pass band is sent to the first square adder 4, and a signal having a frequency component having a negative pass band is a second signal.
- the first square addition unit 4 and the second square addition unit 5 calculate the power value P P of the positive frequency component and the power value P M of the negative frequency component of the pass band of the bandpass filter 3, respectively. .
- the subtraction unit 6 calculates a power difference between the two power values P P and P M.
- FIG. 2 and FIG. 3 show the power value P P of the positive frequency component and the power value P M of the negative frequency component of the pass band of the band pass filter 3 for the received signal converted into this frequency spectrum.
- It is explanatory drawing which shows the outline
- the band pass filter 3 multiplies the signal of each frequency component by the transfer function of the band pass filter 3 for the signal converted into the frequency spectrum in the frequency domain by the FFT unit 1. For example, when a rectangular wave filter is used as the bandpass filter 3, the frequency component of the pass band is multiplied by 1, and the frequency component of the stop band is multiplied by 0.
- the first square adder 4 and the second square adder 5 square the frequency components of the pass band and take the sum thereof to obtain the positive and negative frequency components of the pass band of the band pass filter 3.
- a power value P P and a power value P M that are the total power are calculated.
- the power value P P of the positive frequency component is equal to the power value P M of the negative frequency component due to the symmetry of the frequency spectrum.
- the frequency spectrum is biased in either the positive direction or the negative direction, and the power value P P of the positive frequency component and the power value of the negative frequency component. power difference between P M occurs.
- the first determination unit 7 in FIG. 1 compares the absolute value of the power difference between the power value P P and the power value P M with a predetermined threshold value T H to make a determination of magnitude. To do.
- the first determination unit 7 starts the process of the second determination unit 8 when the absolute value of the power difference between the power value P P and the power value P M is larger than the threshold value T H and the determination is YES. Further, the contact point of the SPDT switch 2 is switched, and the input signal to the band pass filter 3 is changed from the output signal from the FFT unit 1 to the output signal from the frequency spectrum circulation moving unit 9.
- the second determination unit 8 compares the power value P P and the power value P M and determines the magnitude. Based on the result of the determination, that is, the result of determining whether the frequency spectrum is biased in the positive direction or the negative direction, the second determination unit 8 moves in the frequency spectrum circulation moving unit 9 and the counter 10. Select control and notify.
- the frequency spectrum cyclic movement unit 9 moves all frequency components on the frequency axis in a direction in which the bias in the positive direction or the negative direction of the frequency component determined by the second determination unit 8 is reduced.
- the counter 10 counts and holds the number of times of moving in the positive direction and moving in the negative direction.
- this movement is a cyclic movement in which the frequency component number is 1 to N even after the movement. That is, when the frequency component number after movement is less than 1, N is added to the frequency component number, and when the frequency component number after movement exceeds N, N is subtracted from the frequency component number.
- the frequency spectrum circulation moving unit 9 circulates and moves the frequency components one by one in the negative direction. Increase the value d by one.
- the frequency spectrum circulation moving unit 9 circulates and moves the frequency components one by one in the positive direction. The value d of the counter 10 is decreased by 1.
- the contact point of the SPDT switch 2 is changed so that the output signal from the frequency spectrum circulation moving unit 9 is input to the bandpass filter 3.
- the spectrum is input again to the bandpass filter 3, and signal processing is executed in the order of the first square addition unit 4, the second square addition unit 5, and the subtraction unit 6, and the power value P P and the power value P
- the power difference from M is calculated and input to the first determination unit 7.
- This loop signal processing is repeatedly executed until the first determination unit 7 determines NO. That is, the signal processing of the loop is repeated until the absolute value of the power difference between the power values P P and the power value P M is determined to be equal to or lower than a predetermined threshold value T H.
- the signal processing of this loop converges, and the frequency spectrum movement distance detection unit 11 detects the value d of the counter 10 at the time of convergence.
- This value is the sum of the movement distances on the frequency axis of the frequency spectrum, and the frequency offset is estimated based on the frequency corresponding to the movement amount that is the sum of the movement distances.
- the frequency offset estimation is completed, and the first determination unit 7 switches the contact point of the SPDT switch 2 so that the output signal of the FFT unit 1 is input to the bandpass filter 3. Data is input to this loop signal processing and frequency offset estimation is started.
- FIG. 4 is an explanatory diagram showing an example of a result of simulating the operation of the first embodiment
- FIG. 5 is a result of simulating the operation of a frequency offset estimation apparatus using a conventional phase increase algorithm for comparison. It is explanatory drawing which showed an example.
- the horizontal axis represents the set value of the frequency offset
- the vertical axis represents the frequency offset estimation result.
- the modulation method is 112 Gbit / s polarization multiplexed QPSK (Quadrature Phase Shift Keying), and the symbol rate R S is 28 Gbaud.
- the frequency offset estimation apparatus 15 estimates a frequency offset that is a difference between the carrier frequency of the received signal and the frequency of the output signal of the local oscillator.
- the absolute value of the power difference between the positive frequency component power value P P and the negative frequency component power value P M of the frequency band limited in the frequency spectrum of the received signal is equal to or less than a preset threshold value. All frequency components are circulated until the frequency reaches, and the frequency offset is estimated based on the amount of movement.
- the frequency offset estimation device 15 of the first embodiment estimates the frequency offset in a wide band compared to the frequency offset range that can be estimated by the frequency offset estimation device using the conventional phase increase algorithm shown in FIG. can do.
- the method according to the first embodiment is a method that does not require a known pilot symbol, and further, an NCO that is used in the conventional frequency offset estimation apparatus shown in FIG. 18 is unnecessary. Therefore, according to the present embodiment, it is not necessary to capture an input signal having the number of samples equal to the number of data when performing FFT of several hundred to several thousand at each frequency of stepwise frequency change. Therefore, according to the present embodiment, the frequency offset can be estimated with high speed and high accuracy.
- the frequency offset estimating apparatus 15 does not need to be limited to 1 for moving the frequency component in a circulating manner. Therefore, by increasing the size to move in circulation when the absolute value of the power difference between the power values P P and the power value P M is large and the absolute value of the power difference between the power values P P and the power value P M it is possible to reduce the size to move circulating when approaching the threshold T H. As a result, the frequency offset can be estimated at a higher speed.
- the value of the previously determined threshold T H of the first determination unit 7 may be determined by the power of the input signal. For example, assuming that the frequency spectrum of the output signal of the FFT unit 1 is a uniform rectangle, and if the sum of all the power of the frequency spectrum is the power value P TOTAL and the frequency component is cyclically moved by 1 in the negative direction, the power value P P is smaller by P TOTAL / N than before the movement, and the power value P M is larger by P TOTAL / N. Therefore, the difference between the power value P P and the power value P M changes by 2P TOTAL / N. At this time, if the threshold value T H of about 2P TOTAL / N, it is possible to estimate the frequency offset at approximately 1 frequency component of accuracy. If high estimation processing time is required even at the expense of estimation accuracy, the threshold TH may be increased.
- the number N of data when the FFT unit 1 performs FFT can be determined by required estimation accuracy and estimation processing time. In general, the smaller N is, the lower the estimation accuracy is, but the estimation processing time is shortened.
- the frequency offset estimation value output from the frequency spectrum movement distance detection unit 11 may be arithmetically averaged after a plurality of frequency offsets are accumulated, or may be averaged using a forgetting coefficient to follow temporal variations. You may let them.
- the FFT unit 1 that converts an input signal into a frequency spectrum is not limited to fast Fourier transform as a frequency analysis method, and may use short-time Fourier transform, discrete Fourier transform, wavelet transform, or other frequency analysis methods.
- the bandpass filter 3 is not limited to the bandpass filter as a filter for limiting the frequency band, and a lowpass filter, a highpass filter, or another filter for limiting the frequency band is used based on the characteristics of the received signal. Also good.
- the second embodiment is a frequency offset estimation device that uses a conventional phase increase algorithm after roughly estimating a frequency offset by the frequency offset estimation device 15 described in the first embodiment and compensating the frequency offset with the estimated value.
- This is a receiving apparatus having a configuration in which two types of frequency estimation apparatuses for estimating and further compensating for a frequency offset are combined.
- FIG. 6 is a schematic block diagram illustrating a configuration example of a receiving device according to the second embodiment of the present invention.
- the receiving device includes a frequency offset estimation device 15, a first frequency offset compensation unit 28, a polarization separation unit 19, a phase increase frequency offset estimation device 20, a second frequency offset compensation unit 29, a first phase compensation unit 24, A second phase compensation unit 25, a first determination unit 26, and a second determination unit 27 are provided.
- the phase increase frequency offset estimation apparatus 20 has the same configuration as the conventional frequency offset estimation apparatus using the phase increase algorithm shown in the configuration example of FIG.
- the frequency offset estimation apparatus 15 shown in FIG. 6 corresponds to the configuration described in the first embodiment shown in FIG.
- the first frequency offset compensator 28 includes a first NCO 16, a first multiplier 17, and a second multiplier 18.
- the second frequency offset compensation unit 29 includes a second NCO 21, a third multiplication unit 22, and a fourth multiplication unit 23.
- the input signal I + jQ and the input signal I ′ + jQ ′ are complex signals obtained by pre-sampling X-polarized and Y-polarized received signals at a predetermined sampling frequency.
- the frequency offset estimation device 15 roughly estimates the frequency offset in a wide band with respect to the input signal I + jQ and the input signal I ′ + jQ ′.
- the frequency offset estimation device 15 is designed to satisfy the following conditions.
- the condition is that when the frequency offset estimating device 15 roughly estimates the frequency offset, the phase-increasing frequency offset estimating device 20 estimates the frequency offset with respect to the input signal compensated for the roughly estimated frequency offset. In such a case, the condition falls within a frequency range that can be estimated without frequency uncertainty.
- the frequency offset estimation value roughly estimated by the frequency offset estimation device 15 is input to the first frequency offset compensation unit 28.
- the first frequency offset compensation unit 28 compensates for the frequency offset of the input signal. That is, the oscillation frequency of the first NCO 16 included in the first frequency offset compensation unit 28 is adjusted based on the estimated value of the frequency offset that is roughly estimated. Then, the input signal I + jQ and the input signal I ′ + jQ ′ and the adjusted signal having the oscillation frequency of the first NCO 16 are converted into the first multiplier 17 included in the first frequency offset compensator 28. And the second multiplier 18 compensates for the frequency offset of the input signal I + jQ and the input signal I ′ + jQ ′.
- the polarization separation unit 19 performs polarization separation and removal of residual dispersion on the signal whose frequency offset of the input signal I + jQ and the input signal I ′ + jQ ′ is compensated by the first frequency offset compensation unit 28. Execute.
- the phase increase frequency offset estimation apparatus 20 receives the signal output from the polarization separation unit 19 and executes frequency offset estimation.
- the frequency offset estimation value estimated by the phase increase frequency offset estimation device 20 is input to the second frequency offset compensation unit 29.
- the second frequency offset compensator 29 compensates for the frequency offset of the input signal. That is, the oscillation frequency of the second NCO 21 included in the second frequency offset compensator 29 is adjusted based on the frequency offset estimation value estimated by the phase increase frequency offset estimation device 20. Then, the input signal to the second frequency offset compensation unit 29 and the adjusted signal having the oscillation frequency of the second NCO 21 are converted into a third multiplication unit 22 included in the second frequency offset compensation unit 29 and Multiplication is performed by the fourth multiplication unit 23, and the frequency offset of the input signal to the second frequency offset compensation unit 29 is compensated.
- the first phase compensation unit 24 and the second phase compensation unit 25 compensate the phase with respect to the signal whose frequency offset is compensated in the second frequency offset compensation unit 29, and the first determination unit 26 and the second determination unit 27.
- the first determination unit 26 and the second determination unit 27 determine a symbol and generate a demodulated signal.
- the receiving apparatus including the frequency offset estimating apparatus according to the second embodiment first compensates the input signal based on the value roughly estimated by the frequency offset estimating apparatus 15 and then increases the conventional phase increase. Input to the frequency offset estimation apparatus 20. Further, by executing both polarization separation and residual dispersion removal, it is possible to estimate the frequency offset using the phase increase algorithm for an input signal having both a small frequency offset and dispersion value. As a result, the receiving apparatus including the frequency offset estimating apparatus according to the second embodiment can estimate, compensate, and demodulate the received signal with a wide band and high accuracy.
- the frequency offset estimation device using the frequency spectrum
- the frequency offset estimation apparatus of the second embodiment after the frequency offset is roughly estimated and compensated by the frequency offset estimation apparatus 15, the frequency offset is estimated by the conventional phase increase frequency offset estimation apparatus 20. To do. Therefore, in the frequency offset estimation apparatus according to the second embodiment, even when the cutoff frequency of the band-pass filter in the transmission path or the low-pass filter in the receiver is small and the frequency offset is large, the frequency offset Can be prevented from degrading.
- the signal input to the phase increase frequency offset estimation device 20 is replaced with the output signal from the polarization separation unit 19, instead of the first frequency offset compensation unit 28 before being input to the polarization separation unit 19. It may be a signal with compensated frequency offset. However, the frequency offset can be estimated with higher accuracy when the signal from which polarization separation and residual dispersion have been removed by the polarization separation unit 19 is input to the phase increase frequency offset estimation device 20.
- the third embodiment of the present invention will be described below with reference to the drawings.
- the input signal is compensated based on the value roughly estimated by the frequency offset estimating apparatus 15.
- the receiving apparatus of the third embodiment does not compensate the input signal based on the value roughly estimated by the frequency offset estimating apparatus 15 and detects the phase when using the conventional phase increase algorithm.
- This is a receiving apparatus having a configuration that estimates the frequency offset by removing the uncertainty.
- it can replace with the frequency offset estimation apparatus of 1st Embodiment, and can utilize the invention of a nonpatent literature 3.
- FIG. 7 is a schematic block diagram illustrating a configuration example of a receiving device according to the third embodiment of the present invention.
- the receiving device includes a frequency offset estimation device 15, a frequency uncertainty removal control unit 30, a polarization separation unit 19, a phase increase frequency offset estimation device 20, a frequency offset compensation unit 29, a first phase compensation unit 24, a second A phase compensation unit 25, a first determination unit 26, and a second determination unit 27 are provided.
- the frequency offset compensation unit 29 includes an NCO 21, a first multiplication unit 22, and a second multiplication unit 23.
- the frequency offset estimation device 15 has the configuration described in the first embodiment, and a description thereof is omitted. In FIG. 7, the same reference numerals are given to the components corresponding to the respective parts in FIG. 6, and the description thereof is omitted.
- an input signal I + jQ and an input signal I ′ + jQ ′ are complex signals obtained by sampling X-polarized and Y-polarized received signals in advance at a predetermined sampling frequency.
- the frequency offset estimation device 15 roughly estimates the frequency offset in a wide band with respect to the input signal I + jQ and the input signal I ′ + jQ ′, and uses the estimated value as the frequency uncertainty.
- the phase increase frequency offset estimation apparatus 20 estimates a frequency offset for the signal output from the polarization separation unit 19 and outputs the estimated value to the frequency uncertainty removal control unit 30.
- the frequency uncertainty removal control unit 30 estimates the frequency offset based on the estimated values of the frequency offsets input from the frequency offset estimation device 15 and the phase increase frequency offset estimation device 20, and the frequency offset
- the signal is output to the NCO 21 of the compensation unit 29 and the oscillation frequency of the NCO 21 is adjusted.
- the frequency offset compensation unit 29 compensates the frequency offset of the input signal I + jQ and the input signal I ′ + jQ ′
- the first phase compensation unit 24, the second phase compensation unit 25, and the first determination unit 26 and the second determination unit 27 demodulate.
- the frequency uncertainty removal control unit 30 calculates a plurality of frequency offset candidate frequencies based on the estimated frequency offset value input from the phase-increasing frequency offset estimation device 20. Next, the frequency uncertainty removal control unit 30 calculates, as a boundary, a candidate that is adjacent on the frequency axis and a candidate midpoint among the plurality of calculated frequency offset candidates. Further, the frequency uncertainty removal control unit 30 detects that the estimated value of the frequency offset input from the frequency offset estimation device 15 is in a region between adjacent boundaries of the frequency offset candidates. Then, the frequency uncertainty removal control unit 30 selects a frequency offset candidate in the detected area as an estimated value of the frequency offset.
- the frequency uncertainty removal control unit 30 removes the uncertainty of the frequency offset estimation result of the phase increase frequency offset estimation device 20.
- the uncertainty of the frequency offset estimation result is due to the uncertainty of the phase detected by the phase increase algorithm in the phase increase frequency offset estimation apparatus 20, and has periodicity.
- a frequency offset estimate value of the phase increment frequency offset estimation apparatus 20 and the frequency f Mth the frequency f CND candidates for potentially estimates a frequency f Mth frequency offset (k) is given by the following Equation 7 It is done.
- a frequency offset estimate value of the frequency offset estimator 15 and the frequency f CO by the frequency f CO to design the frequency offset estimator 15 so as to satisfy the relationship of Equation 8, uncertainty Can be removed.
- the frequency f CND (k) is arranged from the frequency f Mth to the R S / M period on the frequency axis, and the midpoint of the frequency of the adjacent frequency f CND (k) is set as the adjacent frequency offset candidate. It shows that it is a boundary.
- the frequency uncertainty removal control unit 30 estimates the actual frequency offset as f CND (1).
- the receiving apparatus including the frequency offset estimating apparatus according to the third embodiment first roughly estimates the frequency offset by the frequency offset estimating apparatus 15.
- the frequency uncertainty removal control unit 30 removes the phase uncertainty detected by the phase increase algorithm based on the estimated value and estimates the frequency offset.
- the receiving apparatus including the frequency offset estimating apparatus according to the third embodiment can estimate, compensate, and demodulate the received signal with a wide band and high accuracy.
- an NCO and a multiplier for compensating for the frequency offset with the estimated value of the frequency offset estimation device 15 are not required, and the circuit scale can be reduced.
- the input signal input to the phase increase frequency offset estimation device 20 is an output signal from the polarization separation unit 19 in FIG. 7 of the configuration example of the third embodiment, but is input to the polarization separation unit 19. It may be a signal before being processed.
- FIG. 9 is a block diagram showing a configuration example of a frequency offset estimation apparatus 150 according to the fourth embodiment of the present invention.
- the frequency offset estimation apparatus includes a frequency offset fine estimation unit 41, a frequency offset rough estimation unit 57, and a sweep frequency range control unit 58.
- the frequency offset precision estimation unit 41 includes a first fourth-power circuit 42, a second fourth-power circuit 43, a first multiplication circuit 44, a second multiplication circuit 45, an NCO (numerically-controlled oscillator) 46, a first 1 N symbol addition circuit 47, second N symbol addition circuit 48, first 1 / N division circuit 49, second 1 / N division circuit 50, first absolute value squaring circuit 51, second An absolute value squaring circuit 52, a polarization adding circuit 53, a U frame adding circuit 54, a maximum value sweep frequency detecting circuit 55, and a 1/4 dividing circuit 56 are provided.
- the frequency offset rough estimation unit 57 estimates the frequency offset from the frequency spectrum of the received signal composed of two polarized waves sampled in advance at a predetermined sampling frequency.
- the frequency offset rough estimation unit 57 can use the frequency offset estimation device 15 that uses the frequency spectrum described in the first embodiment. Moreover, it can replace with the frequency offset estimation apparatus 15 described in 1st Embodiment, and the frequency offset estimation apparatus described in the nonpatent literature 3 can be used similarly.
- the sweep frequency range control unit 58 determines the sweep frequency range based on the rough estimated value of the frequency offset rough estimation unit 57.
- the precise frequency offset estimation unit 41 estimates the frequency offset of the received signal in the sweep frequency range determined by the sweep frequency range control unit 58.
- the frequency offset precision estimation unit 41 includes a first calculation unit 41a, a second calculation unit 41b, a third calculation unit 41c, and a fourth calculation unit 41d.
- the first computing unit 41a performs the frequency conversion by subtracting the sweep frequency from the frequency of the received signal on the received signal after each of the two polarized waves in the received signal is raised to the fourth power.
- the second calculation unit 41b performs averaging (calculation of an average value) of N (N: positive integer) symbols on the calculation result of each polarization in the first calculation unit 41a,
- the power of the absolute value is calculated (the power of the absolute value is calculated with respect to the average value).
- the third calculation unit 41c performs addition processing of a U (U: positive integer) frame for a frame composed of N symbols.
- the 4th calculating part 41d detects the sweep frequency from which the calculation result of the 3rd calculating part 41c becomes the maximum value, multiplies this sweep frequency, and outputs it, and estimates a frequency offset.
- the first computing unit 41a includes a first fourth power circuit 42, a second fourth power circuit 43, a first multiplication circuit 44, a second multiplication circuit 45, and an NCO 46.
- the second arithmetic unit 41b includes a first N symbol addition circuit 47, a second N symbol addition circuit 48, a first 1 / N division circuit 49, a second 1 / N division circuit 50, and a first absolute value.
- a value square circuit 51 and a second absolute value square circuit 52 are provided.
- the third arithmetic unit 41 c includes a polarization addition circuit 53 and a U frame addition circuit 54.
- the fourth calculation unit 41 d includes a maximum value sweep frequency detection circuit 55 and a 1 ⁇ 4 division circuit 56.
- Equations 9 and 10 represent the equations representing the operation of the fourth embodiment.
- N is the number of symbols used for estimating one frame
- u is the frame number of a frame consisting of N symbol sequences
- U is the total number of frames used for estimation
- y (u, p, t) is the received signal.
- R S is a symbol rate
- f coarse is a rough estimated frequency of the frequency offset rough estimation unit 57.
- input signals I + jQ and I ′ + jQ ′ are respectively X-polarized wave and Y-polarized wave obtained by pre-sampling the received signal y (u, p, t) of Equation 9 at a predetermined sampling frequency. It is a complex signal.
- the frequency offset precise estimation unit 41 squares this input signal with the first fourth power circuit 42 and the second fourth power circuit 43, and outputs the output signal exp ( ⁇ j2 ⁇ ft) of the NCO 46, the first multiplication circuit 44 and the first power circuit 44. Multiplication is performed in the multiplication circuit 45 of 2. This multiplication corresponds to an operation for performing frequency conversion by subtracting the frequency f from the frequency of the received signal raised to the fourth power.
- the sweep frequency of the NCO 46 is controlled by the sweep frequency range controller 58. Specifically, the sweep frequency range control unit 58 determines the frequency range of the sweep frequency of the NCO 46 so as to satisfy Equation 10 from the coarse estimated frequency f coarse estimated by the frequency offset coarse estimation unit 57 and the symbol rate R S. The frequency f is changed in a predetermined step from the lower limit frequency to the upper limit frequency of Equation 10.
- the frequency offset fine estimator 41 adds the operation results for the N symbols in the first N symbol adder circuit 47 and the second N symbol adder circuit 48 at each frequency f to obtain the first 1 / N
- the division circuit 49 and the second 1 / N division circuit 50 divide by N to obtain N average values, and the first absolute value squaring circuit 51 and the second absolute value squaring circuit 52 obtain an absolute value of 2
- the power is obtained, and the calculation result for the two polarized waves is added in the polarization adding circuit 53.
- the frequency offset fine estimation unit 41 regards the N symbol sequences as the first frame and the next N symbol sequences as the second frame, and adds the calculation result of the U frame in the U frame addition circuit 54. By the addition process, the frequency offset fine estimation unit 41 removes the noise component.
- the evaluation function ⁇ n (f) represented by the following formula 11 is obtained.
- the frequency offset estimator 150 calculates the frequency f at which the evaluation function ⁇ n (f) becomes the maximum value by the maximum value sweep frequency detection circuit 55 and multiplies it by 1 ⁇ 4 by the 1 ⁇ 4 division circuit 56 to obtain the frequency An offset can be estimated.
- the frequency offset rough estimation unit 57 and the sweep frequency range control unit 58 control the sweep frequency range of the NCO 46, so that the frequency offset estimation range can be widened.
- the frequency offset estimation device 150 detects only the peak of 20 GHz and does not detect the peaks on both sides, the frequency offset estimation apparatus 150 detects 20 GHz as the maximum value sweep frequency without uncertainty, and 5/4 that is 1/4 of that is detected. It can be output as a frequency offset estimate.
- the estimable frequency range is [f coarse ⁇ R S / 8 to f coarse + R S / 8], and the coarse estimated frequency f coarse is variable. It is possible to estimate over a wide band.
- the frequency offset rough estimation unit 57 does not require high-precision characteristics.
- the estimation range is [-14 GHz to 14 GHz]
- -8 GHz is detected as the maximum sweep frequency
- -2 GHz which is 1/4, is output as the frequency offset estimation value. .
- FIG. 10 assumes that the same frequency offset as 5 GHz, laser line width 10 MHz, and OSNR 10 dB as in FIGS. 24 and 25, 1028 for the number of symbols N used to estimate one frame, and 80 for the total number of frames U used for estimation.
- the results obtained by performing evaluation function ⁇ n (f) of Formula 11 are shown. Comparing FIG. 10 with FIG. 24 and FIG. 25, the peak at 20 GHz, which is four times the frequency offset, is emphasized, and the local peak that appeared temporarily for each frame is kept relatively low. Therefore, the frequency offset estimation apparatus 150 can calculate a correct estimation result without being affected by a local peak temporarily appearing for each frame due to phase noise or thermal noise.
- the frequency offset estimation apparatus 150 uses the frequency offset coarse estimation unit 57 to roughly estimate the frequency offset based on the result of the rough estimation by the frequency offset coarse estimation unit 57.
- the sweep frequency range control unit 58 determines the frequency range of the sweep frequency of the NCO 46 so as to eliminate the characteristics, and estimates the frequency offset.
- the frequency offset estimation apparatus 150 removes the noise component of the received signal in the calculation process in which the frequency offset precision estimation unit 41 estimates the frequency offset. Thereby, the frequency offset estimation apparatus 150 of the fourth embodiment can estimate the frequency offset with high accuracy with respect to the received signal in a wide band.
- FIG. 11 is a block diagram illustrating a configuration example of the frequency offset estimation apparatus 150 according to the fifth embodiment.
- the same reference numerals are given to the components corresponding to the respective parts in FIG. 9, and the description thereof is omitted.
- the first calculation unit 41a and the fourth calculation unit 41d are the first calculation unit 41a and the fourth calculation unit according to the fourth embodiment.
- the configuration is the same as 41d.
- the second calculation unit 41b and the third calculation unit 41c are the second calculation unit 41b and the third calculation unit according to the fourth embodiment.
- the configuration is different from 41c.
- the second calculation unit 41b includes a first N symbol addition circuit 47, a second N symbol addition circuit 48, a first absolute value circuit 59, and a first absolute value circuit 59. Two absolute value circuits 60 are provided.
- the third arithmetic unit 41c includes a first U frame addition circuit 61, a second U frame addition circuit 62, a first 1 / U division circuit 63, a second 1 / U division circuit 64, and a bias.
- a wave addition circuit 53 is provided.
- the frequency offset fine estimation unit 41 adds the N symbols to the calculation result of each polarization in the first calculation unit 41a in the second calculation unit 41b, and then calculates the absolute value. (Absolute value is calculated for the added value). Further, after the frequency offset precision estimation unit 41 performs the U frame averaging process on the frame of N symbols on the calculation result obtained by the second calculation unit 41b in the third calculation unit 41c. The calculation results of the two polarizations are added. Then, in the fourth calculation unit 41d, the frequency offset precision estimation unit 41 detects the sweep frequency at which the calculation result of the third calculation unit 41c is the maximum value, and outputs the frequency obtained by multiplying this sweep frequency by 1 ⁇ 4. To estimate the frequency offset.
- the frequency offset fine estimation unit 41 in the present embodiment has the first 1 / N division circuit 49 and the second 1 / N division circuit 50 deleted. Yes. Since the first 1 / N division circuit 49 and the second 1 / N division circuit 50 are deleted in this way, each frequency component of the evaluation function ⁇ n (f) becomes N times, but between the frequencies. The relative magnitude relationship does not change. Therefore, the fine frequency offset estimation unit 41 of the present embodiment estimates the frequency offset similarly to the fine frequency offset estimation unit 41 of the fourth embodiment.
- averaging processing is performed, and when processing is performed by the N symbol addition circuit without the 1 / N division circuit, addition processing is performed.
- the first absolute value squaring circuit 51 and the second absolute value squaring circuit 52 in the frequency offset fine estimation unit 41 of the fourth embodiment are the same as those in the frequency offset fine estimation unit 41 of the present embodiment.
- the absolute value circuit 59 and the second absolute value circuit 60 are replaced with each other.
- the fine frequency offset estimation unit 41 of the present embodiment estimates the frequency offset similarly to the fine frequency offset estimation unit 41 of the fourth embodiment.
- the fine frequency offset estimation unit 41 of the present embodiment has a first 1 / U division circuit 63 and a second 1 / U division. A circuit 64 is added. As a result, each frequency component of the evaluation function ⁇ n (f) is 1 / U times, but the relative magnitude relationship between the frequencies does not change. Therefore, the fine frequency offset estimation unit 41 of the present embodiment estimates the frequency offset similarly to the fine frequency offset estimation unit 41 of the fourth embodiment.
- averaging processing is performed, and when processing is performed by the U frame addition circuit by omitting the 1 / U division circuit, addition processing is performed.
- the order from the polarization addition circuit 53 to the U frame addition circuit 54 in the frequency offset fine estimation unit 41 of the fourth embodiment is the same as the first U frame addition circuit in the frequency offset fine estimation unit 41 of the present embodiment.
- 61 (or the second U frame addition circuit 62) is switched in order from the polarization addition circuit 53 via the 1 / U division circuit.
- the fine frequency offset estimation unit 41 of the present embodiment estimates the frequency offset similarly to the fine frequency offset estimation unit 41 of the fourth embodiment. If the total number of frames U used for estimation is large, it is also possible to select only the evaluation function of the polarization having the larger peak value without performing polarization addition.
- the frequency offset estimation apparatus 150 of the fifth embodiment is the same as the configuration of the frequency offset estimation unit 41 of the frequency offset estimation apparatus 150 of the fourth embodiment also in the configuration of the frequency offset estimation unit 41 described above. In addition, it is possible to estimate the frequency offset with high accuracy with respect to the received signal.
- the fourth and fifth embodiments are not limited to signals modulated by QAM, and can be applied to PSK and PSK modulation schemes in which the amplitude of PSK is multistage, or PAM modulation schemes.
- the rotational symmetry of the signal point on the constellation having no frequency offset is 2 ⁇ / W
- the first fourth power circuit 42 and the second fourth power circuit 43 are replaced with a W power circuit and a 1/4 division circuit.
- FIG. 12 is a block diagram illustrating a configuration example of the receiving device 200 according to the sixth embodiment of the present invention.
- the receiving apparatus 200 includes a frequency offset fine estimation unit 41, a frequency offset rough estimation unit 57, a sweep frequency range control unit 58, a polarization separation unit 66, a frequency offset compensation unit 67, a first phase compensation unit 68, a second phase compensation unit 68, and a second phase compensation unit 68.
- a phase compensation unit 69, a first determination unit 70, and a second determination unit 71 are provided.
- the fine frequency offset estimation unit 41, the rough frequency offset estimation unit 57, and the sweep frequency range control unit 58 shown in FIG. 12 correspond to the same reference numerals shown in FIG. 9 and FIG.
- the frequency offset compensation unit 67 compensates the frequency offset of the reception signal based on the value of the frequency offset of the reception signal estimated by the frequency offset estimation device 150.
- the first phase compensation unit 68 and the second phase compensation unit 69 compensate the phase of the reception signal compensated by the frequency offset compensation unit 67.
- the first determination unit 70 and the second determination unit 71 determine the symbol of the received signal whose phase has been compensated.
- the receiving apparatus 200 of the sixth embodiment first sets the sweep frequency of the frequency offset fine estimation unit 41 to the sweep frequency so as to satisfy Equation 10 based on the value roughly estimated by the frequency offset coarse estimation unit 57. This is determined by the range control unit 58.
- the polarization separation unit 66 performs polarization separation and removal of residual dispersion on the input signal I + jQ and the input signal I ′ + jQ ′.
- the frequency offset precision estimation unit 41 receives the signal output from the polarization separation unit 66 and performs frequency offset estimation.
- the frequency offset estimation value estimated by the frequency offset precision estimation unit 41 is input to the frequency offset compensation unit 67, and compensates for the frequency offset of the input signal I + jQ and the input signal I ′ + jQ ′.
- the first phase compensation unit 68 and the second phase compensation unit 69 compensate the phase of the signal whose frequency offset is compensated by the frequency offset compensation unit 67, and the first determination unit 70 and the second determination unit 70, respectively. 2 to the determination unit 71.
- the first determination unit 70 and the second determination unit 71 determine a symbol and generate a demodulated signal.
- the frequency offset estimation unit 41 is operated with respect to an input signal having both a small frequency offset and a dispersion value by executing the polarization offset and the residual dispersion removal together with the frequency offset estimation. be able to.
- the receiving apparatus 200 including the frequency offset estimating apparatus 150 of the sixth embodiment can estimate and compensate for and demodulate the received signal in a wide band and with high accuracy.
- the signal input to the frequency offset fine estimation unit 41 may be the signal before being input to the polarization separation unit 66 instead of the output signal from the polarization separation unit 66.
- the frequency offset can be estimated with higher accuracy when the signal from which polarization separation and residual dispersion have been removed by the polarization separation unit 66 is input to the frequency offset fine estimation unit 41.
- FIG. 13 is a block diagram illustrating a configuration example of a frequency offset estimation apparatus 150 according to the seventh embodiment of the present invention.
- the frequency offset estimation apparatus 150 includes a frequency offset rough estimation unit 57, a frequency offset fine estimation unit 41, and a frequency uncertainty removal control unit 97.
- the frequency offset fine estimation unit 41 includes a first W power circuit 88, a second W power circuit 89, a first FFT unit 90, a second FFT unit 91, a first absolute value squaring circuit 92, a second The absolute value squaring circuit 93, the polarization spectrum adding circuit 94, the U frame spectrum adding circuit 95, and the maximum value frequency detecting circuit 96 are provided.
- W is a number (a positive integer) determined by the modulation method, and the rotational symmetry of the signal point on the constellation having no frequency offset is 2 ⁇ / W.
- W 4
- the frequency offset rough estimation unit 57 can use the frequency offset estimation device 15 that uses the frequency spectrum described in the first embodiment. Moreover, it can replace with the frequency offset estimation apparatus 15 described in 1st Embodiment, and the invention described in the nonpatent literature 3 can be used similarly.
- the present embodiment is a configuration that pays attention to the fact that an evaluation function that is a time average is also a discrete Fourier transform.
- the frequency offset fine estimation unit 41 sets the two polarizations in the received signal to W Convert to frequency spectrum after riding. Then, the frequency offset precise estimation unit 41 performs an absolute value or a power of the absolute value on the conversion result, adds the frequency spectra of these two polarizations, and N (N: positive integer) symbols.
- a U (U: positive integer) frame addition or averaging is performed on the frequency spectrum of each frame consisting of and the frequency at which this calculation result is the maximum value is detected.
- the frequency offset precision estimation unit 41 uses the first FFT unit 90 and the first W power circuit 88 and the second W power circuit 89 to raise the input signals I + jQ and I ′ + jQ ′ to the W power.
- the second FFT unit 91 collectively converts N symbol sequences into a frequency spectrum.
- the frequency offset precise estimation unit 41 obtains the square of the absolute value of the frequency spectrum data by the first absolute value squaring circuit 92 and the second absolute value squaring circuit 93, and polarization polarization
- the spectrum addition circuit 94 adds the frequency spectra of the two polarized waves.
- the precise frequency offset estimation unit 41 regards the N symbol sequences as the first frame and the next N symbol sequences as the second frame, and adds the frequency spectrum of the U frame in the U frame spectrum addition circuit 95. . By the addition process, the frequency offset fine estimation unit 41 removes the noise component.
- the maximum value frequency detection circuit 96 calculates the frequency f taking the maximum value.
- the frequency f at which ⁇ n (f) becomes maximum is detected after the evaluation function ⁇ n (f) is obtained at each frequency by changing the sweep frequency of the NCO 46 in predetermined steps. did.
- a frequency spectrum is obtained, and a frequency f at which the frequency spectrum is maximum in the frequency domain is detected.
- the frequency range of the sweep frequency of the NCO 46 is limited by the sweep frequency range control unit 58 in advance in order to estimate the frequency offset without frequency uncertainty.
- the frequency spectrum used in the seventh embodiment includes frequency uncertainty. This frequency uncertainty is removed by the frequency uncertainty removal control unit 97.
- the frequency uncertainty removal control unit 97 removes the frequency uncertainty of the frequency offset estimated by the frequency offset fine estimation unit 41 based on the value of the frequency offset estimated by the frequency offset rough estimation unit 57, Estimate the offset.
- the frequency uncertainty removal control unit 97 includes the frequency uncertainty based on the frequency offset estimated by the frequency offset refinement estimation unit 41 when the frequency uncertainty is removed and the frequency offset is estimated. A frequency that is a candidate for a frequency offset is calculated. Further, the frequency uncertainty removal control unit 97 sets the midpoint of the frequency that is a candidate for the frequency offset adjacent on the frequency axis as the boundary of the adjacent frequency offset candidate, and is an area based on the boundary on the frequency axis. The region including the value estimated by the frequency offset rough estimation unit 57 is selected from among the regions. The frequency uncertainty removal control unit 97 selects a frequency offset candidate included in the selected region as an estimated value of the frequency offset.
- FIG. 14 is an explanatory diagram illustrating an outline of an operation in which the frequency offset estimation apparatus 150 according to the seventh embodiment estimates a frequency offset.
- the discrete Fourier transform is related to y W (u, p, t). That is, the frequency spectrum is related to y W (u, p, t).
- the frequency range for obtaining the maximum value of the frequency spectrum is limited to [-R S / 2 to R S / 2], which is the first Nyquist zone that is not affected by the aliasing, by Equation 13.
- the frequency f CND candidate frequency offset that may estimate the frequency offset is f MAX (k) is expressed by Equation 14
- f MAX is plotted on the vertical axis and f CND (k) is plotted on the horizontal axis.
- the coarse estimated value of the frequency offset coarse estimator 97 is set as f CO, and the midpoint between adjacent f CND (k) is set as a determination boundary. At this time, the region including fCO among the regions sandwiched between the two determination boundaries gives an actual frequency offset.
- the frequency offset estimation apparatus 150 uses the frequency offset estimated by the frequency offset fine estimation unit 41 based on the frequency offset value estimated by the frequency offset rough estimation unit 57. Remove uncertainty and estimate frequency offset. Thereby, the frequency offset estimation apparatus 150 of the seventh embodiment can estimate the frequency offset with high accuracy with respect to the received signal in a wide band. Further, compared with the fourth embodiment and the fifth embodiment, the sweep frequency range control unit 58 and the NCO 46 are replaced with FFT units 90 and 91 and converted into a frequency spectrum for estimation. It can be shortened.
- the FFT unit 90 and the FFT unit 91 that convert to a frequency spectrum are not limited to the fast Fourier transform as a frequency analysis method, and a short-time Fourier transform, discrete Fourier transform, wavelet transform, or other frequency analysis method may be used. good. Further, even if the squares of the first absolute value squaring circuit 92 and the second absolute value squaring circuit 93 are changed to the first power or other powers, the relative magnitude relationship of each frequency component of the frequency spectrum does not change. Does not affect the operation. It is also possible to select the frequency spectrum of the polarization having the larger peak value without performing the polarization addition in the polarization spectrum adding circuit 94.
- FIG. 15 is an explanatory diagram illustrating an example of an experimental result of the seventh embodiment.
- This experimental result shows the frequency spectrum of the output signal of the U frame spectrum addition circuit 95 when a signal modulated with 64QAM having a symbol rate R S of 10 GHz is received.
- the estimated value f CO of the frequency offset rough estimation unit 57 was 3.7 GHz.
- f MAX in Expression 12 is ⁇ 4 GHz.
- W 4
- FIG. 16 is a block diagram illustrating a configuration example of a receiving device 200 according to the eighth embodiment of the present invention.
- the polarization separation unit 66 performs polarization separation and removal of residual dispersion on the input signal I + jQ and the input signal I ′ + jQ ′.
- the frequency offset precision estimation unit 41 receives the signal output from the polarization separation unit 66 and performs frequency offset estimation.
- the frequency uncertainty removal control unit 97 receives the estimated value of the frequency offset rough estimation unit 57 and the estimated value of the frequency offset fine estimation unit 41, and determines the frequency offset estimation value according to Equations 12 to 16.
- the frequency offset compensation unit 67 receives the estimated frequency offset value output from the frequency uncertainty removal control unit 97 and compensates for the frequency offset of the input signal.
- the first phase compensation unit 68 and the second phase compensation unit 69 compensate the phase of the signal whose frequency offset is compensated by the frequency offset compensation unit 67, and the first determination unit 70 and the second determination unit 70, respectively. 2 to the determination unit 71.
- the first determination unit 70 and the second determination unit 71 determine a symbol and generate a demodulated signal.
- the receiving apparatus 200 according to the eighth embodiment performs the estimation of the frequency offset by the frequency offset estimating apparatus 150 and the combined execution of the polarization separation and the removal of the residual dispersion so that the frequency offset and the dispersion value can be obtained.
- the frequency offset precision estimation unit 41 can be operated for both small input signals. As a result, the receiving apparatus 200 according to the eighth embodiment can estimate, compensate, and demodulate the frequency offset of the received signal with a wide bandwidth and high accuracy.
- the signal input to the frequency offset fine estimation unit 41 may be the signal before being input to the polarization separation unit 66 instead of the output signal from the polarization separation unit 66.
- the frequency offset can be estimated with higher accuracy when the signal from which polarization separation and residual dispersion have been removed by the polarization separation unit 66 is input to the frequency offset fine estimation unit 41.
- the “computer system” includes a homepage providing environment (or display environment) if a WWW (world wide web) system is used.
- the “computer-readable recording medium” means a portable medium such as a flexible disk, a magneto-optical disk, a ROM (read only memory), a CD (compact disc) -ROM, or a hard disk built in the computer system. Refers to the device. Furthermore, the “computer-readable recording medium” dynamically holds a program for a short time like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line.
- a volatile memory in a computer system serving as a server or a client in that case, and a program that holds a program for a certain period of time are also included.
- the program may be a program for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system.
- a configuration for realizing precise estimation of the frequency offset is a phase increase frequency offset estimation apparatus (unit) 20 (FIG. 7) or a frequency. It is not limited to the configuration of the fine offset estimator 41 (FIGS. 13 and 16). Instead, a periodic frequency offset estimation characteristic as shown in a sawtooth waveform in FIGS. 5 and 14 is used. You may use the conventional general structure which has.
- the present invention can be used for, for example, a digital coherent optical receiver or a wireless communication receiver.
- the frequency offset of the received signal can be estimated with a wide band, high speed, and high accuracy.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Quality & Reliability (AREA)
- Circuits Of Receivers In General (AREA)
- Optical Communication System (AREA)
- Superheterodyne Receivers (AREA)
- Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
Abstract
Description
本願は、2010年6月17日に日本へ出願された特願2010-138402号、および、2010年11月10日に日本へ出願された特願2010-251868号に基づき優先権を主張し、それらの内容をここに援用する。
また無線通信においては、送信機と受信機に用いている基準発振器の発振周波数の誤差や送信機と受信機の移動に伴うドップラーシフトにより周波数オフセットが生じる。この場合も受信機において周波数オフセットを推定し、補償する必要がある。
一方、既知のパイロットシンボルを必要としない周波数オフセット推定方法には、1シンボル周期におけるシンボルの位相変化情報を利用する位相増加アルゴリズムや(非特許文献2参照)、周波数スペクトルを利用する方法(非特許文献3参照)が知られている。
図19から図22は、上述の非特許文献4や非特許文献5に係る周波数オフセット推定方法の動作を示す説明図である。
図19に示すものは、64QAMで変調された信号に対して、0またはRS/4の整数倍の周波数オフセットを持っており、また位相オフセットが残留している場合のコンスタレーションである。信号点の周期はシンボルレートの逆数1/RSであり、周波数オフセットがRS/4であればコンスタレーション上の信号点は周波数オフセットが0の時の位置から丁度π/2回転したところに位置にする。すなわち周波数オフセットが0の時とRS/4の整数倍の周波数オフセットの時とでは同じコンスタレーション配置を示す。QAMで変調された信号はπ/2の位相対称性を持っているため、ある1つの信号点の位相を位相θとすると、原点からの距離がこの信号点と原点との距離に等しい点であって、且つこの信号点に対してπ/2の整数倍の位相差を持つ点が他に3点ある。図19においては、これらの点を黒丸k1からk4で示している。これら4点の黒丸k1,k2,k3,k4の位相βは、次の数式4で表される。
すなわち、上記のように受信信号の周波数スペクトルの中の周波数帯域制限された正の周波数成分の電力値と負の周波数成分の電力値との電力差の絶対値が予め設定されたしきい値以下になるまで全周波数成分を移動し、その移動量に基づいて周波数オフセットを推定するため、位相増加アルゴリズムを用いる周波数オフセット推定装置と比較して、広帯域に周波数オフセットを推定することができる。また、既知のパイロットシンボルを必要としない方法であって、さらにNCOも不要であるため、広帯域な他の周波数スペクトルを利用する周波数オフセット推定装置に比較して、高速かつ高精度に周波数オフセットを推定することができる。
以下、図面を参照して、本発明の第1実施形態について説明する。図1は、本発明の第1実施形態による、受信信号の搬送波周波数と受信側の局部発振器の出力信号の周波数との差である周波数オフセットを推定する周波数オフセット推定装置15の構成例を示す概略ブロック図である。
周波数オフセット推定装置15は、FFT部1、SPDT(Single Pole, Double Throw)スイッチ2、バンドパスフィルタ3、および周波数オフセット推定制御部12を備えている。さらに、周波数オフセット推定制御部12は、第1の2乗加算部4、第2の2乗加算部5、減算部6、第1の判定部7、第2の判定部8、周波数スペクトル循環移動部9、カウンタ10、および周波数スペクトル移動距離検出部11を備えている。
FFT部1は、この入力信号を周波数領域の周波数スペクトルに変換する。例えば、FFT部1がFFTする際のデータ数をN(Nは任意の自然数)とすると、周波数の大きさの順に1からNまで順序付けしたN個の番号の周波数成分を持つ周波数スペクトルを出力する。ここで、この周波数スペクトルの1からN/2までの周波数成分番号の周波数成分を負の周波数成分、周波数スペクトルのN/2+1からNまでの周波数成分番号の周波数成分を正の周波数成分としてこの後の信号処理を説明する。
初期設定は第1の判定部7はNOの判定で、SPDTスイッチ2の共通端子は上側の接点に接続されており、SPDTスイッチ2はFFT部1の出力信号を通過させてバンドパスフィルタ3に入力する。
第1の2乗加算部4および第2の2乗加算部5は、それぞれバンドパスフィルタ3の通過帯域の正の周波数成分の電力値PPおよび負の周波数成分の電力値PMを算出する。減算部6は、これら2つの電力値PPと電力値PMとの間の電力差を算出する。
この移動は、FFT部1で変換された周波数スペクトルの周波数成分番号が1からNとすると、移動後も周波数成分番号が1からNになるような循環した移動とする。すなわち移動後の周波数成分番号が1未満の場合は周波数成分番号にNを加算し、移動後の周波数成分番号がNを越える場合は周波数成分番号からNを減算する。
以上で周波数オフセット推定は終了し、第1の判定部7は、SPDTスイッチ2の接点をFFT部1の出力信号がバンドパスフィルタ3に入力されるように切り替え、FFT部1から新たに周波数スペクトルデータをこのループ信号処理に入力し、周波数オフセットの推定は開始される。
図4、図5共に、横軸は周波数オフセットの設定値、縦軸は周波数オフセット推定結果である。変調方式は112Gbit/s偏波多重QPSK(Quadrature Phase Shift Keying)であり、シンボルレートRSは28Gbaudである。
また、周波数スペクトル移動距離検出部11より出力される周波数オフセット推定値は、複数個を蓄積した後に相加平均しても良いし、忘却係数を用いた平均化を行って時間的な変動に追随させても良い。
また、バンドパスフィルタ3は、周波数帯域を制限するフィルタとして、バンドパスフィルタに限らず、受信する信号の特性に基づいて、ローパスフィルタ、ハイパスフィルタ、あるいはその他の周波数帯域を制限するフィルタを用いても良い。
以下、図面を参照して、本発明の第2実施形態について説明する。本第2実施形態は、第1実施形態に記載した周波数オフセット推定装置15により周波数オフセットを粗く推定し、その推定値で周波数オフセットを補償した後に、従来の位相増加アルゴリズムを用いる周波数オフセット推定装置で周波数オフセットを推定し、さらに補償するという2種類の周波数推定装置を組み合わせた構成を持つ受信装置である。
受信装置は、周波数オフセット推定装置15、第1の周波数オフセット補償部28、偏波分離部19、位相増加周波数オフセット推定装置20、第2の周波数オフセット補償部29、第1の位相補償部24、第2の位相補償部25、第1の判定部26、および第2の判定部27を備えている。ここで、位相増加周波数オフセット推定装置20は、図17に構成例を示した位相増加アルゴリズムを用いる従来の周波数オフセット推定装置と同様の構成を有している。図6に示す周波数オフセット推定装置15は、図1に示す第1実施形態に記載の構成に対応しており、その説明を省略する。
第2の周波数オフセット補償部29は、第2のNCO21、第3の乗算部22、および第4の乗算部23を備えている。
入力信号I+jQおよび入力信号I’+jQ’はそれぞれX偏波とY偏波の受信信号が所定のサンプリング周波数で予めサンプリングされた複素信号である。
つまり、粗く推定された周波数オフセットの推定値に基づいて、第1の周波数オフセット補償部28が有する第1のNCO16の発振周波数は調整される。そして、入力信号I+jQおよび入力信号I’+jQ’と、調整された第1のNCO16の発振周波数を持った信号とが、第1の周波数オフセット補償部28が有する第1の乗算部17および第2の乗算部18において乗算され、入力信号I+jQおよび入力信号I’+jQ’の周波数オフセットは補償される。
つまり、位相増加周波数オフセット推定装置20で推定した周波数オフセット推定値に基づいて、第2の周波数オフセット補償部29が有する第2のNCO21の発振周波数は調整される。そして、第2の周波数オフセット補償部29への入力信号と、調整された第2のNCO21の発振周波数を持った信号とが、第2の周波数オフセット補償部29が有する第3の乗算部22および第4の乗算部23において乗算され、第2の周波数オフセット補償部29への入力信号の周波数オフセットは補償される。
以下、図面を参照して、本発明の第3実施形態について説明する。
上記に説明したように、第2実施形態の受信装置においては、周波数オフセット推定装置15が粗く推定した値に基づいて入力信号を補償している。これに対して、本第3実施形態の受信装置は、周波数オフセット推定装置15が粗く推定した値に基づいて、入力信号は補償せずに、従来の位相増加アルゴリズムを用いて検出する際の位相の不確定性を除去して周波数オフセットを推定する構成を持つ受信装置である。なお、本実施形態における受信装置では、第1実施形態の周波数オフセット推定装置に代えて、非特許文献3に記載の発明を利用することも可能である。
受信装置は、周波数オフセット推定装置15、周波数不確定性除去制御部30、偏波分離部19、位相増加周波数オフセット推定装置20、周波数オフセット補償部29、第1の位相補償部24、第2の位相補償部25、第1の判定部26、および第2の判定部27を備えている。周波数オフセット補償部29は、NCO21、第1の乗算部22、および第2の乗算部23を備えている。
周波数オフセット推定装置15は、第1実施形態に記載の構成であり説明を省略する。
図7において図6の各部に対応する構成には同一の符号をつけており、その説明を省略する。
本第3実施形態では、まず、周波数オフセット推定装置15は、入力信号I+jQおよび入力信号I’+jQ’に対して、周波数オフセットを広帯域に粗く推定し、その推定値を周波数不確定性除去制御部30に出力する。
また、位相増加周波数オフセット推定装置20は、偏波分離部19から出力された信号に対して周波数オフセットを推定し、その推定値を周波数不確定性除去制御部30に出力する。
周波数オフセット補償部29で、入力信号I+jQおよび入力信号I’+jQ’の周波数オフセットが補償されたあと、第1の位相補償部24、第2の位相補償部25、第1の判定部26、および第2の判定部27で復調される。
周波数オフセット推定装置15の推定値に基づいて、周波数不確定性除去制御部30は、位相増加周波数オフセット推定装置20の周波数オフセット推定結果の不確定性を除去する。
以下、図面を参照して、本発明の第4実施形態について説明する。
図9は、本発明の第4の実施形態による周波数オフセット推定装置150の構成例を示すブロック図である。
この周波数オフセット推定装置は、周波数オフセット精推定部41、周波数オフセット粗推定部57、および掃引周波数範囲制御部58を備えている。さらに、周波数オフセット精推定部41は、第1の4乗回路42、第2の4乗回路43、第1の乗算回路44、第2の乗算回路45、NCO(numerically-controlled oscillator)46、第1のNシンボル加算回路47、第2のNシンボル加算回路48、第1の1/N除算回路49、第2の1/N除算回路50、第1の絶対値2乗回路51、第2の絶対値2乗回路52、偏波加算回路53、Uフレーム加算回路54、最大値掃引周波数検出回路55、および1/4除算回路56を備えている。
以下、本実施形態における周波数オフセット精推定部41の具体的な動作例を説明する。
例えば、周波数オフセットが5GHzである図23において、粗推定周波数fcoarseが4GHzであったとすると、シンボルレートRSは28GHzであることから、数式10よりNCO46の掃引周波数範囲は[2GHz~30GHz]となる。このため、周波数オフセット推定装置150は、20GHzのピークのみを検出してその両隣のピークは検出しないため、不確定性が無く20GHzを最大値掃引周波数として検出し、その1/4である5GHzを周波数オフセット推定値として出力することができる。
この図10を図24および図25と比べると、周波数オフセットの4倍である20GHzのピークが強調され、フレームごとに一時的に現れていた局所的なピークは相対的に低く抑えられている。従って、周波数オフセット推定装置150は、位相雑音や熱雑音によりフレームごとに一時的に現れていた局所的なピークに影響されることなく、正しい推定結果を算出することが可能となる。
以下、図面を参照して、本発明の第5実施形態について説明する。
図11は、本第5実施形態による周波数オフセット推定装置150の構成例を示すブロック図である。なお、図11において図9の各部に対応する構成には同一の符号をつけており、その説明を省略する。
以下、図面を参照して、本発明の第6実施形態について説明する。
図12は、本発明の第6実施形態による受信装置200の構成例を示すブロック図である。
この受信装置200は、周波数オフセット精推定部41、周波数オフセット粗推定部57、掃引周波数範囲制御部58、偏波分離部66、周波数オフセット補償部67、第1の位相補償部68、第2の位相補償部69、第1の判定部70、および第2の判定部71を備えている。ここで、図12に示す、周波数オフセット精推定部41、周波数オフセット粗推定部57、および掃引周波数範囲制御部58は、図9および図11に示す同一符号の各部とそれぞれ対応している。
偏波分離部66は、入力信号I+jQおよび入力信号I’+jQ’に対して偏波分離および残留分散の除去を実行する。周波数オフセット精推定部41は、偏波分離部66から出力された信号を入力し、周波数オフセットの推定を実行する。
以下、図面を参照して、本発明の第7実施形態について説明する。
図13は、本発明の第7実施形態による周波数オフセット推定装置150の構成例を示すブロック図である。
この周波数オフセット推定装置150は、周波数オフセット粗推定部57、周波数オフセット精推定部41、および周波数不確定性除去制御部97を備えている。周波数オフセット精推定部41は、第1のW乗回路88、第2のW乗回路89、第1のFFT部90、第2のFFT部91、第1の絶対値2乗回路92、第2の絶対値2乗回路93、偏波スペクトル加算回路94、Uフレームスペクトル加算回路95、および最大値周波数検出回路96を備えている。なお、Wは変調方式によって定まる数(正の整数)であり、周波数オフセットの無いコンスタレーション上の信号点の持つ回転対称性を2π/Wとする。QAM変調であればW=4、PAM変調であればW=2、M-PSK変調であればW=Mである。
この実験結果は、シンボルレートRSが10GHzの64QAMで変調された信号を受信した時のUフレームスペクトル加算回路95の出力信号の周波数スペクトルを示している。また周波数オフセット粗推定部57の推定値fCO は3.7GHzであった。図15より、数式12のfMAX は-4GHzである。数式14およびW=4 よりfCND(k)は、-1GHz(k=0)、1.5GHz(k=1)、4GHz(k=2)、6.5GHz(k=3)、…となり、このうち数式15を満たすiは2である。従って、数式16より周波数オフセットは4GHzと定めることができる。以上のように本実施形態により、RS/8=1.25GHzを越える周波数オフセットを推定することが可能となる。
以下、図面を参照して、本発明の第8実施形態について説明する。
図16は、本発明の第8実施形態による受信装置200の構成例を示すブロック図である。なお、図16において図12の各部に対応する構成には同一の符号をつけており、その説明を省略する。
偏波分離部66は、入力信号I+jQおよび入力信号I’+jQ’に対して偏波分離および残留分散の除去を実行する。周波数オフセット精推定部41は、偏波分離部66から出力された信号を入力し、周波数オフセットの推定を実行する。
また、「コンピュータ読み取り可能な記録媒体」とは、フレキシブルディスク、光磁気ディスク、ROM(read only memory)、CD(compact disc)-ROM等の可搬媒体、コンピュータシステムに内蔵されるハードディスク等の記憶装置のことをいう。さらに「コンピュータ読み取り可能な記録媒体」とは、インターネット等のネットワークや電話回線等の通信回線を介してプログラムを送信する場合の通信線のように、短時間の間、動的にプログラムを保持するもの、その場合のサーバやクライアントとなるコンピュータシステム内部の揮発性メモリのように、一定時間プログラムを保持しているものも含むものとする。また上記プログラムは、前述した機能の一部を実現するためのものであっても良く、さらに前述した機能をコンピュータシステムにすでに記録されているプログラムとの組み合わせで実現できるものであっても良い。
Claims (20)
- 受信信号の搬送波周波数と局部発振器の出力信号の周波数との差である周波数オフセットを推定する周波数オフセット推定装置であって、
所定のサンプリング周波数で予めサンプリングされた前記受信信号を周波数変換し、周波数の大きさの順に1からN(Nは任意の自然数)まで順序付けしたN個の周波数成分を持つ周波数スペクトルを出力する周波数変換部と、
前記周波数スペクトルの、1からN/2までの周波数成分番号を持った周波数成分である負の周波数成分とN/2+1からNまでの周波数成分番号を持った周波数成分である正の周波数成分とを、それぞれ周波数帯域制限する周波数帯域制限部と、
前記周波数帯域制限された前記周波数スペクトルの前記正の周波数成分と前記負の周波数成分とをそれぞれ2乗加算してそれぞれの電力を算出し、前記正の周波数成分の電力と前記負の周波数成分の電力とから算出した電力差の絶対値が、予め設定されたしきい値以下になるまで前記周波数スペクトルの全ての周波数成分を周波数軸上で循環して移動させ、前記しきい値以下になるまで移動させた移動量に基づいて前記周波数オフセットを推定する周波数オフセット推定制御部と、
を備える周波数オフセット推定装置。 - 前記周波数オフセット推定制御部は、前記周波数スペクトルの前記全ての周波数成分を前記周波数軸上で循環して移動させる場合に、
前記正の周波数成分の電力が前記負の周波数成分の電力より大きい場合は、前記周波数スペクトルの全周波数成分を負の方向に予め定められた大きさだけ移動し、移動後の周波数成分番号が1未満の場合は周波数成分番号にNを加算し、
前記正の周波数成分の電力が前記負の周波数成分の電力以下の場合は、前記周波数スペクトルの全周波数成分を正の方向に予め定められた大きさだけ移動し、移動後の周波数成分番号がNを越える場合は周波数成分番号からNを減算する請求項1に記載の周波数オフセット推定装置。 - 請求項1あるいは請求項2に記載の周波数オフセット推定装置と、
前記周波数オフセット推定装置によって推定された前記受信信号の前記周波数オフセットの値に基づいて、前記受信信号の前記周波数オフセットを補償する第1の周波数オフセット補償部と、
前記第1の周波数オフセット補償部によって補償された前記受信信号に対して、前記周波数オフセットを位相増加アルゴリズムに基づいて推定する、位相増加周波数オフセット推定部と、
前記位相増加周波数オフセット推定部によって推定された前記受信信号の前記周波数オフセットの値に基づいて、前記周波数オフセットを補償する第2の周波数オフセット補償部と、
を備える受信装置。 - 受信信号の搬送波周波数と局部発振器の出力信号の周波数との差である周波数オフセットを推定する周波数オフセット推定装置において用いられる、周波数オフセット推定方法であって、
所定のサンプリング周波数で予めサンプリングされた前記受信信号を周波数変換し、周波数の大きさの順に1からN(Nは任意の自然数)まで順序付けしたN個の周波数成分を持つ周波数スペクトルを出力する周波数変換手順と、
前記周波数スペクトルの、1からN/2までの周波数成分番号を持った周波数成分である負の周波数成分とN/2+1からNまでの周波数成分番号を持った周波数成分である正の周波数成分とを、それぞれ周波数帯域制限する周波数帯域制限手順と、
前記周波数帯域制限された前記周波数スペクトルの前記正の周波数成分と前記負の周波数成分とをそれぞれ2乗加算してそれぞれの電力を算出し、前記正の周波数成分の電力と前記負の周波数成分の電力とから算出した電力差の絶対値が、予め設定されたしきい値以下になるまで前記周波数スペクトルの全ての周波数成分を周波数軸上で循環して移動させ、前記しきい値以下になるまで移動させた移動量に基づいて前記周波数オフセットを推定する周波数オフセット推定制御手順と、
を備える周波数オフセット推定方法。 - 前記周波数オフセット推定制御手順において、前記周波数スペクトルの前記全ての周波数成分を前記周波数軸上で循環して移動させる場合に、
前記正の周波数成分の電力が前記負の周波数成分の電力より大きい場合は、前記周波数スペクトルの全周波数成分を負の方向に予め定められた大きさだけ移動し、移動後の周波数成分番号が1未満の場合は周波数成分番号にNを加算し、
前記正の周波数成分の電力が前記負の周波数成分の電力以下の場合は、前記周波数スペクトルの全周波数成分を正の方向に予め定められた大きさだけ移動し、移動後の周波数成分番号がNを越える場合は周波数成分番号からNを減算することにより、前記周波数スペクトルを循環して移動する請求項4に記載の周波数オフセット推定方法。 - 請求項4あるいは請求項5に記載の周波数オフセット推定方法による手順と、
前記周波数オフセット推定方法による手順によって推定された前記受信信号の前記周波数オフセットの値に基づいて、前記受信信号の前記周波数オフセットを補償する第1の周波数オフセット補償手順と、
前記第1の周波数オフセット補償手順によって補償された前記受信信号に対して、前記周波数オフセットを位相増加アルゴリズムに基づいて推定する、位相増加周波数オフセット推定手順と、
前記位相増加周波数オフセット推定手順によって推定された前記受信信号の前記周波数オフセットの値に基づいて、前記周波数オフセットを補償する第2の周波数オフセット補償手順と、
を備える受信方法。 - 受信信号の搬送波周波数と局部発振器の出力信号の周波数との差を推定する周波数オフセット推定装置であって、
所定のサンプリング周波数で予めサンプリングされた2つの偏波から成る受信信号の周波数スペクトルから周波数オフセットを推定する周波数オフセット粗推定部と、
前記周波数オフセット粗推定部の粗推定値に基づいて掃引周波数の範囲を決定する掃引周波数範囲制御部と、
前記掃引周波数範囲制御部により決定された前記掃引周波数の範囲において前記受信信号の周波数オフセットを推定する周波数オフセット精推定部と、
を備え、
前記周波数オフセット精推定部は、
前記受信信号の周波数オフセットの無いコンスタレーション上の信号点の持つ回転対称性を2π/W(W:正の整数)と定義した場合に、前記受信信号における前記2つの偏波をそれぞれW乗した後、該受信信号に対して該受信信号の周波数から前記掃引周波数を減算する周波数変換を行う第1の演算部と、
前記第1の演算部における各偏波の演算結果に対して、N(N:正の整数)シンボルの平均化または加算を行った後、絶対値または絶対値のべき乗の演算を行う第2の演算部と、
前記第2の演算部の後段側において、前記2つの偏波の演算結果を加算、またはピーク値の大きい方の偏波の演算結果を選択するとともに、前記NシンボルからなるフレームについてU(U:正の整数)フレームの加算または平均化を行う第3の演算部と、
前記第3の演算部の演算結果が最大値となる掃引周波数を検出し、当該掃引周波数を1/W倍して前記周波数オフセットを推定する第4の演算部と、
を有する周波数オフセット推定装置。 - 受信信号の搬送波周波数と局部発振器の出力信号の周波数との差を推定する周波数オフセット推定方法であって、
所定のサンプリング周波数で予めサンプリングされた2つの偏波から成る受信信号の周波数スペクトルから周波数オフセットを推定する周波数オフセット粗推定手順と、
前記周波数オフセット粗推定手順により推定された粗推定値に基づいて掃引周波数の範囲を決定する掃引周波数範囲制御手順と、
前記掃引周波数範囲制御手順により決定された前記掃引周波数の範囲において前記受信信号の周波数オフセットを推定する周波数オフセット精推定手順と、
を備え、
前記周波数オフセット精推定手順は、
前記受信信号の周波数オフセットの無いコンスタレーション上の信号点の持つ回転対称性を2π/W(W:正の整数)と定義した場合に、前記受信信号における前記2つの偏波をそれぞれW乗した後、該受信信号に対して該受信信号の周波数から前記掃引周波数を減算する周波数変換を行う第1の演算手順と、
前記第1の演算手順における各偏波の演算結果に対して、N(N:正の整数)シンボルの加算または平均化を行った後、絶対値または絶対値のべき乗の演算を行う第2の演算手順と、
前記第2の演算手順の後において、前記2つの偏波の演算結果を加算、またはピーク値の大きい方の偏波の演算結果を選択するとともに、前記NシンボルからなるフレームについてU(U:正の整数)フレームの加算または平均化を行う第3の演算手順と、
前記第3の演算手順の演算結果が最大値となる掃引周波数を検出し、当該掃引周波数を1/W倍して前記周波数オフセットを推定する第4の演算手順と、
を含む周波数オフセット推定方法。 - 受信信号の搬送波周波数と局部発振器の出力信号の周波数との差を推定する周波数オフセット推定装置であって、
所定のサンプリング周波数で予めサンプリングされた2つの偏波から成る受信信号の周波数スペクトルから周波数オフセットを推定する周波数オフセット粗推定部と、
前記受信信号または前記受信信号の分散が補償された信号に対して周期的な周波数オフセット推定特性を有する周波数オフセット精推定部と、
前記周波数オフセット粗推定部によって推定された前記周波数オフセットの値に基づいて、前記周波数オフセット精推定部によって推定された周波数オフセットの周波数不確定性を除去し、前記周波数オフセットを推定する周波数不確定性除去制御部と
を備える周波数オフセット推定装置。 - 前記周波数オフセット精推定部は、前記受信信号または前記受信信号の分散が補償された信号に対して位相増加アルゴリズムに基づいて前記周波数オフセットを推定する請求項9に記載の周波数オフセット推定装置。
- 前記周波数オフセット精推定部は、前記受信信号の周波数オフセットの無いコンスタレーション上の信号点の持つ回転対称性を2π/W(W:正の整数)と定義した場合に、前記受信信号における前記2つの偏波をそれぞれW乗した後に周波数スペクトルに変換し、その変換結果に対して絶対値または絶対値のべき乗の演算を行い、これら2つの偏波の周波数スペクトルを加算、またはピーク値の大きい方の偏波の周波数スペクトルを選択するとともに、N(N:正の整数)シンボルからなるフレームの周波数スペクトルについてU(U:正の整数)フレームの加算または平均化を行い、この演算結果が最大値となる周波数を検出する請求項9に記載の周波数オフセット推定装置。
- 前記周波数不確定性除去制御部は、前記周波数不確定性を除去し前記周波数オフセットを推定する際に、
前記周波数オフセット精推定部によって推定された前記周波数オフセットに基づいて、前記周波数不確定性を含む前記周波数オフセットの候補となる周波数を算出し、
周波数軸上で隣り合う前記周波数オフセットの候補となる周波数の中点を、隣り合う前記周波数オフセットの候補の境界とし、
前記周波数軸上で前記境界に基づいた領域の中から、前記周波数オフセット粗推定部によって推定された前記値が含まれる領域を選択し、
前記選択した領域に含まれる前記周波数オフセットの候補となる周波数を、前記周波数オフセットの推定値として選択する請求項9~11のいずれか1項に記載の周波数オフセット推定装置。 - 前記周波数オフセット粗推定部は、
前記受信信号を周波数変換し、周波数の大きさの順に1からN(Nは任意の自然数)まで順序付けしたN個の周波数成分を持つ周波数スペクトルを出力する周波数変換部と、
前記周波数スペクトルの、1からN/2までの周波数成分番号を持った周波数成分である負の周波数成分とN/2+1からNまでの周波数成分番号を持った周波数成分である正の周波数成分とを、それぞれ周波数帯域制限する周波数帯域制限部と、
前記周波数帯域制限された前記周波数スペクトルの前記正の周波数成分と前記負の周波数成分とをそれぞれ2乗加算してそれぞれの電力を算出し、前記正の周波数成分の電力と前記負の周波数成分の電力とから算出した電力差の絶対値が、予め設定されたしきい値以下になるまで前記周波数スペクトルの全ての周波数成分を周波数軸上で循環して移動させ、前記しきい値以下になるまで移動させた移動量に基づいて前記周波数オフセットを推定する周波数オフセット推定制御部と、
を備える請求項9~12のいずれか1項に記載の周波数オフセット推定装置。 - 請求項7、請求項9、請求項10、請求項11、請求項12、または請求項13に記載の周波数オフセット推定装置と、
前記周波数オフセット推定装置によって推定された前記受信信号の前記周波数オフセットの値に基づいて、前記受信信号の前記周波数オフセットを補償する周波数オフセット補償部と、
前記周波数オフセット補償部によって補償された前記受信信号に対して、位相を補償する位相補償部と、
前記位相を補償された前記受信信号のシンボルの判定を行う判定部と、
を備える受信装置。 - 受信信号の搬送波周波数と局部発振器の出力信号の周波数との差を推定する周波数オフセット推定方法であって、
所定のサンプリング周波数で予めサンプリングされた2つの偏波から成る受信信号の周波数スペクトルから周波数オフセットを推定する周波数オフセット粗推定手順と、
前記受信信号または前記受信信号の分散が補償された信号に対して周期的な周波数オフセット推定特性を有する周波数オフセット精推定手順と、
前記周波数オフセット粗推定手順によって推定された前記周波数オフセットの値に基づいて、前記周波数オフセット精推定手順によって推定された周波数オフセットの周波数不確定性を除去し、前記周波数オフセットを推定する周波数不確定性除去制御手順と、
を備える周波数オフセット推定方法。 - 前記周波数オフセット精推定手順は、前記受信信号または前記受信信号の分散が補償された信号に対して位相増加アルゴリズムに基づいて前記周波数オフセットを推定する請求項15に記載の周波数オフセット推定方法。
- 前記周波数オフセット精推定手順は、前記受信信号の周波数オフセットの無いコンスタレーション上の信号点の持つ回転対称性を2π/W(W:正の整数)と定義した場合に、前記受信信号における前記2つの偏波をそれぞれW乗した後に周波数スペクトルに変換し、その変換結果に対して絶対値または絶対値のべき乗の演算を行い、これら2つの偏波の周波数スペクトルを加算、またはピーク値の大きい方の偏波の周波数スペクトルを選択するとともに、N(N:正の整数)シンボルからなるフレームの周波数スペクトルについてU(U:正の整数)フレームの加算または平均化を行い、この演算結果が最大値となる周波数を検出する請求項15に記載の周波数オフセット推定方法。
- 前記周波数不確定性除去制御手順は、前記周波数不確定性を除去し前記周波数オフセットを推定する際に、
前記周波数オフセット精推定手順によって推定された前記周波数オフセットに基づいて、前記周波数不確定性を含む前記周波数オフセットの候補となる周波数を算出し、
周波数軸上で隣り合う前記周波数オフセットの候補となる周波数の中点を、隣り合う前記周波数オフセットの候補の境界とし、
前記周波数軸上で前記境界に基づいた領域の中から、前記周波数オフセット粗推定手順によって推定された前記値が含まれる領域を選択し、
前記選択した領域に含まれる前記周波数オフセットの候補となる周波数を、前記周波数オフセットの推定値として選択する請求項15~17のいずれか1項に記載の周波数オフセット推定方法。 - 前記周波数オフセット粗推定手順は、
前記受信信号を周波数変換し、周波数の大きさの順に1からN(Nは任意の自然数)まで順序付けしたN個の周波数成分を持つ周波数スペクトルを出力する周波数変換手順と、
前記周波数スペクトルの、1からN/2までの周波数成分番号を持った周波数成分である負の周波数成分とN/2+1からNまでの周波数成分番号を持った周波数成分である正の周波数成分とを、それぞれ周波数帯域制限する周波数帯域制限手順と、
前記周波数帯域制限された前記周波数スペクトルの前記正の周波数成分と前記負の周波数成分とをそれぞれ2乗加算してそれぞれの電力を算出し、前記正の周波数成分の電力と前記負の周波数成分の電力とから算出した電力差の絶対値が、予め設定されたしきい値以下になるまで前記周波数スペクトルの全ての周波数成分を周波数軸上で循環して移動させ、前記しきい値以下になるまで移動させた移動量に基づいて前記周波数オフセットを推定する周波数オフセット推定制御手順と、
を備える請求項15~18のいずれか1項に記載の周波数オフセット推定方法。 - 請求項8、請求項15、請求項16、請求項17、請求項18、または請求項19に記載の周波数オフセット推定方法による手順と、
前記周波数オフセット推定方法による手順によって推定された前記受信信号の前記周波数オフセットの値に基づいて、前記受信信号の前記周波数オフセットを補償する周波数オフセット補償手順と、
前記周波数オフセット補償手順によって補償された前記受信信号に対して、位相を補償する位相補償手順と、
前記位相を補償された前記受信信号のシンボルの判定を行う判定手順と、
を備える受信方法。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2012520505A JP5404926B2 (ja) | 2010-06-17 | 2011-06-17 | 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 |
US13/701,963 US8781029B2 (en) | 2010-06-17 | 2011-06-17 | Frequency offset estimation apparatus, reception apparatus, frequency offset estimation method, and reception method |
EP11795834.8A EP2584720B1 (en) | 2010-06-17 | 2011-06-17 | Frequency offset estimation apparatus, receiver apparatus, frequency offset estimation method, and reception method |
EP15152605.0A EP2903173B1 (en) | 2010-06-17 | 2011-06-17 | Frequency offset estimation apparatus and method |
CN201180028973.0A CN102934378B (zh) | 2010-06-17 | 2011-06-17 | 频率偏移估计装置、接收装置、频率偏移估计方法及接收方法 |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2010-138402 | 2010-06-17 | ||
JP2010138402 | 2010-06-17 | ||
JP2010-251868 | 2010-11-10 | ||
JP2010251868 | 2010-11-10 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2011158932A1 true WO2011158932A1 (ja) | 2011-12-22 |
Family
ID=45348324
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2011/063903 WO2011158932A1 (ja) | 2010-06-17 | 2011-06-17 | 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 |
Country Status (5)
Country | Link |
---|---|
US (1) | US8781029B2 (ja) |
EP (2) | EP2584720B1 (ja) |
JP (1) | JP5404926B2 (ja) |
CN (2) | CN102934378B (ja) |
WO (1) | WO2011158932A1 (ja) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014097583A1 (ja) * | 2012-12-20 | 2014-06-26 | 日本電気株式会社 | 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 |
CN105141564A (zh) * | 2015-07-28 | 2015-12-09 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | 一种高子载波数高阶调制水平ofdm采样频率同步方法 |
JP2016521045A (ja) * | 2013-04-10 | 2016-07-14 | 富士通株式会社 | 周波数オフセット推定方法、装置及びシステム |
JP2018509038A (ja) * | 2015-01-23 | 2018-03-29 | 華為技術有限公司Huawei Technologies Co.,Ltd. | コヒーレント受信機における微細周波数オフセット推定のための推定精度及びロバスト性の向上した方法及び装置 |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102422571B (zh) | 2009-05-18 | 2016-06-15 | 日本电信电话株式会社 | 信号生成电路、光信号发送装置、信号接收电路、光信号同步确立方法以及光信号同步系统 |
US9264278B2 (en) * | 2012-10-19 | 2016-02-16 | Apple Inc. | Robust scalable and adaptive frequency estimation and frequency tracking for wireless systems |
EP2932673B1 (en) * | 2012-12-14 | 2018-08-15 | Telefonaktiebolaget LM Ericsson (publ) | A receiver for multi carrier modulated signals |
CN104521206B (zh) * | 2013-07-31 | 2017-11-24 | 华为技术有限公司 | 一种多波带ofdm接收机、频率偏移补偿方法及系统 |
WO2015025468A1 (ja) * | 2013-08-21 | 2015-02-26 | 日本電気株式会社 | 周波数偏差補償方式、周波数偏差補償方法及び記憶媒体 |
EP2849362A1 (en) * | 2013-09-11 | 2015-03-18 | Alcatel Lucent | Blind frequency offset estimation for pulse-shaped signals |
WO2015052874A1 (ja) * | 2013-10-09 | 2015-04-16 | 日本電信電話株式会社 | 光伝送システム |
US9866364B2 (en) | 2014-08-29 | 2018-01-09 | Huawei Technologies Co., Ltd. | System and method for semi-orthogonal multiple access |
ES2782950T3 (es) * | 2014-09-24 | 2020-09-16 | Guangdong Oppo Mobile Telecommunications Corp Ltd | Método y dispositivo de comunicación inalámbrica para estimar el desplazamiento de frecuencia de la señal recibida |
WO2016161638A1 (zh) * | 2015-04-10 | 2016-10-13 | 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 |
US9819405B2 (en) * | 2015-06-12 | 2017-11-14 | Huawei Technologies Co., Ltd. | Systems and methods for transmission pairing mixed transmission modes |
EP3360476B1 (en) * | 2017-02-13 | 2020-09-30 | Stichting IMEC Nederland | A method and a device for detecting of a vital sign |
BR112020008286A2 (pt) | 2017-11-24 | 2020-10-20 | Huawei Technologies Co., Ltd. | dispositivo de processamento para um nó de acesso à rede para gerar símbolos de modulação compensados por fase, e método associado a esse dispositivo |
JP7064141B2 (ja) * | 2018-09-05 | 2022-05-10 | 日本電信電話株式会社 | 光受信装置、及び周波数オフセット推定方法 |
CN110995632B (zh) * | 2019-11-29 | 2023-03-21 | 深圳市统先科技股份有限公司 | 卫星通信带宽复用电路 |
US11251766B2 (en) * | 2020-01-13 | 2022-02-15 | Maxim Integrated Products, Inc. | Ultra-wide band frequency offset estimation systems and methods for analog coherent receivers |
CN115865239B (zh) * | 2021-09-27 | 2023-08-08 | 中国电信股份有限公司 | 基于载波聚合的信息上报方法、装置、介质及电子设备 |
CN114422314B (zh) * | 2021-12-22 | 2024-05-24 | 江苏科大亨芯半导体技术有限公司 | 多载波鉴频方法及系统 |
CN115333603B (zh) * | 2022-07-14 | 2023-06-27 | 航天恒星科技有限公司 | 载波同步方法、装置、电子设备及存储介质 |
CN115333654B (zh) * | 2022-10-13 | 2023-02-03 | 成都爱旗科技有限公司 | 一种频偏检测方法、系统及电子设备 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06509691A (ja) * | 1991-08-05 | 1994-10-27 | フォード モーター カンパニー | ラジオ受信機用の隣接チャネル・コントローラ |
JP2005524327A (ja) * | 2002-05-01 | 2005-08-11 | アイビキュイティ・デジタル・コーポレイション | Fmデジタル音声放送受信機のための隣接チャンネル干渉の軽減 |
JP2009530918A (ja) * | 2006-03-15 | 2009-08-27 | クゥアルコム・インコーポレイテッド | タイミング同期に適応する周波数トラッキング |
JP2010516076A (ja) * | 2007-01-05 | 2010-05-13 | エルジー エレクトロニクス インコーポレイティド | 周波数オフセットを考慮して循環シフトを設定する方法 |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1482699A3 (en) * | 1998-06-08 | 2005-02-09 | Telefonaktiebolaget LM Ericsson (publ) | Burst structure for multicarrier transmission, and synchronisation of bursts, symbols and frequency |
JP4983365B2 (ja) * | 2006-05-16 | 2012-07-25 | ソニー株式会社 | 無線通信装置 |
CN101442364B (zh) * | 2007-11-19 | 2011-10-19 | 富士通株式会社 | 光相干接收机、光相干接收机用频差估计装置及方法 |
WO2009093579A1 (ja) | 2008-01-21 | 2009-07-30 | Nec Corporation | 通信装置、通信システム、制御方法及び制御プログラム |
US8649750B2 (en) * | 2008-03-11 | 2014-02-11 | Intel Corporation | Techniques for efficient carrier recovery for passband communciation systems |
CN102422571B (zh) * | 2009-05-18 | 2016-06-15 | 日本电信电话株式会社 | 信号生成电路、光信号发送装置、信号接收电路、光信号同步确立方法以及光信号同步系统 |
-
2011
- 2011-06-17 US US13/701,963 patent/US8781029B2/en active Active
- 2011-06-17 WO PCT/JP2011/063903 patent/WO2011158932A1/ja active Application Filing
- 2011-06-17 CN CN201180028973.0A patent/CN102934378B/zh active Active
- 2011-06-17 EP EP11795834.8A patent/EP2584720B1/en active Active
- 2011-06-17 JP JP2012520505A patent/JP5404926B2/ja active Active
- 2011-06-17 CN CN201410812968.4A patent/CN104539342B/zh active Active
- 2011-06-17 EP EP15152605.0A patent/EP2903173B1/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06509691A (ja) * | 1991-08-05 | 1994-10-27 | フォード モーター カンパニー | ラジオ受信機用の隣接チャネル・コントローラ |
JP2005524327A (ja) * | 2002-05-01 | 2005-08-11 | アイビキュイティ・デジタル・コーポレイション | Fmデジタル音声放送受信機のための隣接チャンネル干渉の軽減 |
JP2009530918A (ja) * | 2006-03-15 | 2009-08-27 | クゥアルコム・インコーポレイテッド | タイミング同期に適応する周波数トラッキング |
JP2010516076A (ja) * | 2007-01-05 | 2010-05-13 | エルジー エレクトロニクス インコーポレイティド | 周波数オフセットを考慮して循環シフトを設定する方法 |
Non-Patent Citations (5)
Title |
---|
A. LEVEN ET AL.: "Frequency estimation in intradyne reception", IEEE PHOTONICS TECHNOLOGY LETTERS, vol. 19, March 2007 (2007-03-01), pages 366 - 368 |
CIBLAT ET AL.: "Blind NLLS carrier frequency-offset estimation for QAM, PSK, and PAM modulations: performance at low SNR", IEEE TRANSACTIONS ON COMMUNICATIONS, vol. 54, October 2006 (2006-10-01), pages 1725 - 1730 |
K. PIYAWANNO ET AL.: "Fast and accurate automatic frequency control for coherent receivers", ECOC2009, September 2009 (2009-09-01) |
M. K. NEZAMI ET AL.: "DFT-based frequency acquisition algorithm for large carrier offsets in mobile satellite receivers", ELECTRONICS LETTERS, vol. 37, March 2001 (2001-03-01), pages 386 - 387 |
M. SELMI ET AL.: "Accurate digital frequency offset estimator for coherent PolMux QAM transmission systems", ECOC2009, 8 September 2009 (2009-09-08) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014097583A1 (ja) * | 2012-12-20 | 2014-06-26 | 日本電気株式会社 | 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 |
JPWO2014097583A1 (ja) * | 2012-12-20 | 2017-01-12 | 日本電気株式会社 | 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 |
US9621278B2 (en) | 2012-12-20 | 2017-04-11 | Nec Corporation | Frequency offset compensation apparatus and frequency offset compensation method |
JP2016521045A (ja) * | 2013-04-10 | 2016-07-14 | 富士通株式会社 | 周波数オフセット推定方法、装置及びシステム |
JP2018509038A (ja) * | 2015-01-23 | 2018-03-29 | 華為技術有限公司Huawei Technologies Co.,Ltd. | コヒーレント受信機における微細周波数オフセット推定のための推定精度及びロバスト性の向上した方法及び装置 |
CN105141564A (zh) * | 2015-07-28 | 2015-12-09 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | 一种高子载波数高阶调制水平ofdm采样频率同步方法 |
CN105141564B (zh) * | 2015-07-28 | 2018-10-12 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | 一种高子载波数高阶调制水平ofdm采样频率同步方法 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2011158932A1 (ja) | 2013-08-19 |
EP2903173B1 (en) | 2019-09-04 |
US8781029B2 (en) | 2014-07-15 |
JP5404926B2 (ja) | 2014-02-05 |
CN102934378A (zh) | 2013-02-13 |
CN102934378B (zh) | 2015-01-14 |
CN104539342B (zh) | 2018-01-09 |
EP2584720A1 (en) | 2013-04-24 |
EP2584720A4 (en) | 2013-11-27 |
EP2903173A1 (en) | 2015-08-05 |
US20130070874A1 (en) | 2013-03-21 |
CN104539342A (zh) | 2015-04-22 |
EP2584720B1 (en) | 2015-03-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5404926B2 (ja) | 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 | |
JP5304083B2 (ja) | 周波数オフセットモニタリング装置及びコヒーレント光受信機 | |
US8908809B2 (en) | Complexity reduced feed forward carrier recovery methods for M-QAM modulation formats | |
JP5391660B2 (ja) | 光コヒーレント受信機、周波数オフセット推定装置、及び光コヒーレント受信機の方法 | |
JP5651990B2 (ja) | デジタルコヒーレント受信器および受信方法 | |
EP2583424B1 (en) | Method for phase and oscillator frequency estimation | |
US8514922B2 (en) | Filter coefficient control apparatus and method | |
US8463121B2 (en) | Ultra wide-range frequency offset estimation for digital coherent optical receivers | |
KR0157500B1 (ko) | 자동주파수조절방법 및 그 장치 | |
US9960857B2 (en) | System and method for blind frequency recovery | |
CN105612700B (zh) | 用于表征光接收信号的色度色散的装置 | |
US20170288932A1 (en) | Carrier frequency offset estimation in a receiver | |
US7430247B2 (en) | Carrier frequency detection for N-ary phase modulated signal | |
US8989316B1 (en) | Method for estimating and correcting a carrier frequency offset over dispersive but spectrally flat channels | |
JP2020010195A (ja) | 周波数推定装置および追尾受信機 | |
KR100594269B1 (ko) | 주파수 위상 동기루프회로 및 이를 사용하는 atscdtv 복조기. | |
Zhou et al. | DFT-based carrier recovery for Satellite DVB Receivers | |
KR100665100B1 (ko) | 디지털 초협대역 단말기용 주파수 옵셋 자동 보상 장치 및그 방법과 그를 이용한 수신 시스템 | |
JP2000156655A (ja) | 等化器 | |
JP5656750B2 (ja) | シングルキャリア受信装置 | |
JP2014017609A (ja) | 受信装置および受信方法 | |
KR0157498B1 (ko) | 자동주파수조절장치 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201180028973.0 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 11795834 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012520505 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13701963 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2011795834 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |