WO2011007829A1 - 周波数領域多重信号受信方法及び周波数領域多重信号受信装置 - Google Patents
周波数領域多重信号受信方法及び周波数領域多重信号受信装置 Download PDFInfo
- Publication number
- WO2011007829A1 WO2011007829A1 PCT/JP2010/061972 JP2010061972W WO2011007829A1 WO 2011007829 A1 WO2011007829 A1 WO 2011007829A1 JP 2010061972 W JP2010061972 W JP 2010061972W WO 2011007829 A1 WO2011007829 A1 WO 2011007829A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- frequency
- signal
- frequency domain
- fourier transform
- digital signal
- 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
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2647—Arrangements specific to the receiver only
- H04L27/2655—Synchronisation arrangements
- H04L27/2668—Details of algorithms
- H04L27/2669—Details of algorithms characterised by the domain of operation
- H04L27/2672—Frequency domain
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
- H04L25/03159—Arrangements for removing intersymbol interference operating in the frequency domain
-
- 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
- H04L27/2649—Demodulators
- H04L27/265—Fourier transform demodulators, e.g. fast Fourier transform [FFT] or discrete Fourier transform [DFT] demodulators
- H04L27/26522—Fourier transform demodulators, e.g. fast Fourier transform [FFT] or discrete Fourier transform [DFT] demodulators using partial FFTs
-
- 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
- H04L27/266—Fine or fractional frequency offset determination and synchronisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
- H04L2025/0335—Arrangements for removing intersymbol interference characterised by the type of transmission
- H04L2025/03375—Passband transmission
- H04L2025/03414—Multicarrier
Definitions
- the present invention relates to a frequency domain multiplexed signal receiving method and a frequency domain multiplexed signal receiving apparatus in a communication system that multiplexes signals in the frequency domain.
- the present invention also relates to a frequency domain multiplexed signal receiving method and a frequency domain multiplexed signal receiving apparatus that equalize a received signal multiplexed in the frequency domain without performing discrete Fourier transform.
- OFDM orthogonal frequency division multiplexing
- ADC analog digital converter
- LAN wireless local area network
- IEEE802.11a it is known that a band of about 16.6 MHz is used as a data signal in a signal band of 20 MHz (megahertz).
- single carrier transmission requires an ADC that can adapt to a sampling frequency that is about twice the frequency band of the signal. Therefore, the advantage of being able to set the frequency lower than twice the baud rate in OFDM is the advantage of low sampling frequency especially in optical communication whose application limit is limited by ADC sampling frequency and high frequency wireless communication. It is possible to transmit signals efficiently by ADC. Furthermore, in OFDM, as the number of frequency channels increases, the setting of the guard interval becomes easier, and flat fading can be assumed in each frequency channel. For this reason, the number of frequency channels of about 64 to 1024 is selected in a practical wireless system. In general, an OFDM system has a frequency channel corresponding to a direct current component with a frequency of “0” in the receiving apparatus. In the frequency channel corresponding to this DC component, the characteristics are deteriorated due to the influence of inter-signal interference and noise. Therefore, frequency channels corresponding to these are not generally used.
- frequency division multiplexing FDM: “Frequency Division Division Multiplexing”
- OFDM orthogonal frequency division multiplexing
- the first problem will be described.
- PAPR peak to average power ratio
- a large number of frequency channels cannot be set.
- the number of OFDM frequency channels included in the frequency band of the received signal decreases due to the influence of the filter.
- the number of subcarriers is reduced, it has a frequency channel centered on the DC component as in IEEE802.11a and is not used for data signal transmission because this frequency channel is affected by noise and inter-signal interference. This increases the loss in communication speed.
- a two-carrier OFDM signal obtained by multiplexing a signal having a transmission symbol rate of 5 Gbaud in two frequency bands is received.
- the interval I between adjacent frequency channels is 5 GHz (gigahertz).
- the discrete Fourier transform is a transform that converts a signal in the time domain into a signal in the frequency domain.
- the discrete Fourier transform of 4 points center frequency: ⁇ 2.5 GHz, ⁇ 5 GHz
- it is converted into just two frequency channels.
- the symbol C ⁇ b> 1 indicates the center frequency of the signal channel 1
- the symbol C ⁇ b> 2 indicates the center frequency of the signal channel 2.
- a symbol P indicates a frequency band to which four frequency channels of the conventional Fourier transform correspond.
- the transmission device that generates the transmission signal and the reception device that receives the transmitted signal as a reception signal are connected to different reference signal generation devices. There is usually a frequency shift in the frequency of each reference signal generator. Therefore, the receiving apparatus accurately compensates for the frequency deviation from the received signal when a specific frequency region is cut out from the received signal transmitted in synchronization with the frequency of the transmitting apparatus or when Fourier transform is performed. There is a need. In the receiving apparatus, when the frequency shifts, the signal of the adjacent frequency channel leaks, and the signal quality deteriorates. In particular, in optical communication that performs synchronous detection, a frequency shift between a laser beam used for transmission and a laser beam used for reception is large, and such a problem is likely to occur.
- the receiving apparatus 190 includes a digital signal acquisition circuit 1901, a Fourier transform circuit 1902, and a decoding circuit 1903.
- the received signal is converted into a digital signal by a digital signal acquisition circuit 1901, Fourier-transformed by a Fourier transform circuit 1902, and decoded by a decoding circuit 1903.
- the frequency shift is not completely compensated in the Fourier transform circuit 1902, interference power remains and degrades the signal.
- FIG. 14A and FIG. 14B the influence which this frequency shift has is demonstrated.
- FIG. 14A and 14B show an example in which two signal channels of frequency channels A and B are acquired.
- the center frequencies of frequency channels A and B are denoted by fa and fb, respectively.
- FIG. 14A shows a case where no frequency deviation remains. In this case, if Fourier transform is performed, it is possible to obtain the power at the center of each signal. At this time, the signal of the adjacent frequency channel is 0 at the center of the signal of the different frequency channel. Therefore, no inter-signal interference occurs.
- FIG. 14B in the state where the frequency of the frequency channels A ′ and B ′ of the received signal is shifted to fa ′ and fb ′ with respect to the center frequencies fa and fb, The deviation ⁇ f remains. This inter-signal interference causes a problem that communication quality deteriorates.
- a receiving apparatus that receives a signal obtained by multiplexing signals in the frequency domain needs to perform highly accurate frequency deviation compensation in order to perform discrete Fourier transform.
- frequency shift compensation before the inverse Fourier transform performed on the transmission device side, there arises a problem that the load on the communication system becomes large, such as insertion of a known signal and introduction of a complicated frequency estimation algorithm. .
- a first object of an embodiment of the present invention is to provide a frequency domain multiplexed signal receiving method and a frequency domain multiplexed signal receiving apparatus capable of sampling at a frequency lower than twice the frequency band of the received signal.
- a second object of the embodiment of the present invention is to provide a frequency domain multiplexed signal receiving method and a frequency domain multiplexed signal receiving apparatus that perform decoding without performing Fourier transform independently.
- a frequency domain multiplexed signal receiving method for decoding a received signal multiplexed in the frequency domain acquires a digital signal from the received signal multiplexed in the frequency domain.
- a digital signal acquisition step an offset discrete Fourier transform step for performing an offset discrete Fourier transform of an odd number of discrete points based on the obtained digital signal, and a frequency domain digital of the frequency domain obtained by the offset discrete Fourier transform
- a decoding step of decoding the frequency domain digital signal of one or more frequency channels A decoding step of decoding the frequency domain digital signal of one or more frequency channels.
- a frequency shift compensation step for compensating a frequency shift with respect to the frequency of the obtained digital signal may be further included, and the offset discrete Fourier transform step may include the frequency shift.
- An offset discrete Fourier transform with an odd number of discrete points may be performed on the digital signal frequency-converted to the compensated frequency.
- a residual frequency shift of each frequency channel or a frequency shift common to all frequency channels is estimated from the digital signal converted into the frequency domain by the offset discrete Fourier transform step.
- a frequency deviation estimation step for updating frequency deviation information indicating the frequency deviation, and the frequency deviation compensation step may be performed on the frequency of the obtained digital signal based on the frequency deviation information.
- the frequency deviation may be compensated.
- the offset discrete Fourier transform step performs an offset discrete Fourier transform with an odd number of discrete points on the digital signal by a convolution operation, and calculates and outputs only the frequency channel corresponding to the digital signal. May be.
- the number of discrete points of the offset discrete Fourier transform may be 3, and two frequency channels may be acquired.
- the offset discrete Fourier transform step includes Independent frequency deviation compensation may be performed for each frequency channel.
- a frequency domain multiplex signal receiving apparatus for decoding a received signal multiplexed in the frequency domain according to the B2 embodiment of the present invention
- the digital signal acquisition unit for acquiring a digital signal from the received signal multiplexed in the frequency domain, and the obtained An offset discrete Fourier transform unit for performing an offset discrete Fourier transform with an odd number of discrete points based on the digital signal, and a frequency domain digital signal in the frequency domain obtained by the offset discrete Fourier transform
- a decoding unit for decoding the frequency domain digital signal of the frequency channel.
- the frequency domain multiplexed signal receiver may further include a frequency shift compensation unit that performs frequency shift compensation on the frequency of the obtained digital signal, and the offset discrete Fourier transform The unit may perform an offset discrete Fourier transform with an odd number of discrete points on the digital signal frequency-converted to a frequency in which the frequency shift is compensated.
- a frequency shift estimation unit is further provided that estimates a residual frequency shift in the offset discrete Fourier transform unit from the obtained digital signal and updates frequency shift information indicating the frequency shift.
- the frequency deviation compensation unit may compensate the frequency deviation with respect to the frequency of the obtained digital signal based on the frequency deviation information.
- a frequency domain multiplexed signal receiving method for decoding a received signal multiplexed in the frequency domain includes a digital signal obtaining step for obtaining a digital signal from a received signal multiplexed in the frequency domain, A branch step for branching to the number of frequency channels to be decoded with respect to the digital signal, an initial coefficient storage step for storing different coefficients with low correlation as initial weights for each of the branched signal sequences, and the branch An equalization step for equalizing each signal sequence with a different coefficient, and a decoding step for performing decoding on each equalized signal sequence.
- a frequency domain multiplexed signal receiving method for decoding a received signal multiplexed in the frequency domain according to the A2 embodiment of the present invention, a digital signal obtaining step for obtaining a digital signal from the received signal multiplexed in the frequency domain, A branch step for branching to the number of frequency channels to be decoded with respect to the digital signal, and a coefficient having a high correlation with a coefficient of a discrete Fourier transform corresponding to each frequency channel with respect to each of the branched signal sequences, or An initial coefficient storage step for storing a coefficient including at least a part thereof as an initial weight; an equalization step for performing equalization using the coefficient stored in the initial coefficient storage step as an initial weight; and each equalized signal sequence
- a decoding step for performing decoding is included.
- the branching step may branch after performing frequency conversion so that the center frequency of each frequency channel is near a specific frequency component when the digital signal is branched.
- a coefficient having a high correlation with a coefficient of the discrete Fourier transform corresponding to the specific frequency component in the discrete Fourier transform, or a coefficient including at least a part thereof may be stored as the initial weight.
- the equalization coefficient may be adjusted so that the equalized signal series does not converge to a signal series indicating the same signal in the equalization step.
- a frequency domain multiplexed signal receiving method for decoding a received signal multiplexed in the frequency domain acquires two received signals multiplexed in the frequency domain for different polarization components, A digital signal obtaining step for obtaining a digital signal from the signal, a branching step for branching the obtained two digital signals into the number of frequency channels to be decoded, and among the branched signal sequences, k
- the coefficients a k, 0 , a k, 1 ,..., A k, n ⁇ 1 of the n- point discrete Fourier transform are used.
- a frequency domain signal receiving apparatus for decoding a reception signal multiplexed in the frequency domain according to the A4 embodiment of the present invention, a digital signal acquisition unit for acquiring a digital signal from the reception signal multiplexed in the frequency domain, and the obtained A branching unit that branches into the number of frequency channels to be decoded within the reception band for receiving the received signal with respect to the digital signal, and an initial coefficient storage unit that stores an initial weight used for equalization of the signal of each frequency channel; An equalization unit that performs equalization on each of the branched signal sequences with a coefficient input from the initial coefficient storage unit, and a decoding unit that performs decoding on the equalized signal sequences. Have.
- a frequency domain signal receiving apparatus for decoding an optical signal multiplexed in a frequency domain a photoelectric conversion unit for converting the optical signal into an electrical signal, and acquiring a digital signal from the electrical signal A digital signal acquisition unit; a branching unit that branches to the number of frequency channels to be decoded with respect to the obtained digital signal; and an initial coefficient storage unit that stores initial weights used for equalization of the signals of the frequency channels.
- An equalization unit that performs equalization with the coefficients input from the initial coefficient storage unit with respect to each branched signal sequence, and a decoding unit that performs decoding with respect to each equalized signal sequence, Have
- the digital signal acquisition step acquires a digital signal from the received signal multiplexed in the frequency domain.
- the offset discrete Fourier transform step performs an offset discrete Fourier transform with an odd number of discrete points based on the obtained digital signal.
- the decoding step performs decoding on the frequency domain digital signal of the frequency domain obtained by the offset discrete Fourier transform, and the frequency domain digital signal of one or more frequency channels.
- the digital signal acquisition step acquires a digital signal from the received signal multiplexed in the frequency domain.
- the obtained digital signal is branched into the number of frequency channels to be decoded.
- different coefficients having low correlation are stored as initial weights for each branched signal sequence.
- equalization equalization is performed on the branched signal sequences with different coefficients.
- decoding decoding is performed on each equalized signal sequence.
- the digital signal acquisition step acquires a digital signal from the received signal multiplexed in the frequency domain.
- the obtained digital signal is branched into the number of frequency channels to be decoded.
- the initial coefficient storing step a coefficient having a high correlation with a coefficient of the discrete Fourier transform corresponding to each frequency channel or a coefficient including at least a part thereof is stored as an initial weight for each branched signal sequence.
- equalization step equalization is performed using the coefficients stored in the initial coefficient storage step as initial weights.
- decoding step decoding is performed on each equalized signal sequence.
- FIG. 1 is a configuration diagram showing a configuration of a receiving apparatus according to the B1 embodiment of the present invention.
- the receiving apparatus 210 shown in this figure includes a digital signal acquisition circuit 2101, an odd-offset discrete Fourier transform circuit 2102, and a decoding circuit 2103.
- the digital signal acquisition circuit 2101 converts the received signal (analog signal) into a digital signal.
- the odd offset discrete Fourier transform circuit 2102 performs an offset discrete Fourier transform described later on the digital signal converted according to the received analog signal, and outputs a frequency channel corresponding to the received signal.
- the decoding circuit 2103 decodes the transmitted signal from the reception signal of each frequency channel converted by the odd-offset discrete Fourier transform circuit 2102.
- j represents an imaginary unit.
- the received signal can be converted into a frequency domain signal without contradiction.
- one of the frequency channels corresponds to a DC component. Therefore, one of the frequency channels cannot be used due to the interference power, resulting in a problem that the throughput is lowered.
- FIG. 2 is a schematic diagram showing the intensity distribution of a signal when synchronous detection is performed using a reference signal having a center frequency of the signal.
- a symbol Q indicates a frequency band to which the three frequency channels in the present embodiment correspond.
- the interval I between adjacent frequency channels is 5 GHz (gigahertz).
- the digital signal acquisition circuit 2101 performs conversion using an analog-digital converter (ADC) with a sampling frequency of 15 GS / s (gigasample / second), and an odd-offset discrete Fourier transform circuit 2102. Can be decoded by the decoding circuit 2103. Since FIG. 2 represents the spectrum of a signal in the baseband, a negative frequency less than the direct current component (frequency 0) is defined. When the ADC is 15 GHz, -7.5 GHz to 0 GHz is equivalent to 7.5 GHz to 15 GHz.
- ADC analog-digital converter
- an offset discrete Fourier transform is performed.
- n complex number sequences Z 0 ,..., Z n-1 can be obtained from the arithmetic expression shown in Expression (B2).
- j is an imaginary unit
- n is an odd number.
- the total number (n) of frequency channels obtained by the discrete Fourier transform is referred to as “point number” of the discrete Fourier transform.
- the number of frequency channels multiplexed in the frequency domain by OFDM is an even number (see Non-Patent Document 2), and it is easier to perform data signal processing if one or even number of frequency channels are handled. Therefore, in order to acquire an even frequency channel signal with an ADC having a low sampling rate, it is effective to set the number of points of the discrete Fourier transform to an odd number. In this case, the odd-offset discrete Fourier transform circuit 2102 outputs the result of the center even number of frequency channels to the decoding circuit 2103.
- the phenomenon shown in FIG. 2 in which one of the three frequencies extends over a plurality of frequencies also occurs when the number of points of the discrete Fourier transform is larger than 3. That is, in the formula (B2), the frequency domain signal Z (n-1) / 2 corresponding to this frequency band cannot express a specific frequency, so that the conversion to the original frequency domain is incomplete. By discarding such incomplete frequency channel information, the odd-offset discrete Fourier transform circuit 2102 can acquire a signal existing in a specific frequency region.
- the offset discrete Fourier transform in the odd-offset discrete Fourier transform circuit 2102 may be performed for each block or may be performed as a convolution operation.
- X k received sequence
- X d the coefficient multiplied by Xi in equation (B2) can be convolved with the received signal.
- G the head position of the signal.
- FIG. 3 is a flowchart showing a processing procedure in the present embodiment.
- the digital signal acquisition circuit 2101 converts it into a digital signal according to the received reception signal (step S11).
- the odd offset discrete Fourier transform circuit 2102 performs an offset discrete Fourier transform with an odd number of points (step S12).
- the decoding circuit 2103 decodes the signal of each frequency channel corresponding to the obtained signal (step S13). In the reception method according to the above processing procedure, it is possible to decode a desired reception signal even when an ADC having a sampling clock frequency lower than twice the frequency band is used.
- the odd-offset discrete Fourier transform circuit 2102 can also compensate for the frequency shift according to the frequency shift information indicating the frequency shift of the converted digital signal.
- the decoding circuit 2103 can also feed back the estimated frequency shift signal to the frequency shift compensation circuit 2102.
- FIG. 4 is a block diagram showing a configuration of a receiving apparatus according to the second B2 embodiment of the present invention.
- the receiving apparatus 220 shown in this figure includes a digital signal acquisition circuit 2201, an odd-offset discrete Fourier transform circuit 2202, a decoding circuit 2203, and a frequency shift compensation circuit 2204.
- the digital signal acquisition circuit 2201 converts the received signal (analog signal) into a digital signal.
- the frequency shift compensation circuit 2204 receives the received signal using one of the received signal corresponding to the test signal inserted into the transmitted signal, the characteristics of the modulation method of the transmitted signal, and the frequency shift information input from the other receiving circuit block. Compensates for frequency shifts in the signal.
- the frequency deviation occurring in the reception signal indicates a frequency deviation between the frequency at which the transmission signal is generated and the frequency that the reception device 220 uses as a reference.
- the odd-offset discrete Fourier transform circuit 2202 is converted according to the received analog signal, and the frequency shift compensation circuit 2204 performs an offset discrete Fourier transform described later on the digital signal compensated for the frequency shift, and a frequency corresponding to the received signal. Output the channel.
- the decoding circuit 2203 decodes the transmitted signal from the reception signal of each frequency channel converted by the odd-offset discrete Fourier transform circuit 2202.
- the frequency shift compensation circuit 2204 uses the equation (B3) for the frequency shift compensated received signal X ′ k in which the frequency shift is compensated for the received signal having the frequency shift. To calculate.
- the odd-offset discrete Fourier transform circuit 2202 performs offset discrete Fourier transform on the frequency deviation compensated received signal X ′ k to compensate for the orthogonality collapse due to the frequency deviation, and then transmit at an appropriate frequency position.
- the signal can be separated.
- the frequency shift compensation circuit 2204 the frequency conversion shown in the equation (B4) can be performed in advance.
- the frequency conversion shown in Expression (B4) is performed by converting the frequency to a frequency shifted by half the frequency width of the frequency channel occupation frequency band.
- the frequency shift compensation circuit 2204 can perform the Fourier transform in the odd-offset discrete Fourier transform circuit 2202 after performing the frequency transform shown in Expression (B4).
- the offset discrete Fourier transform can be realized by the frequency shift compensation circuit 2204 and the odd discrete Fourier transform circuit 2202. As ⁇ f, the frequency shift signal estimated in the decoding circuit 2203 may be fed back.
- FIG. 5 is a flowchart showing a processing procedure in the present embodiment.
- the receiving device 220 converts the digital signal acquisition circuit 2201 into a digital signal according to the received received signal (step S21).
- the frequency deviation compensation circuit 2204 compensates for the frequency deviation according to the frequency deviation information indicating the frequency deviation of the converted digital signal (step S22).
- the odd-offset discrete Fourier transform circuit 2202 performs an offset discrete Fourier transform with an odd number of points (step S23).
- the decoding circuit 2203 decodes the signal of each frequency channel corresponding to the obtained signal.
- the decoding circuit 2203 feeds back the estimated frequency shift signal to the frequency shift compensation circuit 2204 (step S24).
- the frequency shift information recorded in step S24 is derived by repeated arithmetic processing and used as reference information in the frequency shift compensation circuit 2204. In the reception method according to the above processing procedure, it is possible to decode a desired reception signal even when an ADC having a sampling clock frequency lower than twice the frequency band is used.
- FIG. 6 is a block diagram showing a configuration of a receiving apparatus according to the B3 embodiment of the present invention.
- the receiving apparatus 230 shown in this figure includes a digital signal acquisition circuit 2301, an odd-offset discrete Fourier transform circuit 2302, a decoding circuit 2303, and a frequency shift estimation circuit 2304.
- the digital signal acquisition circuit 2301 converts the received signal (analog signal) into a digital signal.
- the frequency shift estimation circuit 2304 receives the received signal using one of the received signal corresponding to the test signal inserted into the transmitted signal, the characteristics of the modulation method of the transmitted signal, and the frequency shift information input from other receiving circuit blocks. Compensates for frequency shifts in the signal.
- the frequency deviation occurring in the reception signal indicates a frequency deviation between the frequency at which the transmission signal is generated and the frequency that is used as a reference by the reception device 230.
- the odd-offset discrete Fourier transform circuit 2302 is converted according to the received analog signal, performs an odd-offset discrete Fourier transform in consideration of the frequency shift, and outputs a frequency channel corresponding to the received signal.
- the frequency shift estimation circuit 2304 estimates residual frequency shift information using the frequency channel digital signal output from the odd offset discrete Fourier transform circuit 2302, and outputs it to the offset Fourier transform circuit 2302. Further, the frequency shift estimation circuit 2304 outputs the digital signal to the decoding circuit 2303.
- the decoding circuit 2303 decodes the transmitted signal from the received signal of each frequency channel, which is a digital signal output from the frequency shift estimation circuit 2304 and converted by the odd-offset discrete Fourier transform circuit 2302.
- the output signal of the kth frequency channel is input to the frequency shift estimation circuit 2304, and the frequency shift is estimated.
- the frequency deviation compensation method can be estimated by using, for example, a blind algorithm, a frequency deviation compensation method using a known signal proposed in wireless communication, or the like (for example, see Non-Patent Document 4). Since the estimated frequency shift ⁇ f 0 is a frequency shift remaining in the offset discrete Fourier transform circuit 2302, the offset discrete Fourier transform circuit 2302 updates the frequency shift information as ( ⁇ f + ⁇ f 0 ), An increase in interference power due to frequency shift can be prevented. Alternatively, the odd-offset discrete Fourier transform circuit has a function of compensating for the frequency deviation in this way, so that the residual frequency deviation in each frequency channel can be individually compensated.
- the frequency shift estimation circuit 2304 can estimate the frequency shift for each frequency channel and output the frequency shift information to the odd-offset discrete Fourier transform circuit 2302 to compensate for the frequency shift in each frequency channel. Also, the decoding circuit 2303 can estimate the frequency shift of each frequency channel and output it to the odd offset discrete Fourier transform circuit 2302. In this case, ⁇ f in equation (B5) is different for each frequency channel, and ⁇ f k can be used as the frequency shift of the kth frequency channel.
- FIG. 7 is a flowchart showing a processing procedure in the present embodiment.
- the receiving device 230 converts the digital signal acquisition circuit 2301 into a digital signal according to the received received signal (step S31).
- the odd offset discrete Fourier transform circuit 2302 compensates for the frequency shift according to the frequency shift information indicating the frequency shift of the converted digital signal, and performs the offset discrete Fourier transform with the odd number of points (step S32).
- the frequency shift estimation circuit 2304 estimates residual frequency shift information using the frequency channel digital signal output from the odd offset discrete Fourier transform circuit 2302, and outputs it to the offset Fourier transform circuit 2302.
- the odd offset discrete Fourier transform circuit 2302 records the estimated frequency shift information in the internal storage unit (step S33).
- the frequency shift estimation circuit 2304 outputs the digital signal to the decoding circuit 2303.
- the decoding circuit 2303 decodes the signal of each frequency channel corresponding to the obtained signal (step S34).
- the frequency shift information recorded in step S34 is derived and updated by arithmetic processing repeatedly performed by the frequency shift estimation circuit 2304. In the reception method according to the above processing procedure, a desired received signal can be decoded even when an ADC with a low sampling clock frequency is used.
- FIG. 8A shows an example in which the offset discrete Fourier transform circuit 2302 performs a three-point offset discrete Fourier transform on two frequency channels as shown in FIG.
- FIG. 2 shows that in the case of performing a three-point offset discrete Fourier transform on two frequency channels, it is possible to decode a signal with 1.5 times oversampling.
- the ADC clock can be set 25% lower, and offset discrete Fourier transform can be used particularly effectively.
- the offset discrete Fourier transform is particularly effective when performing odd-numbered offset discrete Fourier transform that can be centered on the signal position.
- FIG. 8B shows a case where four frequency channels are received and 7-point offset discrete Fourier transform is performed. In this case, the sampling frequency of the ADC used for conversion of the received signal is 1.75 times the bandwidth of one frequency channel, and the ADC clock can be set 12.5% lower.
- FIG. 8C shows a case where four frequency channels are similarly received and a 5-point offset discrete Fourier transform is performed. In this case, the sampling frequency of the ADC used for conversion of the received signal is 1.25 times the bandwidth of one frequency channel.
- the ADC clock can be set to 37.5% lower, but since the oversampling is low, there is a possibility that the characteristics will deteriorate.
- 7-point offset discrete Fourier transform it is possible to receive 6 frequency channels, and it is also possible to perform 7-point, 9-point, and 11-point offset discrete Fourier transform.
- k can be obtained from 0 to (n ⁇ 1).
- M operations in which signals are input are performed, and Z that does not correspond to the signal domain
- the calculation load can be reduced by not performing k calculations.
- n 3 offset discrete Fourier transform can be performed, and only Z 0 and Z 2 can be output to the decoding circuit, and Z 1 can not be computed.
- the transmission position is shifted by shifting the conversion position from the position of Fourier transform. Proper separation is possible, and the ADC operating clock can be set low.
- the frequency is “0”, that is, if no frequency channel is provided for the DC component and the DC component is set to be the boundary of the frequency channel, noise corresponding to the DC component or Signal quality degradation due to interference can be reduced.
- FIG. 9 is a diagram illustrating a reception result of a signal transmitted with multiplexed polarization.
- the vertical axis represents the Q factor (factor) (dB (decibel)), and the horizontal axis represents the optical signal-to-noise ratio (OSNR) (dB).
- the reception result shown in this figure is a received signal in communication using 111 Gb / s (Gigabit / second) QPSK (Quadrature Phase Shift Keying) -2 carrier OFDM (Orthogonal Frequency Division Multiplexing) signal transmitted with multiplexed polarization. Is measured.
- the transmission apparatus transmits a signal obtained by multiplexing two 13.5 Gbaud QPSK signals in the frequency direction for the 111 Gb / s signal and further multiplexing the two polarization signals.
- the receiving apparatus In order to receive a signal of 2 carriers ⁇ 13.5 Gbaud in the receiving apparatus, the receiving apparatus generally has an ADC of 55.5 GS / s (gigasample / second), extracts the corresponding frequency channel, and performs decoding. There is a need. On the other hand, when the receiving apparatus according to the present embodiment is used, decoding can be performed with a 41.6 GS / s ADC.
- FIG. 7 shows a result (graph 2Sb) obtained by re-sampling a signal received at 50 GS / s to 55.5 GS / s as a result of the 4-point offset discrete Fourier transform (graph 2Sb).
- a digital signal is acquired from the received signal multiplexed in the frequency domain.
- the offset discrete Fourier transform step performs an offset discrete Fourier transform with an odd number of discrete points based on the obtained digital signal.
- the decoding step performs decoding on the frequency domain digital signal of the frequency domain obtained by the offset discrete Fourier transform, and the frequency domain digital signal of one or more frequency channels. Accordingly, when the odd-offset discrete Fourier transform circuit 2102 receives a small number of frequency channels, even if the received signal is converted into a frequency domain signal using the Fourier transform, the frequency channel corresponding to the transmitted signal is appropriately set. You can choose. Further, the digital signal acquisition circuit 2101 can reduce the sampling frequency of the ADC for acquiring the signal position using Fourier transform.
- this makes it possible to prevent the DC component interference power and noise from being detected as a frequency channel signal by using the offset discrete Fourier transform with the frequency channel boundary set to a specific frequency of frequency “0”. Can do.
- frequency deviation compensation is performed on the frequency of the obtained digital signal.
- the offset discrete Fourier transform step an offset discrete Fourier transform with an odd number of discrete points is performed on the digital signal frequency-transformed to a frequency in which the frequency shift is compensated.
- the odd-offset discrete Fourier transform circuit 2102 can perform offset discrete Fourier transform on the digital signal compensated for the frequency shift, so that the accuracy of the offset discrete Fourier transform can be increased.
- the frequency deviation estimation step of this embodiment the residual frequency deviation of each frequency channel or the frequency deviation common to all frequency channels is estimated from the digital signal converted into the frequency domain by the offset discrete Fourier transform step, and the frequency deviation is estimated.
- the frequency shift information shown is updated.
- frequency deviation compensation step frequency deviation compensation is performed on the frequency of the obtained digital signal based on the frequency deviation information.
- the frequency shift estimation circuit 2304 can estimate the frequency shift based on the calculation result and perform the offset discrete Fourier transform on the compensated digital signal based on the estimated frequency shift information, thereby further improving the accuracy of the offset discrete Fourier transform. Can be increased.
- an offset discrete Fourier transform with an odd number of discrete points is performed on the digital signal by a convolution operation, and only the frequency channel corresponding to the digital signal is calculated and output.
- the odd-offset discrete Fourier transform circuit 2102 separates necessary information without being affected by a DC component or the like by performing convolution operation on the information of the frequency channel corresponding to the transmitted signal. Can do.
- the number of discrete points of the offset discrete Fourier transform is set to 3, and two frequency channels are acquired.
- the odd-offset discrete Fourier transform circuit 2102 can aggregate unnecessary information such as a direct current component in one frequency channel, and can improve the quality of information of the two frequency channels to be acquired.
- the ADC can convert it into a digital signal at a sampling clock frequency that is not more than twice the frequency band.
- the offset discrete Fourier transform step performs independent frequency shift compensation for each frequency channel.
- necessary information can be converted at a frequency suitable for each frequency channel by compensating for the frequency shift detected independently for each frequency channel.
- FIG. 10 is a block diagram showing the configuration of the receiving apparatus according to the A1 embodiment of the present invention.
- the receiving apparatus 110 shown in this figure includes a digital signal acquisition circuit 1101, a branch circuit 1102, an equalization circuit 1103, a decoding circuit 1104, and an initial coefficient storage circuit 1105.
- the digital signal acquisition circuit 1101 converts the received analog signal into a digital signal.
- the branch circuit 1102 is a frequency channel included in the received signal received by the digital signal acquisition circuit 1101 and duplicates the digital signal converted by the digital signal acquisition circuit 1101 to the number of channels (L series) of frequency channels to be decoded. Then branch.
- the equalization circuit 1103 equalizes each of the L series received signals branched by the branch circuit 1102.
- the least square error method (MMSE), the maximum SNR method (MSN), the constrained output power minimization method (CMP), the constant envelope are used by using the prior knowledge of the desired signal.
- An equalization algorithm such as a line signal algorithm (CMA) can be used (see Non-Patent Document 4).
- the equalization for the signal of the kth frequency channel can be expressed by the following equation (A1) using the equalization weight w k, i .
- Formula (A1) In Expression (A1), S k, i is an estimated transmission signal corresponding to the i-th discrete time of the k-th frequency.
- M is the number of equalization weight taps.
- X i + nM is a reception signal at reception timing (i + nM).
- w k, i is the equalization coefficient of the equalization algorithm of the kth subcarrier.
- signals S X, k, i and S Y, k, i for two orthogonal polarizations can be obtained by the following equation (A2), respectively.
- Formula (A2) In the formula (A2), XX, i + nM and XY, i + nM are reception signals corresponding to the X polarization and Y polarization at the reception timing (i + nM), respectively.
- w XX, k, i are equalization coefficients of the equalization algorithm used for the received signal of the X polarization in order to obtain the signal transmitted by the X polarization of the kth subcarrier.
- w YX, k, i are equalization coefficients of the equalization algorithm used for the received signal of Y polarization in order to obtain a signal transmitted by X polarization of the kth subcarrier.
- w YY, k, i and w XY, k, i are the equalization coefficient and Y of the equalization algorithm used for the X polarization received signal to obtain the signal transmitted with the Y polarization of the kth subcarrier. This is an equalization coefficient of an equalization algorithm used for a polarization reception signal.
- the initial coefficient storage circuit 1105 stores a coefficient having a low correlation as an initial weight so that equalization is performed with different weights for each signal series.
- the equalization circuit can equalize signals corresponding to different frequency bands from the branched signal sequences.
- a coefficient for extracting a corresponding frequency channel by discrete Fourier transform can be selected.
- the discrete Fourier transform is given by the following equation. When the discrete Fourier transform is performed on the n complex number sequences X 0 ,..., X n ⁇ 1 , the signal Y k of the k-th frequency channel is expressed by the equation (A3).
- d is 0.5 or ⁇ 0.5 when the DC component is at the center of the two frequency channels.
- a k, i is an equalization coefficient in the discrete Fourier transform, and this value can be used as an initial value of the equalization algorithm. Further, in consideration of the characteristics of the apparatus and propagation path information, the initial coefficient can be set to the initial coefficient by calculating the coefficient of the discrete Fourier transform.
- the equalization weight w k [w k, 1 , w k, 2 ,..., w k, M ] T with the number of taps used for the k-th frequency channel is [0, 0,. , a k, 0 , a k, 1 ,..., a k, n ⁇ 1 , 0, 0,... 0] T.
- the superscript T is an operator for transposing.
- any number of a k, 0 , a k, 1 ,..., A k, n ⁇ 1 can be replaced with “0” in order to reduce the initial weight constraint condition.
- the equalization circuit 1103 can equalize the signal of each frequency channel.
- a relationship that satisfies the orthogonal condition can be given between the polarized waves.
- the initial values of w XX, k and w YX, k are [0, 0,..., 0, a k, 0 , a k, 1 ,.
- orthogonalization can be set while making equalization weights corresponding to the same frequency channel.
- w XX, k [0, 0,..., 0, b xx a k, 0 , b xx a k, 1 ,..., b xx a k, n-1 , 0, 0,... 0] T
- w YX, k [0, 0,..., 0, b yx a k, 0 , b yx a k, 1 ,..., b yx a k, n-1 , 0, 0,... 0] T
- w XY , k [0, 0,..., 0, b xy a k, 0 , b xy a k, 1 ,..., b xy a k, n-1 , 0, 0,... 0] T
- w YY, k [0, 0,..., 0, b yy a k, 0 , b yy a k, 1 ,..., b yy a
- the decoding circuit 1104 decodes the signal equalized by the equalization circuit 1103 in accordance with the encoding method / modulation method. However, the compensation performance with respect to the time spread of the signal originally possessed by the equalizer is degraded by the number of frequency channels n. That is, the receiving apparatus 10 consumes equalization performance with respect to time spread of the received signal for Fourier transform by the number of frequency channels n. This eliminates the need for frequency deviation estimation and compensation for Fourier transform in the receiving apparatus 110.
- the branch circuit 1102 can shift and output the frequency for each frequency channel. In this case, the k th received signal branched for decoding the k th frequency channel is output from the branch circuit 102 after performing the following calculation.
- X ′ k is a signal output from the branch circuit.
- Fs is a sampling frequency.
- F k is the center frequency of the kth frequency channel in the receiver.
- the center frequency of the k-th frequency channel used here does not need to be accurate, and an approximate value obtained by storing in advance or roughly estimating the center position of each frequency channel can be used.
- the equalization circuit can set an equalization weight for the center frequency channel for the frequency-shifted signal, and the initial value of the equalization weight is [0, 0, ..., 0, 1, 1, ... 1, 0, 0,..., 0] can be used.
- the number of “1” is n
- the number of “0” is Nz.
- any number of “1” s can be converted to “0” in order to reduce the constraint condition.
- FIG. 11 is a block diagram showing the configuration of the receiving apparatus according to the A2 embodiment of the present invention.
- the receiving apparatus 120 shown in this figure includes a photoelectric conversion circuit 1200, a digital signal acquisition circuit 1201, a branch circuit 1202, an equalization circuit 1203, a decoding circuit 1204, and an initial coefficient storage circuit 1205.
- the receiving device 120 receives an optical signal.
- the photoelectric conversion circuit 1200 converts the received optical signal into an electrical signal.
- the digital signal acquisition circuit 1201 converts the electrical signal converted by the photoelectric conversion circuit 1200 into a digital signal.
- the branch circuit 1202 is a frequency channel included in the reception signal converted by the digital signal acquisition circuit 1201, and duplicates the digital signal converted by the digital signal acquisition circuit 1201 to the number of channels (L series) of frequency channels to be decoded. Then branch.
- the initial coefficient storage circuit 1205 stores a coefficient that can be expressed using a discrete Fourier transform coefficient represented by the formula (A3), a coefficient having a high correlation with the discrete Fourier transform coefficient, or a part thereof.
- the equalization circuit 1203 equalizes the L-sequence signal branched by the branch circuit 1202 using the initial weight stored in the initial coefficient storage circuit 1205.
- the prior knowledge of the desired signal is used, so that the least square error method (MMSE), the maximum SNR method (MSN), the constrained output power minimization method (CMP), the constant envelope
- MMSE least square error method
- MSN maximum SNR method
- CMP constrained output power minimization method
- CMA line signal algorithm
- a constraint condition is given so as to obtain an equalization coefficient having a low correlation between different frequency channels, or a constraint condition is set so that the correlation of the output signal becomes low.
- the decoding circuit 1204 decodes the signal of the specific frequency channel extracted by the equalization circuit 1203.
- the branch circuit 1202 can shift and output the frequency of the signal after branching similarly to the equation (A4). Since the maximum value of the frequency shift in optical communication may be larger than the bandwidth of the frequency channel, the branch circuit 1202 performs a rough estimation on the frequency shift from the known signal included in the received signal and the spectrum distribution, It is possible to improve the transmission quality of the signal after branching by shifting the frequency using the formula (A4) for all sequences.
- FIG. 12 is a diagram illustrating a reception result of a signal transmitted with multiplexed polarization.
- the vertical axis indicates the Q factor (dB)
- the horizontal axis indicates the optical signal-to-noise ratio (OSNR) (dB).
- the reception result shown in this figure is a received signal in communication using 111 Gb / s (Gigabit / second) QPSK (Quadrature Phase Shift Keying) -2 carrier OFDM (Orthogonal Frequency Division Multiplexing) signal transmitted with multiplexed polarization. Is measured.
- the transmission apparatus transmits a signal obtained by multiplexing two 13.5 Gbaud (Gigabaud) QPSK signals in the frequency direction for the 111 Gb / s signal and two for the polarization.
- the signal received at 50 GS / s (gigasample / second) is converted to 55.5 GS / s by offline processing, and decoding according to this embodiment is performed.
- the graph curve 1Sa shows the result of this embodiment.
- Graph 1Sb compares and shows the result by the system which decodes after performing the conventional Fourier transform.
- FIG. 12 shows that the same characteristics are obtained when the Fourier transform is performed (graph 1Sb) and when equalization is performed without performing the graph (graph 1Sa).
- graph 1Sb the Fourier transform and the frequency shift estimation are not performed in the preceding stage, the characteristics equivalent to the result obtained by performing the discrete Fourier transform are obtained. Can be confirmed.
- the greater the frequency deviation the greater the effect.
- the digital signal acquisition circuit 1101 acquires a digital signal from the received signal multiplexed in the frequency domain.
- the branch circuit 1102 branches the obtained digital signal into the number of frequency channels to be decoded.
- the initial coefficient storage circuit 1105 stores a different coefficient having a low correlation as an initial weight for each branched signal sequence.
- the equalization circuit 1103 equalizes each branched signal series with a different coefficient.
- the decoding circuit 1104 performs decoding on each equalized signal sequence. Thereby, the signal transmitted by the orthogonal frequency division multiplexing method can be decoded without performing the discrete Fourier transform.
- the decoding circuit 1104 can eliminate the need for frequency deviation estimation and compensation of discrete Fourier transform.
- the digital signal acquisition circuit 1101 acquires a digital signal from the received signal multiplexed in the frequency domain.
- the branch circuit 1102 branches the obtained digital signal into the number of frequency channels to be decoded.
- the initial coefficient storage circuit 1105 stores, as an initial weight, a coefficient having a high correlation with a coefficient of a discrete Fourier transform corresponding to each frequency channel, or a coefficient including at least a part of each branched signal series.
- the equalization circuit 1103 performs equalization using the coefficients stored in the initial coefficient storage step as initial weights.
- the decoding circuit 1104 performs decoding on each equalized signal sequence.
- the equalization circuit 1103 performs blind equalization using the coefficient of the discrete Fourier transform for acquiring the center frequency of each frequency channel as an initial value without performing the discrete Fourier transform on the frequency channel to be equalized.
- a signal transmitted by the orthogonal frequency division multiplexing method can be decoded.
- the branch circuit 1102 branches a digital signal
- the branch is performed after performing frequency conversion so that the center frequency of each frequency channel becomes a DC component
- the initial coefficient storage circuit 1105. Stores a coefficient having a high correlation with a coefficient of the discrete Fourier transform corresponding to the specific frequency component in the discrete Fourier transform, or a coefficient including at least a part thereof as an initial weight.
- the frequency of the branched signal is converted and the frequency shift can be corrected, so that the reception quality of the equalization processing and decoding processing can be ensured, so that the signal is transmitted by orthogonal frequency division multiplexing without performing discrete Fourier transform.
- the received signal can be decoded.
- the equalization coefficient is adjusted so that the signal sequences equalized in the equalization circuit 1103 do not converge to signal sequences indicating the same signal.
- the present invention can be applied to a frequency domain multiplexed signal receiving method and a frequency domain multiplexed signal receiving apparatus. According to these frequency domain multiplexed signal receiving methods and frequency domain multiplexed signal receiving apparatuses, sampling can be performed at a frequency lower than twice the frequency band of the received signal.
Abstract
Description
また、本発明は、周波数領域に多重された受信信号を、離散フーリエ変換を行わずに等化する周波数領域多重信号受信方法及び周波数領域多重信号受信装置に関する。
本願は、2009年7月17日に日本に出願された特願2009-169460号、及び、2009年7月17日に日本に出願された特願2009-169455号に基づき優先権を主張し、その内容をここに援用する。
OFDMでは、受信信号を離散化するアナログデジタル変換器(ADC:Analog Digital Converter)におけるサンプリング周波数が、受信信号のBaud rate(ボーレート)の2倍の周波数より低い周波数に設定できる利点がある。このため、例えば、IEEE802.11aに適用される無線LAN(Local Area Network)では、信号帯域20MHz(メガヘルツ)のうち、16.6MHz程度の帯域をデータ信号として用いることが知られている。
さらに、OFDMにおいて、周波数チャネル数を増やすほど、ガードインターバルの設定が容易となり、各周波数チャネルにおいてフラットフェージングを仮定できる。このため、実用されている無線システムでは64~1024程度の周波数チャネル数が選ばれている。また、一般にOFDMシステムでは、受信装置における周波数が「0」、つまり直流成分に対応する周波数チャネルを有する。この直流成分に対応する周波数チャネルでは、信号間干渉や雑音の影響で特性が劣化する。そのため、これらに対応する周波数チャネルは、一般に用いられていない。
受信装置において、送信信号のPeak to average power ratio (PAPR)が通信品質に大きな影響をもたらす場合などは、多数の周波数チャネル数を設定できない。又は、フィルタを用いてOFDMの一部を分離した場合においても、受信信号の周波数帯域に含まれるOFDMの周波数チャネル数は、フィルタの影響によって少なくなる。
このような場合において、ガードインターバルを用いずとも、受信側で各周波数チャネルに対し、等化処理を行うことで復号することも可能である。ただし、サブキャリア数が少なくなるため、IEEE802.11aのように直流成分を中心とする周波数チャネルを有し、かつこの周波数チャネルが雑音や信号間干渉の影響を受けるためデータ信号の伝送に用いないと、通信速度における損失が大きくなる。
送信信号を生成した送信装置と、送信された信号を受信信号として受信を行う受信装置は、それぞれ異なる基準信号発生装置に接続される。それぞれの基準信号発生装置の周波数には、通常周波数ずれが存在する。そのため、受信装置は、送信装置の周波数に同期して送信された受信信号から、フィルタで特定の周波数領域を切り出す場合や、フーリエ変換を行う場合に、受信信号との周波数ずれを精度よく補償する必要がある。受信装置では、その周波数がずれると、隣接周波数チャネルの信号がもれ込むことにより、信号品質が劣化する。特に同期検波を行う光通信においては、送信に用いるレーザ光と、受信に用いるレーザ光の間に生じる周波数ずれが大きく、このような問題が生じやすい。
本発明の実施形態の第1の目的は、受信信号の周波数帯域の2倍の周波数より低い周波数で標本化できる周波数領域多重信号受信方法と周波数領域多重信号受信装置を提供することである。
本発明の実施形態の第2の目的は、フーリエ変換を独立して行うことなく、復号を行う周波数領域多重信号受信方法及び周波数領域多重信号受信装置を提供することである。
前記各周波数チャネルに独立の周波数ずれの補償を行ってもよい。
これにより、少ない周波数チャネルを受信する場合において、フーリエ変換を用いて受信信号を周波数領域の信号に変換しても、送信した信号に対応する周波数チャネルを適切に選択できる。また、フーリエ変換を用いて信号位置を取得するためのADCのサンプリング周波数を低下させることができる。
これにより、離散フーリエ変換を行うことなく、直交周波数分割多重方式で送信された信号を復号することができる。
これにより、周波数領域に多重された受信信号を、離散フーリエ変換を行うことなく等化することで、復号する周波数チャネルの独立した離散フーリエ変換部を省き、受信装置構成を簡易化することが可能となる。
図1は、本発明のB1実施形態による受信装置の構成を示す構成図である。
この図に示される、受信装置210は、デジタル信号取得回路2101、奇数オフセット離散フーリエ変換回路2102、及び、復号回路2103を備える。
デジタル信号取得回路2101は、受信した信号(アナログ信号)をデジタル信号に変換する。
奇数オフセット離散フーリエ変換回路2102は、受信したアナログ信号に応じて変換されたデジタル信号に、後述するオフセット離散フーリエ変換を行い、受信した信号に対応する周波数チャネルを出力する。復号回路2103は、奇数オフセット離散フーリエ変換回路2102によって変換された各周波数チャネルの受信信号から、送信された信号の復号を行う。
n個の複素数列X0, …, Xn-1に対して離散フーリエ変換を行うと、n個の複素数列Y0, …, Yn-1が得られる。k番目の周波数チャネルの信号Ykを、式(B1)に示す。
図2は、信号の中心周波数の基準信号で同期検波した際の信号の強度分布を示す模式図である。図2において、符号Qは、本実施形態における3つの周波数チャネルが対応する周波数帯域を示している。
例えば、5Gbaud(ギガボー)の信号からなる、2つの周波数帯に対応する2キャリアOFDM信号を受信する場合を仮定する。隣接する周波数チャネルの間隔Iは、5GHz(ギガヘルツ)になる。
図2は、ベースバンドにおける信号のスペクトラムを表しているため、直流成分(周波数0)以下のマイナスの周波数が定義されている。ADCが15GHzである場合は、-7.5GHz~0GHzは、7.5GHz~15GHzと等価である。
一般に、OFDMで周波数領域に多重される周波数チャネルの数は偶数であり(非特許文献2参照)、1又は偶数個の周波数チャネルごとに扱う方が、データの信号処理を行いやすい。このため、偶数の周波数チャネルの信号を、低いサンプリングレートのADCで取得するためには、離散フーリエ変換のポイント数を奇数とすることが有効となる。この場合には、奇数オフセット離散フーリエ変換回路2102は、中央の偶数個の周波数チャネルの結果を、復号回路2103に出力する。
ブロック毎に行う場合は、受信系列Xkに対し、(Xd, … , Xd+(n-1)), (Xd+G, … , Xd+(G+n-1)), (Xd+2G, … , Xd+(2G+n-1))… とGシンボル毎にオフセット離散フーリエ変換を行う。ここでdは信号の先頭位置である。畳み込み演算を行う場合には、式(B2)においてXiに乗算している係数を受信信号に畳み込み演算することができる。また、G=1としてブロック演算を行っても同様である。
図3は、本実施形態における処理手順を示すフローチャートである。
受信装置210は、受信信号を受信すると、デジタル信号取得回路2101は、受信した受信信号に応じてデジタル信号に変換する(ステップS11)。奇数オフセット離散フーリエ変換回路2102は、奇数ポイント数のオフセット離散フーリエ変換を行う(ステップS12)。
復号回路2103は、得られた信号に対応する各周波数チャネルの信号を復号する(ステップS13)。
以上の処理手順にしたがった受信方法では、サンプリングクロック周波数を周波数帯域の2倍の周波数より低下させたADCを用いた場合でも、所望の受信信号を復号することが可能になる。上記ステップにおいて、奇数オフセット離散フーリエ変換回路2102は、変換されたデジタル信号の周波数ずれを示す周波数ずれ情報に応じて周波数ずれを補償することもできる。また、復号回路2103は、推定された周波数ずれ信号を、周波数ずれ補償回路2102にフィードバックすることもできる。
図4は、本発明の第B2実施形態による受信装置の構成を示す構成図である。
この図に示される、受信装置220は、デジタル信号取得回路2201、奇数オフセット離散フーリエ変換回路2202、復号回路2203、及び、周波数ずれ補償回路2204を備える。
デジタル信号取得回路2201は、受信した信号(アナログ信号)をデジタル信号に変換する。
周波数ずれ補償回路2204は、送信信号に挿入された検定信号に対応する受信信号、送信された信号の変調方式の特徴、他受信回路ブロックから入力された周波数ずれ情報、のいずれかを用いて受信信号に生じている周波数ずれを補償する。受信信号に生じている周波数ずれは、送信信号を生成した周波数と、受信装置220が基準とする周波数との周波数偏差を示す。
奇数オフセット離散フーリエ変換回路2202は、受信したアナログ信号に応じて変換され、周波数ずれ補償回路2204が周波数ずれを補償したデジタル信号に、後述するオフセット離散フーリエ変換を行い、受信した信号に対応する周波数チャネルを出力する。
復号回路2203は、奇数オフセット離散フーリエ変換回路2202によって変換された各周波数チャネルの受信信号から、送信された信号の復号を行う。
周波数ずれ補償回路2204において、式(B4)に示す周波数変換を予め行うこともできる。
Δfとして、復号回路2203において推定された周波数ずれ信号をフィードバックしてもよい。
図5は、本実施形態における処理手順を示すフローチャートである。
受信装置220は、受信信号を受信すると、デジタル信号取得回路2201は、受信した受信信号に応じてデジタル信号に変換する(ステップS21)。周波数ずれ補償回路2204は、変換されたデジタル信号の周波数ずれを示す周波数ずれ情報に応じて周波数ずれを補償する(ステップS22)。奇数オフセット離散フーリエ変換回路2202は、奇数ポイント数のオフセット離散フーリエ変換を行う(ステップS23)。
復号回路2203は、得られた信号に対応する各周波数チャネルの信号を復号する。また、復号回路2203は、推定された周波数ずれ信号を、周波数ずれ補償回路2204にフィードバックする(ステップS24)。
ステップS24で記録された周波数ずれ情報は、繰り返し行われる演算処理によって導かれ、周波数ずれ補償回路2204における参照情報として利用される。
以上の処理手順にしたがった受信方法では、サンプリングクロック周波数を周波数帯域の2倍の周波数より低下させたADCを用いた場合でも、所望の受信信号を復号することが可能になる。
図6は、本発明の第B3実施形態による受信装置の構成を示す構成図である。
この図に示される、受信装置230は、デジタル信号取得回路2301、奇数オフセット離散フーリエ変換回路2302、復号回路2303、及び、周波数ずれ推定回路2304を備える。
デジタル信号取得回路2301は、受信した信号(アナログ信号)をデジタル信号に変換する。周波数ずれ推定回路2304は、送信信号に挿入された検定信号に対応する受信信号、送信された信号の変調方式の特徴、他受信回路ブロックから入力された周波数ずれ情報、のいずれかを用いて受信信号に生じている周波数ずれを補償する。受信信号に生じている周波数ずれは、送信信号を生成した周波数と、受信装置230が基準とする周波数との周波数偏差を示す。
奇数オフセット離散フーリエ変換回路2302は、受信したアナログ信号に応じて変換され、周波数ずれを考慮した奇数オフセット離散フーリエ変換を行い、受信した信号に対応する周波数チャネルを出力する。
周波数ずれ推定回路2304は、奇数オフセット離散フーリエ変換回路2302から出力された周波数チャネルのデジタル信号を用いて、残留する周波数ずれ情報を推定し、オフセットフーリエ変換回路2302に出力する。また、周波数ずれ推定回路2304は、デジタル信号を復号回路2303へ出力する。
復号回路2303は、周波数ずれ推定回路2304が出力したデジタル信号であって、奇数オフセット離散フーリエ変換回路2302によって変換された各周波数チャネルの受信信号から、送信された信号の復号を行う。
又は、このように周波数ずれを補償する機能を奇数オフセット離散フーリエ変換回路が有することで、各周波数チャネルにおける残留周波数ずれが個別に補償できる。よって、周波数ずれ推定回路2304は、各周波数チャネルに対して周波数ずれを推定し、周波数ずれ情報をそれぞれ奇数オフセット離散フーリエ変換回路2302に出力し、各周波数チャネルにおいて周波数ずれを補償することもできる。また、復号回路2303は、各周波数チャネルの周波数ずれを推定し、奇数オフセット離散フーリエ変換回路2302に出力することもできる。この場合式(B5)におけるΔfは各周波数チャネルに異なり、Δfkをk番目の周波数チャネルの周波数ずれとして用いることができる。
図7は、本実施形態における処理手順を示すフローチャートである。
受信装置230は、受信信号を受信すると、デジタル信号取得回路2301は、受信した受信信号に応じてデジタル信号に変換する(ステップS31)。奇数オフセット離散フーリエ変換回路2302は、変換されたデジタル信号の周波数ずれを示す周波数ずれ情報に応じて周波数ずれを補償し、奇数ポイント数のオフセット離散フーリエ変換を行う(ステップS32)。
周波数ずれ推定回路2304は、奇数オフセット離散フーリエ変換回路2302から出力された周波数チャネルのデジタル信号を用いて、残留する周波数ずれ情報を推定し、オフセットフーリエ変換回路2302に出力する。奇数オフセット離散フーリエ変換回路2302は、推定された周波数ずれ情報を内部の記憶部に記録する(ステップS33)。
周波数ずれ推定回路2304は、デジタル信号を復号回路2303へ出力する。
復号回路2303は、得られた信号に対応する各周波数チャネルの信号を復号する(ステップS34)。
ステップS34で記録された周波数ずれ情報は、周波数ずれ推定回路2304によって繰り返し行われる演算処理によって導かれ、更新される。
以上の処理手順にしたがった受信方法では、サンプリングクロック周波数が低いADCを用いた場合でも、所望の受信信号を復号することが可能になる。
図8A~8Cは、周波数チャネルの配置とオフセット離散フーリエ変換のポイント数を示す図である。図8Aは、オフセット離散フーリエ変換回路2302が、図2に示したように2つの周波数チャネルを3ポイントのオフセット離散フーリエ変換を行う場合の例を示す。
特に図2に示したように2つの周波数チャネルを3ポイントのオフセット離散フーリエ変換を行う場合では、1.5倍のオーバサンプルで信号を復号できるため、2倍のオーバサンプルを必要とする場合に比べ、ADCのクロックを25%低く設定することができ、特に有効にオフセット離散フーリエ変換を活用できる。
図8Bでは、4つの周波数チャネルを受信し、7ポイントのオフセット離散フーリエ変換を行う場合を示す。その場合の受信信号の変換に用いるADCのサンプリング周波数は、1つの周波数チャネルの帯域幅の1.75倍となり、ADCのクロックを12.5%低く設定できる。
図8Cでは、同じく4つの周波数チャネルを受信し、5ポイントのオフセット離散フーリエ変換を行う場合を示す。その場合の受信信号の変換に用いるADCのサンプリング周波数は、1つの周波数チャネルの帯域幅の1.25倍となる。この場合、ADCのクロックを37.5%低く設定できるが、オーバサンプルが低いため、特性の劣化が生じる可能性がある。
また、7ポイントのオフセット離散フーリエ変換を行う場合では、6つの周波数チャネルを受信することができ、さらに、7ポイント、9ポイント、11ポイントのオフセット離散フーリエ変換を行うこともできる。
以上に示した本発明の実施形態によれば、少数の周波数チャネルを含んだ受信信号を、フーリエ変換を用いて周波数変換する際に、変換位置をフーリエ変換の位置とずらすことで、送信信号を適切に分離することができ、ADCの動作クロックを低く設定することができる。
図9は、偏波を多重して送信した信号の受信結果を示す図である。
この図の縦軸は、Qファクター(factor)(dB(デシベル))を示し、横軸は、光信号対雑音比(OSNR)(dB)を示す。
この図に示す受信結果は、偏波を多重して送信した111Gb/s(ギガビット/秒)のQPSK(Quadrature Phase Shift Keying)-2キャリアOFDM(Orthogonal Frequency Division Multiplexing)信号を用いた通信における受信信号を測定したものである。
図7において、4ポイントオフセット離散フーリエ変換の結果として、50GS/sで受信した信号を55.5GS/sにリサンプルして復号した結果(グラフ2Sb)を示し、それに対し本実施形態による結果として、46.1GS/sの受信信号に、3ポイントオフセット離散フーリエ変換を用いた場合のQ-factorをOSNRに対して算出した結果(グラフ2Sa)を示す。図に示されるように、本実施形態(グラフ2Sa)では、ADCのサンプリングクロックの周波数を大きく低下させているにもかかわらず、55.5GS/sのデータを用いた4ポイントオフセット離散フーリエ変換の結果(グラフ2Sb)とほぼ同等の特性が得られていることが確認できる。
これにより、奇数オフセット離散フーリエ変換回路2102は、少ない周波数チャネルを受信する場合において、フーリエ変換を用いて受信信号を周波数領域の信号に変換しても、送信した信号に対応する周波数チャネルを適切に選択できる。また、デジタル信号取得回路2101は、フーリエ変換を用いて信号位置を取得するためのADCのサンプリング周波数を低下させることができる。
これにより、奇数オフセット離散フーリエ変換回路2102は、周波数ずれを補償したデジタル信号をオフセット離散フーリエ変換できることから、オフセット離散フーリエ変換の精度を高めることができる。
これにより、周波数ずれ推定回路2304は、演算結果に基づいて周波数ずれを推定し、推定された周波数ずれ情報に基づいて補償したデジタル信号をオフセット離散フーリエ変換できることから、オフセット離散フーリエ変換の精度をさらに高めることができる。
これにより、奇数オフセット離散フーリエ変換回路2102は、送信された信号に対応する周波数チャネルの情報を、畳み込み演算を用いて行うことにより、直流成分などに影響されることなく必要な情報を分離することができる。
これにより、奇数オフセット離散フーリエ変換回路2102は、1つの周波数チャネルに直流成分などの不要な情報を集約することができ、取得する2つの周波数チャネルの情報の品質を上げることができる。また、周波数帯域の2倍の周波数以下のサンプリングクロック周波数によってADCがデジタル信号に変換することができる。
これにより、各周波数チャネルに独立して検出された周波数ずれを補償することにより、それぞれの周波数チャネルに適した周波数で必要な情報を変換することができる。
図10は、本発明のA1実施形態による受信装置の構成を示す構成図である。
この図に示される、受信装置110は、デジタル信号取得回路1101、分岐回路1102、等化回路1103、復号回路1104、及び、初期係数記憶回路1105を備える。
デジタル信号取得回路1101は、受信したアナログ信号をデジタル信号に変換する。
分岐回路1102は、デジタル信号取得回路1101が受信した受信信号に含まれる周波数チャネルであって、復号を行う周波数チャネルのチャネル数(L系列)に、デジタル信号取得回路1101が変換したデジタル信号を複製し分岐する。
k番目の周波数チャネルの信号に対する等化は、等化ウエイトwk,iを用いて、以下の式(A1)で表すことができる。
式(A1)
式(A1)において、Sk,iはk番目の周波数のi番目の離散時間に対応する推定された送信信号である。Mは等化ウエイトのタップ数である。Xi+n-Mは受信タイミング(i+n-M)の受信信号である。wk,iはk番目のサブキャリアの等化アルゴリズムの等化係数である。送信において偏波多重を行った場合には、2つの直交する偏波に対する信号SX,k,iとSY,k,iはそれぞれ以下の式(A2)で得ることができる。
式(A2)
式(A2)において、XX,i+n-MとXY,i+n-Mはそれぞれ受信タイミング(i+n-M)のX偏波とY偏波に対応する受信信号である。wXX,k,iはk番目のサブキャリアのX偏波で送信した信号を得るためにX偏波の受信信号に用いる等化アルゴリズムの等化係数である。wYX,k,iはk番目のサブキャリアのX偏波で送信した信号を得るためにY偏波の受信信号に用いる等化アルゴリズムの等化係数である。同様にwYY,k,iとwXY,k,iはk番目のサブキャリアのY偏波で送信した信号を得るためにX偏波の受信信号に用いる等化アルゴリズムの等化係数とY偏波の受信信号に用いる等化アルゴリズムの等化係数である。
初期ウエイトとしては、離散フーリエ変換で対応する周波数チャネルを抜き出すための係数を選ぶことができる。離散フーリエ変換は、以下の式で与えられる。n個の複素数列X0,…,Xn-1に対して離散フーリエ変換を行うと、k番目の周波数チャネルの信号Ykは、式(A3)に示される。
偏波多重を用いて、送信においてX偏波とY偏波に異なる信号を送信する場合の初期ウエイトとしては、wXX,k = [wXX,k,1, wXX,k,2, …, wXX,k,M]T とwYX,k= [wYX,k,1, wYX,k,2, …, wYX,k,M]Tの初期値として、[0, 0, …, 0, ak,0, ak,1, …, ak,n-1, 0, 0, …0]Tと[0, 0, ..., 0]Tを設定し、wXY,k = [wXY,k,1, wXY,k,2, …, wXY,k,M]T とwYY,k= [wYY,k,1, wYY,k,2, …, wYY,k,M]Tを [0, 0, ..., 0]Tと[0, 0, …, 0, ak,0, ak,1, …, ak,n-1, 0, 0, …0] Tとなるように設定することで同じ周波数チャネルに対応する等化ウエイトとしながら、直交するように設定できる。または、直交条件を満たすような関係を偏波間に与えることもできる。例えば2×2のウォルシュの直交符号を用いる場合には、wXX,k とwYX,kの初期値として、[0, 0, …, 0, ak,0, ak,1, …, ak,n-1, 0, 0, …0]Tと[0, 0, …, 0, ak,0, ak,1, …, ak,n-1, 0, 0, …0] Tを設定し、wXY,kとwYY,k Tを [0, 0, …, 0, ak,0, ak,1, …, ak,n-1, 0, 0, …0] Tと[0, 0, …, 0, -ak,0, -ak,1, …, -ak,n-1, 0, 0, …0] Tとなるように設定することで同じ周波数チャネルに対応する等化ウエイトとしながら、直交するように設定できる。
つまり、wXX,k= [0, 0, …, 0, bxxak,0, bxxak,1, …, bxxak,n-1, 0, 0, …0]T、wYX,k = [0, 0, …, 0, byxak,0, byxak,1, …, byxak,n-1, 0, 0, …0]T、wXY,k = [0, 0, …, 0, bxyak,0, bxyak,1, …, bxyak,n-1, 0, 0, …0]T、wYY,k= [0, 0, …, 0, byyak,0, byyak,1, …, byyak,n-1, 0, 0, …0]T、とした際に、bxxbxy+byxbyyが0もしくは0に近似できる小さい値になるように、bxx、byx、bxy、byyを設定することができる。また、bxx、byx、bxy、byyは各周波数チャネルで独立に設定することもできる。
または、分岐回路1102において、各周波数チャネルに対し、それぞれ周波数をシフトさせ、出力することができる。この場合には、k番目の周波数チャネルの復号用として分岐したk番目の受信信号は、以下の演算を行った上で分岐回路102から出力される。
図を参照して本発明の第A2実施形態について説明する。
図11は、本発明の第A2実施形態による受信装置の構成を示す構成図である。
この図に示される、受信装置120は、光電気変換回路1200、デジタル信号取得回路1201、分岐回路1202、等化回路1203、復号回路1204、及び、初期係数記憶回路1205を備える。この受信装置120は、光信号を受信する。
デジタル信号取得回路1201は、光電気変換回路1200が変換した電気信号をデジタル信号に変換する。
分岐回路1202は、デジタル信号取得回路1201が変換した受信信号に含まれる周波数チャネルであって、復号を行う周波数チャネルのチャネル数(L系列)に、デジタル信号取得回路1201が変換したデジタル信号を複製し分岐する。
初期係数記憶回路1205は、式(A3)で示される離散フーリエ変換係数、または離散フーリエ変換係数と高い相関を持つ係数、またはそれらの一部を用いて表せる係数を記憶する。
復号回路1204は、等化回路1203が取り出した特定の周波数チャネルの信号を復号する。同期検波を行う光通信では、特に送信装置と受信装置でのレーザ間の波長ずれが大きいため、フーリエ変換のための周波数ずれ補償機能を必要としないことで、回路規模を小さくすることができる。
分岐回路1202は、式(A4)と同様に、分岐後の信号の周波数をそれぞれシフトさせ、出力することができる。
光通信における周波数ずれの最大値は周波数チャネルの帯域幅より大きくなることも考えられるため、分岐回路1202は、受信信号に含まれる既知信号や、スペクトラムの分布から、周波数ずれについて粗推定を行い、全系列に対して式(A4)を用いて周波数をシフトさせ、分岐後の信号の伝送品質を向上することができる。
図12は、偏波を多重して送信した信号の受信結果を示す図である。
この図の縦軸は、Qファクター(factor)(dB)を示し、横軸は、光信号対雑音比(OSNR)(dB)を示す。
この図に示す受信結果は、偏波を多重して送信した111Gb/s(ギガビット/秒)のQPSK(Quadrature Phase Shift Keying)-2キャリアOFDM(Orthogonal Frequency Division Multiplexing)信号を用いた通信における受信信号を測定したものである。
本実施形態に示す方法では、周波数ずれが大きくなるほど、その効果が大きくなる。また、本実施形態に示す受信装置では、フーリエ変換のために消費するチャネルを設けることにより、偏波モード分散に対する耐力の低下が生じる場合があるが、等化のためのタップ数を多くすることにより、その耐力の低下を容易に回避することができる。
これにより、離散フーリエ変換を行うことなく、直交周波数分割多重方式で送信された信号を復号することができる。復号回路1104では、離散フーリエ変換の周波数ずれ推定と補償を不要とすることができる。
これにより、等化回路1103が、各周波数チャネルの中心周波数を取得する離散フーリエ変換の係数を初期値としてブラインドで等化を行うことにより、等化する周波数チャネルに離散フーリエ変換を行うことなく、直交周波数分割多重方式で送信された信号を復号することができる。
これにより、分岐された信号の周波数が変換され、周波数ずれを補正できることから、等化処理、復号処理の受信品質を確保できることから、離散フーリエ変換を行うことなく、直交周波数分割多重方式で送信された信号を復号することができる。
これにより、等化された信号が同じ信号となることを防いで離散フーリエ変換を行うことなく、直交周波数分割多重方式で送信された信号を復号することができる。
受信装置120についても同様の効果を得ることができる。
1101 デジタル信号取得回路
1102 分岐回路
1103 等化回路
1104 復号回路
1105 初期係数記憶回路
210 受信装置
2101 デジタル信号取得回路
2102 奇数オフセット離散フーリエ変換回路
2103 復号回路
Claims (16)
- 周波数領域に多重された受信信号を復号する周波数領域多重信号受信方法であって、
周波数領域に多重された受信信号からデジタル信号を取得するデジタル信号取得ステップと、
前記得られたデジタル信号に基づいて、奇数の離散ポイント数のオフセット離散フーリエ変換を行うオフセット離散フーリエ変換ステップと、
前記オフセット離散フーリエ変換で得られた周波数領域の周波数領域デジタル信号であって、一つ以上の周波数チャネルの前記周波数領域デジタル信号に対し復号を行う復号ステップと、
を有する周波数領域多重信号受信方法。 - 前記得られたデジタル信号の周波数に対して、周波数ずれの補償を行う周波数ずれ補償ステップと、
をさらに備え、
前記オフセット離散フーリエ変換ステップは、
前記周波数ずれが補償された周波数に周波数変換されたデジタル信号に対して、奇数の離散ポイント数のオフセット離散フーリエ変換を行う
請求項1に記載の周波数領域多重信号受信方法。 - 前記オフセット離散フーリエ変換ステップによって周波数領域に変換されたデジタル信号から、各周波数チャネルの残留周波数ずれ、もしくは全周波数チャネル共通の周波数ずれを推定し、前記周波数ずれを示す周波数ずれ情報を更新する周波数ずれ推定ステップと、
をさらに備え、
前記周波数ずれ補償ステップは、
前記周波数ずれ情報に基づいて、前記得られたデジタル信号の周波数に対して、前記周波数ずれの補償を行う
請求項1又は請求項2に記載の周波数領域多重信号受信方法。 - 前記オフセット離散フーリエ変換ステップは、
前記デジタル信号に対して、奇数の離散ポイント数のオフセット離散フーリエ変換を畳み込み演算により行い、前記デジタル信号に対応する周波数チャネルのみ演算して出力する
請求項1から請求項3のいずれか一項に記載の周波数領域多重信号受信方法。 - 前記オフセット離散フーリエ変換ステップは、
前記オフセット離散フーリエ変換の離散ポイント数を3とし、このうち2つの周波数チャネルを取得する
請求項1から請求項4のいずれか一項に記載の周波数領域多重信号受信方法。 - 前記オフセット離散フーリエ変換ステップは、
前記各周波数チャネルに独立の周波数ずれの補償を行う
請求項1から請求項4のいずれか一項に記載の周波数領域多重信号受信方法。 - 周波数領域に多重された受信信号を復号する周波数領域多重信号受信装置であって、
周波数領域に多重された受信信号からデジタル信号を取得するデジタル信号取得部と、
前記得られたデジタル信号に基づいて、奇数の離散ポイント数のオフセット離散フーリエ変換を行うオフセット離散フーリエ変換部と、
前記オフセット離散フーリエ変換で得られた周波数領域の周波数領域デジタル信号であって、一つ以上の周波数チャネルの前記周波数領域デジタル信号に対し復号を行う復号部と、
を有する周波数領域多重信号受信装置。 - 前記得られたデジタル信号の周波数に対して、周波数ずれの補償を行う周波数ずれ補償部と、
をさらに備え、
前記オフセット離散フーリエ変換部は、
前記周波数ずれが補償された周波数に周波数変換されたデジタル信号に対して、奇数の離散ポイント数のオフセット離散フーリエ変換を行う
請求項7に記載の周波数領域多重信号受信装置。 - 前記得られたデジタル信号から、前記オフセット離散フーリエ変換部における残留周波数ずれを推定し、前記周波数ずれを示す周波数ずれ情報を更新する周波数ずれ推定部と、
をさらに備え、
前記周波数ずれ補償部は、
前記周波数ずれ情報に基づいて、前記得られたデジタル信号の周波数に対して、前記周波数ずれの補償を行う
請求項7又は請求項8に記載の周波数領域多重信号受信装置。 - 周波数領域に多重された受信信号を復号する周波数領域多重信号受信方法であって、
周波数領域に多重された受信信号からデジタル信号を取得するデジタル信号取得ステップと、
前記得られたデジタル信号に対して、前記復号を行う周波数チャネル数に分岐する分岐ステップと、
前記分岐された各信号系列に対し、相関の低い異なる係数を初期ウエイトとして記憶する初期係数記憶ステップと、
前記分岐された各信号系列に対し、前記異なる係数で等化を行う等化ステップと、
前記等化された各信号系列に対し、復号を行う復号ステップと、
を有する周波数領域多重信号受信方法。 - 周波数領域に多重された受信信号を復号する周波数領域多重信号受信方法であって、
周波数領域に多重された受信信号からデジタル信号を取得するデジタル信号取得ステップと、
前記得られたデジタル信号に対して、前記復号を行う周波数チャネル数に分岐する分岐ステップと、
前記分岐された各信号系列に対し、各周波数チャネルに対応する離散フーリエ変換の係数と高い相関を有する係数、またはその少なくとも一部を含む係数を初期ウエイトとして記憶する初期係数記憶ステップと、
前記初期係数記憶ステップにおいて記憶された係数を前記初期ウエイトとして等化を行う等化ステップと、
前記等化された各信号系列に対し、復号を行う復号ステップと、を有する周波数領域多重信号受信方法。 - 前記分岐ステップは、
前記デジタル信号を分岐する際に、前記各周波数チャネルの中心周波数が特定の周波数成分付近となるように周波数変換を行ってから分岐を行い、
前記初期係数記憶ステップは、
離散フーリエ変換において前記特定の周波数成分に対応する離散フーリエ変換の係数と高い相関を有する係数、またはその少なくとも一部を含む係数を前記初期ウエイトとして記憶する
請求項11に記載の周波数領域多重信号受信方法。 - 前記等化ステップにおいて前記等化された各信号系列が同じ信号を示す信号系列に収束しないように、等化係数を調整する
請求項10から請求項12のいずれか一項に記載の周波数領域多重信号受信方法。 - 周波数領域に多重された受信信号を復号する周波数領域多重信号受信方法であって、
周波数領域に多重された受信信号を、異なる偏波成分に対して2つ取得し、それぞれからデジタル信号を取得するデジタル信号取得ステップと、
前記得られた2つのデジタル信号に対して、前記復号を行う周波数チャンネル数にそれぞれ分岐する分岐ステップと、
前記分岐された各信号系列のうち、k番目の周波数チャネルに対応する2つの信号系列に対し、nポイントの離散フーリエ変換の係数ak,0, ak,1, …, ak,n-1,を用い、2つの偏波成分の信号系列の初期ウエイトとして、一つの偏波に対応する信号の復号のために、[0, …, 0, bxxak,0, bxxak,1, …, bxxak,n-1, 0, … 0]と[0, …, 0, byxak,0, byxak,1, …, byxak,n-1, 0, … 0]を、異なる偏波に対応する信号の復号のために、[0, …, 0, bxyak,0, bxyak,1, …, bxyak,n-1, 0, … 0]と[0, …, 0, byyak,0, byyak,1, …, byyak,n-1, 0, …0]を、それぞれ、bxxbxy+byxbyy=0を満たすように記憶する初期係数記憶ステップと、
前記初期係数記憶ステップにおいて記憶された係数を、2つの偏波成分に対する受信信号に対し前記初期ウエイトとして用い、等化を行う等化ステップと、
前記等化された各信号系列に対し、復号を行う復号ステップと、
を有する周波数領域多重信号受信方法。 - 周波数領域に多重された受信信号を復号する周波数領域信号受信装置であって、
周波数領域に多重された受信信号からデジタル信号を取得するデジタル信号取得部と、
前記得られたデジタル信号に対して、前記復号を行う周波数チャネル数に分岐する分岐部と、
前記各周波数チャネルの信号の等化に用いる初期ウエイトを記憶する初期係数記憶部と、
前記分岐された各信号系列に対し、前記初期係数記憶部から入力された係数で等化を行う等化部と、
前記等化された各信号系列に対し、復号を行う復号部と、
を有する周波数領域多重信号受信装置。 - 周波数領域に多重された光信号を復号する周波数領域信号受信装置であって、
前記光信号を電気信号に変換する光電気変換部と、
前記電気信号からデジタル信号を取得するデジタル信号取得部と、
前記得られたデジタル信号に対して、前記復号を行う周波数チャネル数に分岐する分岐部と、
前記各周波数チャネルの信号の等化に用いる初期ウエイトを記憶する初期係数記憶部と、
前記分岐された各信号系列に対し、前記初期係数記憶部から入力された係数で等化を行う等化部と、
前記等化された各信号系列に対し、復号を行う復号部と、
を有する周波数領域多重信号受信装置。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP10799887.4A EP2456107B1 (en) | 2009-07-17 | 2010-07-15 | Method for receiving frequency domain multiplexed signal and device for receiving frequency domain multiplexed signal |
JP2011522847A JP5318207B2 (ja) | 2009-07-17 | 2010-07-15 | 周波数領域多重信号受信方法及び周波数領域多重信号受信装置 |
CN201080028997.1A CN102484549B (zh) | 2009-07-17 | 2010-07-15 | 频域复用信号接收方法及频域复用信号接收装置 |
US13/381,114 US9160600B2 (en) | 2009-07-17 | 2010-07-15 | Method for receiving frequency domain multiplexed signal and device for receiving frequency domain multiplexed signal |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2009-169455 | 2009-07-17 | ||
JP2009169460 | 2009-07-17 | ||
JP2009-169460 | 2009-07-17 | ||
JP2009169455 | 2009-07-17 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2011007829A1 true WO2011007829A1 (ja) | 2011-01-20 |
Family
ID=43449438
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2010/061972 WO2011007829A1 (ja) | 2009-07-17 | 2010-07-15 | 周波数領域多重信号受信方法及び周波数領域多重信号受信装置 |
Country Status (5)
Country | Link |
---|---|
US (1) | US9160600B2 (ja) |
EP (2) | EP2658198B1 (ja) |
JP (2) | JP5318207B2 (ja) |
CN (1) | CN102484549B (ja) |
WO (1) | WO2011007829A1 (ja) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5583788B2 (ja) * | 2010-11-29 | 2014-09-03 | 株式会社日立製作所 | 光通信システム、光送信器及びトランスポンダ |
CN103916148B (zh) * | 2013-01-05 | 2016-08-03 | 华为技术有限公司 | 一种自适应射频干扰抵消装置、方法、接收机及通信系统 |
EP3032790B1 (en) * | 2014-12-08 | 2018-02-21 | Vodafone GmbH | Generalized frequency division multiplexing radio transmission using frequency domain offset-qam |
US9973364B2 (en) * | 2016-06-27 | 2018-05-15 | Intel IP Corporation | Generalized frequency division multiplexing (GFDM) frame strucutre for IEEE 802.11AY |
CN108226598B (zh) * | 2018-01-05 | 2020-11-10 | 深圳市道通科技股份有限公司 | 一种示波器信号译码方法、装置及示波器 |
CN111698026B (zh) * | 2019-03-12 | 2023-05-12 | 富士通株式会社 | 偏振态变化监测方法、装置和接收机 |
CN112101179A (zh) * | 2020-09-10 | 2020-12-18 | 国网江苏省电力有限公司信息通信分公司 | 一种抗信道干扰瞬态射频指纹特征提取方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002101069A (ja) * | 2000-07-31 | 2002-04-05 | Thomson Licensing Sa | 直交周波数分割多重信号を処理するための方法およびシステム |
JP2003032222A (ja) * | 2001-07-19 | 2003-01-31 | Fujitsu General Ltd | Ofdm復調器用等化器 |
JP2009141514A (ja) * | 2007-12-04 | 2009-06-25 | Nippon Telegr & Teleph Corp <Ntt> | チャネル推定装置及び無線通信システム |
JP2009169460A (ja) | 2008-01-10 | 2009-07-30 | Sharp Corp | コンテンツ自動コピーシステム |
JP2009169455A (ja) | 2008-01-10 | 2009-07-30 | Casio Comput Co Ltd | サーバ装置、クライアント装置、サーバベースコンピューティングシステム、サーバ制御プログラム、クライアント制御プログラム |
Family Cites Families (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4300229A (en) | 1979-02-21 | 1981-11-10 | Nippon Electric Co., Ltd. | Transmitter and receiver for an othogonally multiplexed QAM signal of a sampling rate N times that of PAM signals, comprising an N/2-point offset fourier transform processor |
US5390185A (en) * | 1992-10-09 | 1995-02-14 | U.S. Philips Corporation | Transmission system for a combination of a main signal and an auxiliary signal |
US7440498B2 (en) * | 2002-12-17 | 2008-10-21 | Tellabs Operations, Inc. | Time domain equalization for discrete multi-tone systems |
US6631175B2 (en) * | 1998-04-03 | 2003-10-07 | Tellabs Operations, Inc. | Spectrally constrained impulse shortening filter for a discrete multi-tone receiver |
EP1063824B1 (en) * | 1999-06-22 | 2006-08-02 | Matsushita Electric Industrial Co., Ltd. | Symbol synchronisation in multicarrier receivers |
US6532258B1 (en) * | 1999-06-24 | 2003-03-11 | Ibiquity Digital Corporation | Method for estimating signal-to-noise ratio of digital carriers in an AM compatible digital audio broadcasting system |
US7352806B2 (en) * | 2001-12-06 | 2008-04-01 | Tensorcom, Inc. | Systems and methods for transmitting data in a wireless communication network |
US7317750B2 (en) | 2002-10-31 | 2008-01-08 | Lot 41 Acquisition Foundation, Llc | Orthogonal superposition coding for direct-sequence communications |
KR100463544B1 (ko) * | 2002-12-14 | 2004-12-29 | 엘지전자 주식회사 | 채널 등화 장치 및 이를 이용한 디지털 tv 수신기 |
US7580454B2 (en) * | 2004-07-14 | 2009-08-25 | Fundacion Tarpuy | Multidimensional decision-directed trained adaptive equalization |
US8665799B2 (en) * | 2006-09-14 | 2014-03-04 | Qualcomm Incorporated | Beacon assisted cell search in a wireless communication system |
US20080273480A1 (en) * | 2007-05-02 | 2008-11-06 | Mediaphy Corporation | Dynamic adjustment of training time for wireless receiver |
US8125885B2 (en) * | 2008-07-11 | 2012-02-28 | Texas Instruments Incorporated | Frequency offset estimation in orthogonal frequency division multiple access wireless networks |
-
2010
- 2010-07-15 US US13/381,114 patent/US9160600B2/en active Active
- 2010-07-15 EP EP13177394.7A patent/EP2658198B1/en active Active
- 2010-07-15 JP JP2011522847A patent/JP5318207B2/ja active Active
- 2010-07-15 WO PCT/JP2010/061972 patent/WO2011007829A1/ja active Application Filing
- 2010-07-15 CN CN201080028997.1A patent/CN102484549B/zh active Active
- 2010-07-15 EP EP10799887.4A patent/EP2456107B1/en active Active
-
2013
- 2013-06-14 JP JP2013126142A patent/JP5576533B2/ja active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002101069A (ja) * | 2000-07-31 | 2002-04-05 | Thomson Licensing Sa | 直交周波数分割多重信号を処理するための方法およびシステム |
JP2003032222A (ja) * | 2001-07-19 | 2003-01-31 | Fujitsu General Ltd | Ofdm復調器用等化器 |
JP2009141514A (ja) * | 2007-12-04 | 2009-06-25 | Nippon Telegr & Teleph Corp <Ntt> | チャネル推定装置及び無線通信システム |
JP2009169460A (ja) | 2008-01-10 | 2009-07-30 | Sharp Corp | コンテンツ自動コピーシステム |
JP2009169455A (ja) | 2008-01-10 | 2009-07-30 | Casio Comput Co Ltd | サーバ装置、クライアント装置、サーバベースコンピューティングシステム、サーバ制御プログラム、クライアント制御プログラム |
Non-Patent Citations (6)
Title |
---|
"Proceedings of the 2009 IEICE General Conference, Tsushin 2, 04 March 2009 (04.03.2009)", vol. B-10-46, article AKIHIDE SANO ET AL.: "13.4 Tb/s (134 x 111 Gb/s /ch) no-guard-interval PDM CO-OFDM transmission over 3, 600 km of SMF", pages: 365, XP008148276 * |
MASAHIRO MORIKURA; SHUJI KUBOTA: "Revised Third Version 802.11 High Speed Wireless LAN Textbook", 27 March 2008, IMPRESS R&D, pages: 156 - 157 |
MASAHIRO MORIKURA; SHUJI KUBOTA: "Revised Version 802.11 High Speed Wireless LAN Textbook", 27 March 2008, IMPRESS R&D |
NOBUYOSHI KIKUMA: "Adaptive Signal Processing with Array Antenna", 1998, KAGAKU GIJUTSU SHUPPAN, pages: 17 - 21 |
See also references of EP2456107A4 |
TAKESHI HATTORI: "OFDM/OFDMA Textbook", 11 September 2008, IMPRESS R&D |
Also Published As
Publication number | Publication date |
---|---|
US20120106618A1 (en) | 2012-05-03 |
CN102484549B (zh) | 2015-02-18 |
JPWO2011007829A1 (ja) | 2012-12-27 |
JP5318207B2 (ja) | 2013-10-16 |
EP2658198B1 (en) | 2016-10-19 |
EP2456107A4 (en) | 2013-02-20 |
US9160600B2 (en) | 2015-10-13 |
EP2658198A1 (en) | 2013-10-30 |
EP2456107B1 (en) | 2014-02-26 |
CN102484549A (zh) | 2012-05-30 |
EP2456107A1 (en) | 2012-05-23 |
JP2013179705A (ja) | 2013-09-09 |
JP5576533B2 (ja) | 2014-08-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5576533B2 (ja) | 周波数領域多重信号受信方法及び周波数領域多重信号受信装置 | |
CN110337788B (zh) | 光传输特性估计方法、光传输特性补偿方法、光传输特性估计系统及光传输特性补偿系统 | |
EP2434665B1 (en) | Signal generation circuit, optical signal transmitter, signal reception circuit, optical signal synchronization establishment method, and optical signal synchronization system | |
US8279953B2 (en) | Transmitting apparatus, receiving apparatus, and wireless communication method | |
JP5576300B2 (ja) | 直交周波数分割多重の受信機におけるi/q不均衡パラメーターを推定する装置及びその方法 | |
JP2010021941A (ja) | 無線通信装置及び無線通信方法、並びにコンピュータ・プログラム | |
US20170302482A1 (en) | Linear Equalization For Use In Low Latency High Speed Communication Systems | |
US10523283B2 (en) | LOS-MIMO demodulation apparatus, communication apparatus, LOS-MIMO transmission system, LOS-MIMO demodulation method and program | |
US8107817B2 (en) | Method for receiving optical orthogonal frequency-division multiplexing signal and receiver thereof | |
JP2001086092A (ja) | Ofdm通信装置および検波方法 | |
KR20080077755A (ko) | 다중 안테나 시스템에서 신호 보정 장치 및 방법 | |
EP2592801A1 (en) | Wireless communication receiver with phase noise estimation and phase noise compensation performed after channel estimation and related wireless communication receiving method | |
JP3910956B2 (ja) | Ofdm無線通信システムのための伝搬路推定器及びこれを用いた受信装置 | |
KR101599074B1 (ko) | 직교주파수분할다중화 방식을 지원하는 수신기에서 위상 잡음 보상장치 및 방법 | |
WO2012132191A1 (ja) | 受信装置 | |
EP1838030A2 (en) | Uplink signal receiving method and apparatus using successive interference cancellation in wireless transmission system based on OFDMA | |
US11228366B2 (en) | Optical transmission characteristics estimation method, optical transmission characteristics estimation system, and optical transmission characteristics compensation system | |
WO2022201387A1 (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及びプログラム | |
JP5116636B2 (ja) | 光信号受信方法、及び光信号受信装置 | |
EP3133751A1 (en) | Method for nonlinearity compensation in optical transmission systems | |
JP2012244421A (ja) | 受信装置、通信システム、及び通信方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201080028997.1 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 10799887 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2011522847 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2010799887 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13381114 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |