JP2003273829A - Ofdm receiver, ofdm communication method and ofdm communication program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an OFDM communication apparatus for realizing phase correction processing which has high accuracy and does not gradually lose accuracy, without increasing the scale of hardware. <P>SOLUTION: An Arctan calculation vector phase rotator 104 calculates the phase of a vector signal. A phase correction information estimator 105 determines an estimated multipath-fading deviation value, on the basis of the phase of a vector in a symbol for phase error difference estimation, and determines an estimated phase deviation value due to a factor other than that on the basis of a vector in a pilot signal. An adder 106 adds these two estimated phase deviation values and outputs it as a total phase correction value. The rotator 104 rotates the phase of a vector to be demodulated, on the basis of the total phase correction value to execute phase correction processing. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a receiver for OFDM communication in a digital wireless communication system, and more particularly to a phase compensator and a method of estimating phase compensation information for OFDM communication.

[0002]

2. Description of the Related Art A main cause of deterioration of transmission characteristics in a transmission path of wireless communication is multipath fading. Strong OFDM (Ort) against this multipath fading
hogonal Frequency Divisio
In recent years, the n-Multiplexing) transmission method has been drawing attention. This OFDM system is a system for multiplexing a large number (tens to thousands) of digital modulated waves that are orthogonal to each other in a certain frequency band. The signal received on the receiving side is observed in a state in which the phase has changed due to the influence of noise or the like. Therefore, in the communication according to this method, the phase correction process is performed after the OFDM demodulation process is performed by the FFT.
There are the following four factors as the factors of the phase error.

The first cause of phase error is due to multipath fading. In addition to the direct wave of the transmission signal, the input to the receiving-side antenna is a composite wave of a plurality of indirect waves that are reflected by the surrounding structures on the receiving side and jump into the antenna. The indirect wave has an individual time difference from the direct wave. The states of the direct wave and the indirect wave change depending on the environment between the transmitting side and the receiving side. On the receiving side, a signal whose phase is changing is received by receiving a composite wave of the direct wave and the indirect wave group having an individual time difference. This phenomenon is called multipath fading.

If the state of the transmission path during communication is constant, the influence of multipath fading on the phase is uniquely determined by the frequency band of communication data. Also, of course, if the state of the transmission line during communication changes,
The state of phase shift also changes. For example, in mobile communication, the transmission side and the reception side move from one place to another, so that the state of the transmission path constantly changes. Therefore, the influence of the multipath fading on the signal phase also changes with time.

The second cause of the phase error is an error caused by the frequency shift of the quadrature detection section. If there is a difference between the frequency of the quadrature detector and the frequency of the received signal,
A uniform phase shift occurs in the signal after detection.

The third cause of the phase error is the fluctuation of the carrier (the carrier wave between the transmitting side and the receiving side). An unstable waveform of the carrier itself may cause an error during sampling. Due to the shift of the sampling position, a uniform phase shift occurs in the sampled data.

The fourth cause of the phase error is the deviation of the sampling frequency itself. If the sampling frequency on the receiving side deviates from the frequency on the transmitting side, the sampling position gradually deviates, and the phase of the vector after sampling gradually deviates. This tends to increase the phase shift as it goes to the subsequent vector data.

FIG. 17 is a diagram for explaining the phase error due to the second to fourth factors. The circuit shown in FIG. 17 is provided before the FFT circuit in the OFDM communication device. As shown in FIG. 17A, the phase shift θ between the frequency f of the signal and the frequency f1 of the quadrature detection unit 999.
When there is 1, a phase error due to the second factor occurs.
In addition, there is a phase shift θ due to the third factor, and
When there is a phase shift φ due to the factor of, the data after sampling has a phase difference of θ + nφ as shown in FIG.
The phase error (n is an integer) is multiplied. As will be described later, nφ
With respect to, by obtaining φ from the correlation between adjacent vectors, correction can be performed before FFT. θ can be corrected by the phase correction process after FFT.

Next, conventional solutions to the above-mentioned first to fourth factors will be described. FIG. 18 is a block diagram showing the configuration of a conventional OFDM communication device. In FIG. 18, O
The FDM communication device includes a receiving circuit 991 and an FFT circuit 99.
2, the transmission path estimation circuit 993, and the phase compensation circuit 994.
And a demodulation error correction circuit 995 (for example, see Patent Document 1).

First, a conventional solution to the first factor will be described. To solve the first factor, a known signal prepared in the header part of the communication packet is used. Figure 1
9 is a diagram showing the structure of a communication packet. As shown in FIG. 19, the header portion 206 in the communication packet includes symbols 207 other than the channel estimation, and a phase error estimation symbol 208 due to the channel. The phase error estimation symbol 208 by the transmission path includes a known signal for solving the first factor. The data of the known signal is O
It is known in advance by the FDM communication device side. In the OFDM communication apparatus, after the OFDM demodulation (FFT operation) in the FFT circuit 992, in the transmission path estimation circuit 993, the data of this known signal and the complex conjugate of the received data are complex-multiplied to obtain A transmission line response vector indicating a transfer function due to multipath fading is obtained. After that, the transmission path estimation circuit 993 rotates the phase of the reception signal by complexly multiplying the reception signal after that by the transmission path response vector, and corrects the phase shift due to the first factor. .

Next, conventional solutions to the second to fourth factors will be described. When the subcarrier signals in the OFDM symbol are arranged on the frequency axis and the phase shift amount is set in the vertical axis direction, the distribution of the phase shift amount of the signal when viewed in the frequency axis direction is as shown in FIG. It can be seen that it is represented by a linear function due to the properties of the second to fourth factors. Therefore, the phase compensation circuit 994 in the OFDM communication device is
The phase shift amount of the phase error estimation pilot signal 209, which is a known signal included in the symbols other than the header portion of the communication packet, is measured first, and the phase shift amount of each transmission / reception signal is calculated from the phase shift amount of each pilot signal. presume. Then, the phase compensation circuit 994 performs a phase rotation process using the estimated phase shift amount as the phase rotation amount of each vector by the CORDIC algorithm, and corrects the phase shift due to the second to fourth factors.

Here, the solution to the second to fourth factors is processing in units of OFDM symbols, but for OFDM symbols in which burst noise or error has occurred, the reliability of the signal contained therein is Is significantly reduced.
In estimating the phase shift amount in the OFDM symbol, it is considered that the reliability of the phase correction process is low only by the estimation result of the phase shift amount obtained from the pilot signal in the target symbol. Therefore, the phase compensation circuit 994 is
The cumulative average value of the phase correction information obtained in the past is used as the phase correction information of the target symbol.

[0013]

[Patent Document 1] JP 2001-86092 A (4th
(Page to Page 5, Figure 1)

[0014]

However, the above-mentioned conventional solutions have the following three problems.

(First Problem) In order to solve the above first to fourth factors at the same time, a phase correction process for a received signal,
That is, it is necessary to perform the vector rotation processing twice by the transmission path estimation circuit 993 and the phase compensation circuit 994.
Therefore, the conventional phase correction process has a problem that the hardware scale of the OFDM communication device is increased.

(Second Problem) With respect to the solutions to the second to fourth factors of the phase shift of the signal, the estimated value of the phase shift amount in the OFDM symbol is calculated by the cumulative average value from the packet head. If there is an OFDM symbol with a large amount of noise in the past, the estimated value of the phase shift amount will continue to be affected by the unreliable phase error estimated amount obtained from this symbol. In addition, the estimated amount is located far from the target OFDM symbol in terms of time, and the correlation of the phase correction amount is considered to be small.
It will also be affected by the symbol. Therefore, the phase correction amount estimated by the method of calculating the estimated value from the cumulative average lacks accuracy, and the effect of the phase correction processing by this method is low.

(Third Problem) As to the solution to the first factor of the phase shift of the signal, the estimation of the transfer function by the multipath fading is performed only once at the beginning of a long packet. This estimate is the OFD that is backward in time.
It also applies to M symbols. This method cannot follow the change in the multipath state during packet transfer, that is, the change in the transfer function indicating the state of the transmission path. Therefore, the accuracy of the phase correction process by this method may gradually be lost.

Therefore, an object of the present invention is to provide an OFD that realizes a phase correction process with high accuracy and without gradual loss of accuracy without increasing the hardware scale.
An M communication device is provided.

[0019]

According to a first aspect of the present invention, a received OFDM signal is subjected to a Discrete Fourier Transform (FFT: Fast Fourier Transform).
rm) An OFDM communication apparatus for correcting the phase of a vector signal obtained by performing a process, comprising: a phase calculation unit that calculates the phase of the vector signal; and a plurality of OFDM signals included in the OFDM signal calculated by the phase calculation unit. Regarding the phase of each vector signal for the known signal, by comparing each with the known phase, a phase correction information estimation unit that outputs each phase shift estimated value for each factor of multiple phase shifts, and a phase correction A plurality of phase shift estimation values output by the information estimation unit are added together, and a summing unit that outputs as a total phase correction value, and the phase of the vector signal is rotated by the total phase correction value output by the summation unit, and the phase is And a vector phase rotation unit that outputs a corrected vector signal.

According to the first aspect of the present invention, the phase shift estimated value is calculated for each factor, the phase shift estimated values are summed, the total phase correction value is calculated, and the phase rotation processing of the received vector signal is performed. Since it is performed at once, the scale of hardware required for the phase correction processing is reduced as compared with the case where the phase rotation processing is performed for each factor. This solves the first problem.

A second aspect of the invention is an aspect dependent on the first aspect of the invention, wherein the phase correction information estimator is an OF to be demodulated.
Each phase shift estimation is performed by obtaining a weighted average of the phase shift estimated value obtained from the known signal included in the DM symbol and the phase shift estimated value obtained from the known signal included in the OFDM symbol other than the demodulation target. It is characterized by outputting a value.

According to the second aspect of the present invention, the final estimated phase shift value is calculated by calculating the estimated phase shift value from the OFDM symbols other than the demodulation target and obtaining the weighted average. Therefore, in estimating the phase shift, it is possible to disperse the influence of the phase shift amount estimated from the OFDM symbol having a large amount of noise, so that it is possible to further improve the communication quality. For example, since a large amount of noise is present in the OFDM symbol to be demodulated, it is possible to avoid a phase shift estimation value having low reliability. As a result, the second
The problem of is solved.

A third invention is an invention dependent on the second invention, wherein the phase correction information estimation unit has a phase shift estimated value obtained from an OFDM symbol adjacent to an OFDM symbol currently being demodulated, The weighted value is increased to obtain a weighted average.

According to the third aspect of the present invention, the phase shift estimated values obtained from the OFDM symbols adjacent to the OFDM symbol currently being demodulated are weighted and averaged with a larger weight, so that they are temporally averaged. OF which is considered to have little correlation with the estimated phase shift value due to the distance
It is possible to reduce the influence of the phase shift estimated value estimated from the DM symbol. Therefore, the estimated value error of the total phase correction value can be reduced, and the communication quality can be improved. This solves the second problem described above.

A fourth invention is an invention dependent on the third invention, wherein a phase shift estimated value obtained from a known signal included in an OFDM symbol other than a demodulation target is before a demodulation target OFDM symbol. It is characterized in that it is an estimated value of phase shift obtained from OFDM symbols for a certain section.

According to the fourth aspect of the present invention, the estimated phase shift value obtained from the OFDM symbols for a certain period before the OFDM symbol to be demodulated is used for weighted averaging, so that noise is large in the past. OFD on board
The influence of the unreliable phase shift estimation value estimated from the M symbols will eventually be cut. This solves the second problem described above.

A fifth invention is an invention dependent on the third invention, wherein a phase shift estimation value obtained from a known signal included in an OFDM symbol other than the demodulation target is before the OFDM symbol to be demodulated. It is characterized in that it is a phase shift estimation value obtained from all OFDM symbols.

According to the fifth aspect of the invention, since the phase shift estimated values obtained from all the OFDM symbols before the OFDM symbol to be demodulated are used for the weighted average, a large amount of noise has been added in the past. It is possible to reduce the influence of the unreliable phase shift estimation value estimated from the OFDM symbol. This solves the second problem described above.

A sixth invention is according to the third invention, wherein the phase shift estimated value obtained from the known signal included in the OFDM symbol other than the demodulation target is the OFDM symbol after the OFDM symbol which is the demodulation target. It is characterized in that it is an estimated value of phase shift obtained from the OFDM symbol.

According to the sixth aspect of the invention, the OFD after the OFDM symbol to be demodulated (the target of the phase correction processing)
By taking into consideration the estimated value of phase shift estimated from M symbols, it becomes possible to cope with the tendency of temporal changes in phase shift, which improves the accuracy of the total phase correction value and improves communication quality. Becomes This solves the second problem described above.

A seventh invention is an invention dependent on the third invention, wherein a phase shift estimated value obtained from a known signal included in an OFDM symbol other than the demodulation target is before the OFDM symbol to be demodulated. The phase shift estimation value obtained from the OFDM symbol and the phase shift estimation value obtained from the OFDM symbol after the OFDM symbol to be demodulated are characterized.

According to the seventh invention, past OFDM
Since the estimated phase shift value is obtained using the estimated phase shift value estimated from the symbol and the estimated phase shift value estimated from the OFDM symbols after the demodulation target, it is necessary to handle the tendency of the temporal shift of the phase shift. It is possible to improve the accuracy of the total phase correction value and improve the communication quality. This solves the second problem described above.

An eighth invention is an invention according to the second invention, further comprising a reception state discrimination section for discriminating a reception state of the received OFDM signal, wherein the reception state discrimination section receives the received OFDM signal. It is characterized in that the weight value of the weighted average by the phase correction information estimation unit is changed according to the reception state of the OFDM symbol included in the signal.

According to the eighth aspect of the invention, the reception state of the received signal is discriminated and the weight value is changed according to the reception state. Therefore, the phase shift estimated from the OFDM symbol which is judged to have low reliability. Since the influence of the estimated value can be reduced and the influence of the phase shift estimated value estimated from the OFDM symbol judged to have high reliability can be increased, more reliable and accurate phase correction can be achieved. It becomes possible and the communication quality can be improved. This solves the second problem described above.

A ninth invention is according to the first invention, further comprising a re-encoding unit for re-encoding the demodulated OFDM symbol periodically, and the phase calculation unit re-encodes. The unit obtains the phase of the vector for the encoded OFDM symbol, and the phase correction information estimation unit has already calculated the phase of the vector for the known signal included in the re-encoded OFDM symbol obtained by the phase calculation unit. It is characterized in that the calculated phase shift estimated value is corrected.

According to the ninth aspect of the invention, the data after demodulation is periodically re-encoded to form a reference for the received signal, so that the latest information on the influence of the transmission path on the signal can always be obtained. Is possible. Therefore, for example, it becomes possible to update the phase shift estimated value that is not frequently determined, such as the amount of phase shift due to multipath fading, and improve the accuracy of the phase shift estimated value.
It is possible to improve communication quality. This solves the third problem.

A tenth invention is an invention according to the first invention, characterized in that the phase calculating section calculates only the phase of the vector signal of the known signal.

According to the tenth aspect of the invention, since only the phase of the vector signal for the known signal needs to be calculated,
It is possible to reduce the processing of the device.

An eleventh invention is an invention subordinate to the first invention, and the phase correction information estimating section uses a polarity inverter for comparison with a known phase.

According to the eleventh aspect of the present invention, the phase shift estimated value can be easily obtained by using the polarity reversal device.

The twelfth invention is an invention dependent on the first invention, and the factors of the phase shift are classified into those due to multipath fading and those other than multipath fading. , Known symbols for phase error estimation due to multipath fading,
A plurality of known pilot signals included in each data symbol are included, and the phase correction information estimation unit subtracts the known phase from the phase of the vector for the known symbol for phase error estimation, thereby performing multipath A first phase shift estimated value calculation unit that calculates a phase shift estimated value due to fading, a known phase is subtracted from the phase of the vector with respect to the pilot signal, and a phase shift estimated value due to multipath fading is subtracted, Based on the pilot signal phase shift amount calculation unit that calculates the phase shift amount of the pilot signal due to factors other than path fading and the phase shift amount calculated by the pilot signal phase shift amount calculation unit, the phase shift caused by factors other than multipath fading A straight line for estimating A second phase shift estimated value calculation unit for calculating a phase shift estimated value due to a factor other than the multipath fading, and the summation unit with the phase shift estimated value due to a factor other than the multipath fading. The total phase correction value is calculated by adding up.

According to the twelfth aspect of the invention, the total phase correction value is obtained by summing the phase shift estimation value due to multipath fading and the phase shift estimation value due to factors other than multipath fading, and the phase correction processing is performed at once. Since this is done, it is possible to reduce the hardware scale. This solves the first problem.

A thirteenth invention is according to the twelfth invention, wherein the second phase shift estimated value calculating section has a linear parameter obtained from the pilot signal included in the OFDM symbol to be demodulated. And OFD other than the demodulation target
By obtaining a weighted average with the parameters of the straight line obtained from the pilot signal included in M symbols,
It is characterized in that the weighted average straight line is obtained and the phase shift estimation value due to factors other than multipath fading is calculated.

The effect of the thirteenth invention is similar to the effect of the second invention.

A fourteenth invention is an invention dependent on the thirteenth invention, wherein the second phase shift estimated value calculation unit detects the pilot signal adjacent to the pilot signal included in the OFDM symbol currently being demodulated. It is characterized in that the weighted value is increased for the obtained parameters and the weighted average is obtained.

The effect of the fourteenth invention is similar to that of the third invention.

A fifteenth invention is according to the fourteenth invention, wherein the second phase shift estimated value calculating section has a linear parameter obtained from a pilot signal included in an OFDM symbol other than a demodulation target. As the O to be demodulated
It is characterized in that a weighted average is obtained by storing parameters obtained from pilot signals included in OFDM symbols before FDM symbols.

The effect of the fifteenth invention is similar to that of the fourth invention.

A sixteenth invention is an invention dependent on the fifteenth invention, further comprising a vector storage section for temporarily storing a vector for an OFDM symbol to be demodulated, and an OFDM stored in the vector storage section. Among the vectors for the symbols, the pilot signal phase shift amount calculation unit is provided with only the pilot signal vector in advance, and the pilot signal phase shift amount calculation unit is the vector for the pilot signal given in advance. Of the phase shift amount is calculated and given to the second phase shift estimated value calculation unit.
As a linear parameter obtained from the pilot signal included in the FDM symbol, based on the phase shift amount of the preceding pilot signal provided from the pilot signal phase shift amount calculation unit, the OFDM symbols after the OFDM symbol to be demodulated The parameter obtained from the symbol is stored, and the parameter and the OFDM to be demodulated are stored.
It is characterized in that a weighted average is obtained by using a parameter obtained from a pilot signal included in the OFDM symbol before the symbol.

According to the sixteenth invention, the phase shift estimated value is calculated in advance based on the pilot signals included in the OFDM symbols to be demodulated and thereafter. Therefore, the same as in the seventh invention. Will have the effect of.

A seventeenth invention is an invention according to the fourteenth invention, and further includes a vector storage unit for temporarily storing a vector for an OFDM symbol to be demodulated, and an OFDM stored in the vector storage unit. Among the vectors for the symbols, the pilot signal phase shift amount calculation unit is provided with only the pilot signal vector in advance, and the pilot signal phase shift amount calculation unit is the vector for the pilot signal given in advance. Of the phase shift amount is calculated and given to the second phase shift estimated value calculation unit.
As a linear parameter obtained from the pilot signal included in the FDM symbol, based on the phase shift amount of the preceding pilot signal provided from the pilot signal phase shift amount calculation unit, the OFDM symbols after the OFDM symbol to be demodulated It is characterized in that a weighted average is obtained by storing parameters obtained from the symbols.

According to the seventeenth invention, the phase shift estimated value is calculated in advance based on the pilot signals included in the OFDM symbols to be demodulated and thereafter. Therefore, similar to the sixth invention. Will have the effect of.

An eighteenth invention is an invention according to the thirteenth invention, further comprising a reception state discrimination section for discriminating the reception state of the received OFDM signal, and the reception state discrimination section is provided with the received OFDM signal. It is characterized in that the weight value of the weighted average by the second phase shift estimated value calculation unit is changed according to the reception state of the OFDM symbol included in the signal.

The effects of the eighteenth invention are similar to those of the eighth invention.

A nineteenth invention is an invention according to the twelfth invention, further comprising a re-encoding unit for re-encoding the demodulated OFDM symbol periodically, and the phase calculation unit includes:
The re-encoding unit calculates the phase of the vector for the OFDM symbol encoded, and the pilot signal phase shift calculation unit calculates the phase of the vector for the pilot signal included in the re-encoded OFDM symbol calculated by the phase calculation unit. It is characterized in that the phase of the vector of the pilot signal corresponding to the OFDM symbol already calculated is subtracted, and the phase shift estimated value due to multipath fading is corrected based on the result.

The effects of the nineteenth invention are similar to those of the ninth invention.

A twentieth aspect of the present invention is to perform a discrete Fourier transform (FFT) on the received OFDM signal.
r Transform) is an OFDM communication device for correcting the phase of a vector signal, the phase calculating section calculating the phase of the vector signal, and the pilot signal included in the OFDM symbol calculated by the phase calculating section. Based on the phase of the vector, the phase shift estimated value calculation unit that obtains the phase shift estimated value for each OFDM symbol after obtaining the phase shift estimated line due to factors other than multipath fading, and the phase shift estimated value calculation unit The phase shift estimated value calculation unit rotates the phase of the vector signal by a phase shift estimated value due to a factor other than the multipath fading calculated by, and outputs a vector signal whose phase is corrected. , Phase obtained from pilot signal included in OFDM symbol to be demodulated The phase shift estimated value is obtained by obtaining a weighted average of the parameters of the shift estimation straight line and the parameters of the phase shift estimation straight line obtained from the pilot signals included in the OFDM symbols other than the demodulation target.

According to the twentieth aspect of the present invention, the final phase shift estimated value is calculated by calculating the phase shift estimated value from the OFDM symbols other than the demodulation target and obtaining the weighted average. Therefore, in estimating the phase shift, it is possible to disperse the influence of the phase shift amount estimated from the OFDM symbol having a large amount of noise, so that it is possible to further improve the communication quality. For example, since a large amount of noise is present in the OFDM symbol to be demodulated, it is possible to avoid a phase shift estimation value having low reliability. As a result, the second
The problem of is solved.

A twenty-first aspect of the invention is a subject matter of the twentieth aspect of the invention, wherein the phase-shift-estimation-value calculating unit has a parameter of a phase shift estimation straight line obtained from a pilot signal included in an OFDM symbol other than a demodulation target. Is characterized in that the parameter of the estimated straight line obtained from the pilot signal included in the OFDM symbol before the demodulation target is stored and the weighted average with the past parameter is obtained.

According to the twenty-first aspect, it is possible to reduce the influence of the low-reliability phase shift estimated value estimated from the OFDM symbol having a large amount of noise in the past.

A twenty-second invention is an invention dependent on the twentieth invention, and further comprises a vector storage section for temporarily storing a vector for an OFDM symbol to be demodulated, and an OFDM stored in the vector storage section. Of the vectors for the symbols, a control unit that gives only the vector of the pilot signal to the phase shift estimated value calculation unit in advance, the phase shift estimated value calculation unit estimates based on the pilot signal given in advance. It is characterized in that the weighted average with the parameters of the estimated straight line after the demodulation target is obtained by obtaining and storing the parameters of the straight line.

According to the twenty-second aspect, it is possible to cope with the tendency of the phase shift to change with time, improve the accuracy of the total phase correction value, and improve the communication quality.

The twenty-third aspect of the present invention is to perform a discrete Fourier transform (FFT) on the received OFDM signal.
r Transform) is an OFDM communication device for correcting the phase of a vector signal, the phase calculating unit calculating the phase of the vector signal, and the multipath included in the OFDM symbol obtained by the phase calculating unit. Based on the phase of the vector for the known symbol for phase error estimation due to fading, a phase shift estimated value calculation unit that obtains a phase shift estimated value due to multipath fading, and a phase due to multipath fading calculated by the phase shift estimated value calculation unit A vector phase rotation unit that rotates the phase of the vector signal by the estimated deviation value and outputs a vector signal with the phase corrected, and a demodulated OFDM
And a re-encoding unit that periodically re-encodes the symbol, the phase calculation unit obtains the phase of the vector for the OFDM symbol encoded by the re-encoding unit, and the phase shift estimated value calculation unit calculates the phase. Re-encoded OFD obtained by the department
From the phase of the vector of the pilot signal included in the M symbol, the phase of the vector of the pilot signal corresponding to the OFDM symbol already calculated is subtracted, and the phase shift estimated value due to multipath fading is calculated based on the result. It is characterized by correction.

According to the twenty-third aspect of the invention, the data after demodulation is periodically re-encoded to form a reference for the received signal, so that the latest information on the influence of the transmission path on the signal can always be obtained. Is possible. Therefore, it is possible to update the phase shift amount due to multipath fading, improve the accuracy of the phase shift estimated value, and improve the communication quality. This solves the third problem.

The twenty-fourth invention is a discrete Fourier transform (FFT: Fast Fourier) of the received OFDM signal.
r Transform) processing to correct the phase of the vector signal,
The step of calculating the phase of the vector signal and the phase of each vector signal for the plurality of known signals included in the calculated OFDM signal are compared with the known phase, respectively, and a plurality of factors of phase shift exist. Calculating each phase shift estimation value for each
A step of setting a value obtained by adding a plurality of calculated phase shift estimated values as a total phase correction value, and a step of rotating the phase of the vector signal by the total phase correction value to obtain a vector signal with the phase corrected Prepare

The effects of the 24th invention are similar to those of the first invention.

A twenty-fifth aspect of the present invention is to perform a discrete Fourier transform (FFT) on the received OFDM signal.
r Transform) processing to correct the phase of the vector signal,
Based on the step of calculating the phase of the vector signal and the phase of the vector for the pilot signal included in the calculated OFDM symbol, the phase shift estimation line due to factors other than multipath fading is obtained, and then each OFD is calculated.
A step of calculating a phase shift estimated value for M symbols, and a step of rotating the phase of the vector signal by the calculated phase shift estimated value due to factors other than multipath fading to obtain a phase-corrected vector signal. In the step of calculating the estimated phase shift value, the phase shift estimation straight line parameter obtained from the pilot signal included in the OFDM symbol to be demodulated and the pilot signal included in the OFDM symbol other than the demodulated target are obtained. The phase shift estimation value is obtained by obtaining a weighted average with the parameters of the phase shift estimation line.

The effect of the 25th invention is the same as that of the 20th invention.

The twenty-sixth aspect of the present invention is to apply a discrete Fourier transform (FFT: Fast Fourier) to the received OFDM signal.
r Transform) processing to correct the phase of the vector signal,
Calculating the phase of the vector signal, and calculating the phase shift estimated value due to multipath fading based on the phase of the vector for the known symbol for phase error estimation due to multipath fading included in the calculated OFDM symbol A step of rotating the phase of the vector signal by the calculated phase shift estimated value due to the multipath fading to obtain a vector signal with the phase corrected, and a step of periodically re-encoding the demodulated OFDM symbol. And recoded OFDM
Calculating the phase of the vector for the symbol, and subtracting the phase of the vector of the pilot signal corresponding to the OFDM symbol already calculated from the phase of the vector for the pilot signal included in the recoded OFDM symbol. , Based on the results
Updating the phase shift estimation value due to multipath fading.

The effects of the 26th invention are similar to those of the 23rd invention.

The twenty-seventh aspect of the present invention is to apply a discrete Fourier transform (FFT: Fast Fourier) to the received OFDM signal.
A program executed by an OFDM communication apparatus that corrects the phase of a vector signal obtained by performing r.Transform) processing, the step of calculating the phase of the vector signal, and a plurality of known values included in the calculated OFDM signal. For each vector signal phase for a signal, by comparing each with a known phase,
For each of the multiple phase shift factors that exist, a step for calculating each phase shift estimated value, a step for adding the calculated multiple phase shift estimated values to a total phase correction value, and a total phase correction value Only, the phase of the vector signal is rotated, and the phase-corrected vector signal is obtained.

The effects of the twenty-seventh invention are similar to those of the first invention.

The twenty-eighth aspect of the invention is to perform a discrete Fourier transform (FFT) on the received OFDM signal.
r Transform) is a program executed in an OFDM communication apparatus for correcting the phase of a vector signal obtained by performing a r signal transform process, the step of calculating the phase of the vector signal, and the pilot signal included in the calculated OFDM symbol. Based on the phase of the vector, after calculating the phase shift estimation line due to factors other than multipath fading, calculating the phase shift estimated value for each OFDM symbol, and calculating the phase shift due to factors other than multipath fading. Rotating the phase of the vector signal by the estimated value to obtain a vector signal with the phase corrected, and calculating the phase shift estimated value is obtained from the pilot signal included in the OFDM symbol to be demodulated. Of the estimated phase shift line The phase shift estimation value is obtained by obtaining a weighted average of the parameters of the phase shift estimation straight line obtained from the pilot signals included in the OFDM symbols other than the demodulation target.

The effects of the 28th invention are similar to those of the 20th invention.

The twenty-ninth invention is a discrete Fourier transform (FFT: Fast Fourier) of the received OFDM signal.
r Transform) is a program executed by an OFDM communication apparatus for correcting the phase of a vector signal obtained by performing r.Transform) processing, which comprises a step of calculating the phase of the vector signal and a multipath fading included in the calculated OFDM symbol. Based on the phase of the vector for the known symbol for phase error estimation, the step of calculating the estimated phase shift value due to multipath fading, and rotating the phase of the vector signal by the calculated estimated phase shift value due to multipath fading. Let
Obtaining a phase-corrected vector signal, periodically re-encoding the demodulated OFDM symbol, obtaining a vector phase for the re-encoded OFDM symbol, and re-encoding OFD
From the phase of the vector of the pilot signal included in the M symbol, the phase of the vector of the pilot signal corresponding to the OFDM symbol already calculated is subtracted, and the phase shift estimated value due to multipath fading is calculated based on the result. Updating.

The effects of the twenty-ninth invention are similar to those of the twenty-third invention.

[0077]

BEST MODE FOR CARRYING OUT THE INVENTION (First Embodiment) FIG. 1 is a functional block diagram showing the configuration of an OFDM communication apparatus 1 according to the first embodiment of the present invention. In FIG. 1, an OFDM communication device 1 includes an antenna 101, a receiving circuit 102, an F
FT circuit 103, Arctan calculation vector phase rotator 104, phase correction information estimator 105, and adder 10
6 and a demodulation error corrector 107.

In the first embodiment, a communication packet having the structure shown in FIG. 19 is used as a signal used in OFDM communication for estimating and compensating the residual phase error using pilot symbols. That is, the packet signal used in the first embodiment is a symbol 207 other than the channel estimation and a symbol for phase error estimation by the channel (hereinafter, simply referred to as a symbol for phase error estimation) 20.
8 and a plurality of transmission / reception data symbols (hereinafter, simply referred to as data symbols) 211 having data. Each data symbol 211 has a plurality of phase error estimation pilot signals (hereinafter, simply referred to as pilot signals) 209 arranged at regular intervals.
And a plurality of transmission / reception data 210 on which data is added. In FIG. 19, the pilot signal is shown by a black frame and the data signal is shown by a white frame. Note that, in FIG. 19, reference numerals of the data symbol 211 and the transmission / reception data 210 are partially omitted. The communication packet used in the other embodiments is also the same as that shown in FIG. 19, so that FIG. 19 is also referred to in the other embodiments.

OFD received via antenna 101
The M signal is subjected to normal wireless reception processing in the receiving circuit 102 and becomes a digital signal. This digital signal is
FFT (Fast Fourier Transform)
m) An FFT operation is performed by the circuit 103. The FFT circuit 103 outputs the subcarrier signal vector (in-phase component and quadrature component) assigned to each subcarrier and inputs it to the Arctan calculation vector phase rotator 104.

Arctan calculation vector phase rotator 10
4 rotates the phase of the vector input from the FFT circuit 103 based on the Arctan calculation processing function of calculating the arctan of the phase of the vector input from the FFT circuit 103 and the phase correction amount input from the adder 106. And a vector phase rotation processing function. These two functions are realized by the CORDIC algorithm described later. Further, as will be described later, most of the circuit configurations that realize these two functions are shared.

Arctan calculation vector phase rotator 10
4 is a vector of a known signal included in the phase error estimation symbol 208 (hereinafter,
The phase of the known signal vector) is calculated and input to the phase correction information estimator 105.

As shown in FIG. 19, there are two types of subcarrier signal vectors, a vector corresponding to pilot signal 209 (hereinafter referred to as pilot signal vector) and a vector corresponding to transmission / reception data 210 (hereinafter referred to as data vector). are categorized. The Arctan calculation vector phase rotator 104 obtains the phase of the pilot signal vector among the subcarrier signal vectors by the Arctan calculation process, and calculates the phase correction information estimator 105.
To enter. Arctan calculation vector phase rotator 10
4, the timing for obtaining the phases of the known signal vector and the pilot signal vector among the input subcarrier signal vectors is controlled by a control circuit (not shown) and a switch circuit (not shown) that generate regular timings. Done.

The phase of the known signal vector is known. The phase correction information estimator 105 compares the known phase with the phase of the received known signal vector to obtain the phase shift amount of the received known signal vector. This phase shift amount estimates the influence of the phase on the signal due to the multipath fading (due to the first factor).
This estimated phase shift amount is called a phase shift estimated value due to multipath fading. Phase correction information estimator 10
5 temporarily stores the calculated phase shift estimated value due to multipath fading in the memory, and inputs it to the adder 106 as necessary.

The phase of the pilot signal vector is also known. The phase correction information estimator 105 compares the known phase with the phase of the received pilot signal vector to obtain the phase shift amount of the received pilot signal vector. The phase correction information estimator 105 subtracts the phase shift estimation value due to multipath fading stored in the memory from the calculated phase shift amount of the pilot signal vector, and determines the second to fourth factors for the pilot signal vector. Phase shift estimation value due to only (hereinafter referred to as phase shift estimation value due to factors other than multipath fading)
Ask for. Then, the phase correction information estimator 105 obtains a linear functional expression of the phase shift amount on the frequency-phase shift amount plane as shown in FIG. 20, calculates the phase correction amount for each subcarrier, and outputs it to the adder 106. input.

The adder 106 adds the estimated value of phase shift due to multipath fading and the estimated value of phase shift due to factors other than multipath fading, and inputs it to the Arctan calculation vector phase rotator 104. This total value is called the total phase correction value.

Arctan calculation vector phase rotator 10
4 is the FFT circuit 10 by the vector phase rotation processing.
The phase of the data vector output from 3 is calculated by the adder 10
The total phase correction value input from 6 is rotated to collectively correct the phase shift due to a plurality of factors included in the data vector, and the corrected data vector is sent to the demodulation error corrector 107.

The demodulation error corrector 107 performs demodulation processing and error correction processing on the corrected data vector input from the Arctan calculation vector phase rotator 104, and outputs communication content data.

Next, detailed functions of the Arctan calculation vector phase rotator 104 (vector rotation processing function and A
The rctan calculation processing function) will be described. Arct
The an-computation vector phase rotator 104 is hardware capable of processing the CORDIC algorithm and performs phase calculation of an input vector and arbitrary phase rotation processing. To simply rotate the phase of the complex vector by θ, (c
osθ + jsinθ) may be multiplied. This calculation method requires a table for sin θ and a table for cos θ, but the table size becomes enormous in proportion to the square of the required calculation accuracy. Further, since rotation requires complex multiplication, hardware for that purpose is also required. To avoid these
In the CORDIC algorithm, the input vector is rotated by the required angle by repeating the complex multiplication on the input vector while changing the sign of the complex vector shown in (Equation 1). Table 1 shows complex vector groups used in the CORDIC algorithm.

[0089]

[Equation 1]

[0090]

[Table 1]

As an example, the input vector (1,0)
The case of performing a rotation process of 30 degrees will be described. The state of the rotation processing is shown in Table 2 and will be described step by step.

[Table 2]

Since the required rotation phase is 30 degrees, Ar
The ctan calculation vector phase rotator 104 uses n =
The input vector is multiplied by the multiplier vector value (1 + j) noted 0, and the input vector is rotated 45 degrees counterclockwise. Next, when the required rotation angle of 30 degrees is compared with the current phase displacement of 45 degrees of the input vector, the current phase displacement of the input vector is rotated more than the required rotation angle. Vector phase rotator 10
No. 4 multiplies the multiplication vector value (1-j / 2), and in the next stage, rotates clockwise by 26.6 degrees. At this stage, the phase displacement of the input vector becomes 18.4 degrees, and the amplitude is 1.414 × 1.118 =
It becomes 1.580 times.

As described above, the Arctan calculation vector phase rotator 104 is used by the CORDIC algorithm.
Compares the total rotation amount of the input vector with the required rotation amount for each rotation process, and gradually drives the total rotation amount to the required rotation amount. This causes the Arctan calculation vector phase rotator 104 to rotate the input vector by the initially required phase to obtain a vector whose amplitude is approximately 1.646 times. Finally, the Arctan calculation vector phase rotator 10
4 is 1 / for the real and imaginary components of this vector
The amplitude is corrected by multiplying 1.646 = 0.607, and the rotation process of the input vector ends.

Next, the Arctan of the input vector
The calculation process will be described. In the Arctan calculation process, since only the process of driving the input vector to 0 degrees is performed, the vector phase rotation process and the vector Arc process are performed.
The tan calculation process can be realized by almost the same circuit.

In the above vector phase rotation processing, A
The rctan calculation vector phase rotator 104 compares the total rotation amount of the vector with the required rotation amount for each rotation process, determines the next rotation direction, and drives the phase of the input vector. I do. On the other hand, Arc
In the case of the tan calculation processing, the Arctan calculation vector phase rotator 104 determines the phase of the input vector by looking at the sign of the imaginary number component (Y coordinate in the XY plane) of the input vector for each rotation processing. The phase of the input vector is driven to 0 degree by determining whether it is above or below 0 degree and determining the sign of the next rotation processing. Arc
The tan calculation vector phase rotator 104 determines the Arct of the input vector by observing the total rotation amount of the input vector phase after the rotation calculation of a predetermined number of times is completed.
ask for an. The vector phase rotation process and the Arctan calculation process are different only in the condition of code determination in each rotation process. Therefore, most of the hardware that performs the vector phase rotation process and the Arctan calculation process can be shared.

Next, vector phase rotation processing and Arc
The advantage of using the CORDIC algorithm for the tan calculation process will be described. The first advantage is that the size of the table to be prepared is proportional to the required bit precision. If you want the accuracy of the calculation to be 16 bits, the number of vectors in the prepared table is 1.
Only 7 required. The second advantage is that the complex multiplication process for vector phase rotation can be realized only by the addition / subtraction process and the shift process. As a result, the hardware scale required for the processing becomes considerably smaller than that in the case where the multiplication processing is performed.

FIG. 2 shows Ar in the OFDM communication apparatus 1.
It is a functional block diagram which shows a part of internal structure of the ctan calculation vector phase rotator 104. The entire circuit of the Arctan calculation vector phase rotator 104 has a plurality of circuits shown in FIG. 2, and each circuit is connected in a pipeline manner. The entire circuit of the Arctan calculation vector phase rotator 104 includes adder / subtractors 611, 612 and
The output of 613 may be fed back to each input.

In FIG. 2, the Arctan calculation vector phase rotator 104 includes a code selection unit 604, an angle change constant table unit 605, and bit shift units 606 and 607.
, Sign addition units 608, 609, 610, and adder / subtractor 6
11,612,613.

The code selection unit 604 changes the code to be output while referring to the Y coordinate of the vector (X (n), Y (n)) during the Arctan calculation process, and changes the code to be output by the code addition unit 608. 609 and 610. On the other hand, in the vector rotation processing, the code selection unit 604
Is the phase correction amount from the adder 106 and the vector (X
(N), Y (n)) while comparing the phase to be output, the code to be output is changed, and the code is added to the code adding units 608, 60.
Input to 9,610.

The angle change constant table section 605 contains Table 1
The complex vector group as shown in is stored.

The total rotation amount Z (n) in the rotation processing up to the nth stage is input to one input of the adder / subtractor 611. The total rotation amount Z (n) is calculated by a total rotation amount calculation unit (not shown). The other input of the adder / subtractor 611 is connected to the angle change constant table unit 6
The angle to which the sign adding unit 608 adds a sign to the multiplier vector angle stored in 05 is input. The adder / subtractor 611 adds the sign addition unit 6 to the rotation amount Z (n) up to the nth stage.
The angle from 08 is added and the total rotation amount up to the (n + 1) th stage is output.

The X coordinate of the input vector in the rotation processing up to the nth stage is input to one input of the adder / subtractor 612. The other input of the adder / subtractor 612 is supplied with Y of the input vector in the rotation processing up to the nth stage which has been bit-shifted.
A value obtained by adding a code to the coordinate by the code adding unit 609 is input. The adder / subtractor 612 adds the two input values and outputs it as the X coordinate up to the (n + 1) th stage.

The Y coordinate of the input vector in the rotation processing up to the nth stage is input to one input of the adder / subtractor 613. The other input of the adder / subtractor 613 is the X of the input vector in the rotation processing up to the n-th stage that has been bit-shifted.
A value obtained by adding a code to the coordinate by the code adding unit 610 is input. The adder / subtractor 613 adds the two input values and outputs it as the Y coordinate up to the (n + 1) th stage.

The bit shift units 606 and 607, the code addition units 609 and 610, and the adder / subtractors 612 and 613 realize complex multiplication of the complex vector to be multiplied at the (n + 1) th stage. The complex vector to be multiplied at the (n + 1) th stage is stored in the angle change constant table unit 605, and the arrow from the angle change constant table unit 605 to the bit shift units 606 and 607 is omitted in the drawing.

In the Arctan calculation process, Arct
The an-calculation vector phase rotator 104 outputs the output (total rotation amount) of the adder / subtractor 611 at the final stage as the phase of the input vector, and inputs it to the phase correction information estimator 105. A block for inputting the total rotation amount to the phase correction information estimator 105 is omitted in the drawing.

In the vector phase rotation processing, Arct
The an-calculation vector phase rotator 104 includes an adder / subtractor 61 at the final stage that rotates the phase based on the phase correction amount.
The outputs of 2, 613 are unitized (multiplied by the inverse of the amplitude multiple), output as a vector after phase rotation, and input to the demodulation error corrector 107. A block for unitizing a vector is omitted in the drawing.

FIG. 3 is a functional block diagram showing the configuration of the phase correction information estimator 105. 3, the phase correction information estimator 105 includes a multipath fading phase shift amount storage unit 701, a subtractor 702, a phase shift amount calculation circuit 703, switch circuits 704 and 705, a control circuit 706, and a transmission line. The phase error estimation unit 707 and the pilot signal phase shift amount calculation unit 708 are included. The phase correction information estimator 105 removes the phase shift estimation amount due to multipath fading from the phase shift amount of the pilot signal vector, and calculates the phase shift estimation value for the data vector due to factors other than multipath fading.

At the start of communication, the control circuit 706 controls the phase error estimation symbol 208 while it is being sent.
While closing the switch circuit 704, the switch circuit 70
Open 5. Further, the control circuit 706 closes the switch circuit 705 and opens the switch circuit 704 while the data vector is being sent and the pilot signal is being sent. The control circuit 706 opens both the switch circuits 704 and 705 during the other period.

While the switch circuit 704 is closed, the phase of the known signal vector is input to the transmission line phase error estimation unit 707. The transmission line phase error estimation unit 707 compares the phase of the input known signal vector with the known phase of the known signal vector stored in advance to calculate a phase shift estimation value due to multipath fading, and calculates the multipath fading phase. It is stored in the shift amount storage unit 701.

While the switch circuit 705 is closed, the phase of the pilot signal vector is input to the pilot signal phase shift amount calculating section 708. The pilot signal phase shift amount calculation unit 708 compares the phase of the input pilot signal vector with the known phase of the pilot signal vector stored in advance, and calculates the phase shift amount of the pilot signal.

The subtractor 702 calculates the multipath fading phase shift amount storage unit 70 based on the pilot signal phase shift amount output from the pilot signal phase shift amount calculation unit 708.
The phase shift estimated value due to the multipath fading stored in 1 is subtracted and input to the phase shift amount calculation circuit 703. As a result, the subtractor 702 outputs the phase shift amount obtained by removing the estimated phase shift value due to multipath fading from the pilot signal phase shift amount.

FIG. 4 is a functional block diagram showing the internal structure of the phase shift amount calculation circuit 703. In FIG. 4, the phase shift amount calculation circuit 703 includes a subtractor 801, a first averaging circuit 802, a delay device 803, a carrier number generator 804, a multiplier 805, a cumulative adder 806, Two
Averaging circuit 807 and adder 808. As described with reference to FIG. 20, the phase shift amount calculation circuit 703 linearly interpolates the phase shift amount between pilot signals for which the phase shift amount obtained by removing the phase shift amount due to multipath fading has been obtained. By performing the above, the phase shift amount of the data vector arranged between the pilot signals is estimated, and the phase shift estimated value due to factors other than multipath fading is calculated.

The delay device 803 delays the phase shift amount of the pilot signal vector from the subtractor 702 to delay the subtractor 8
Enter 01. The subtractor 801 includes a phase shift amount of the pilot signal vector from the subtractor 702 and a delay unit 80.
The phase shift amount difference between the adjacent pilot signals is calculated based on the delayed phase shift amount of the pilot signal vector input from No. 3, and the calculation result is input to the first averaging circuit 802.

The first averaging circuit 802 calculates the average value of the amount of phase shift per one data symbol and calculates the multiplier 805.
To enter. The average value of the phase shift amount becomes the slope value of the phase shift amount estimation line shown in FIG.

Cumulative adder 806 cumulatively adds the phase shift amounts of all pilot signal vectors in one data symbol and inputs the result to second averaging circuit 807. The second averaging circuit 807 averages the cumulatively added phase shift amounts and inputs the averaged amount to the adder 808. This average value becomes the intercept value of the phase shift amount estimation line shown in FIG.

As described above, since the slope value and the intercept value are obtained as the parameters of the phase shift estimation line, the phase shift amount calculation circuit 703 next calculates the phase shift amount for each data vector. . This is shown in FIG.
The carrier number generator 8
It is obtained by substituting the numbers of the respective transmission / reception signal vectors output by 04.

The multiplier 805 multiplies the gradient value of the phase shift amount estimation line by the vector number generated by the carrier number generator 804. The adder 808 adds the multiplication result of the multiplier 805 and the intercept value of the straight line from the second averaging circuit 807, and outputs the phase shift amount for each data vector. This amount of phase shift becomes an estimated value of phase shift due to factors other than multipath fading.

The adder 106 includes a phase shift amount calculation circuit 70.
By adding the phase shift estimation value due to factors other than multipath fading output from No. 3 and the phase shift estimation value due to multipath fading stored in the multipath fading phase shift amount storage unit 701,
The correction value of the phase shift due to the first to fourth factors is input to the Arctan calculation vector phase rotator 104 as the total phase correction value.

Arctan Calculation Vector Phase Rotator 10
4 rotates the vector from the FFT circuit 103 by the total phase correction value input from the adder 106, and inputs the vector in which the phase shift due to the first to fourth factors is corrected to the demodulation error corrector 107.

As described above, in the first embodiment, the phase shift estimation value due to multipath fading is calculated, and the phase shift estimation value due to other than multipath fading is calculated, and these are summed to obtain the first to first The total phase correction value due to the fourth factor is calculated. Then, the vector of the received signal is phase-corrected at one time based on the total phase correction value. Therefore, compared with the conventional case where the phase rotation processing is performed twice, the hardware It is possible to reduce the scale and the amount of calculation.

In the above embodiment, the known phase,
Although the transmission path phase error estimation unit 707 and the pilot signal phase shift amount calculation unit 708 are used when comparing the phase of the received vector, a polarity inverter may be used here. Here, for example, at the time of transmission, a known vector P having a phase of 30 degrees is
If it is received as a vector Q having a phase of 70 degrees at the time of reception, the difference between the phases of P and Q becomes the amount of phase shift due to the transmission path. R (phase: -30 degrees) and vector Q are complex-multiplied to obtain a vector S having a phase of 40 degrees. The polarity inverter outputs the value of 40 degrees as the phase shift amount due to the transmission path.

The phase correction information estimator 105 estimates the linear function expression (phase deviation amount estimation line) of the phase deviation amount from the correlation between the pilot signal vectors.
A linear function expression of the phase shift amount may be estimated. Specifically, the N pilot signal vectors are set to PC (0) to PC (0).
If (N-1), the phase correction information estimator 105
Vector A = PC (0) + PC (1) + PC (2) + ...
+ PC (N-1), vector B = PC (0) × PC
(1) + PC (1) x PC (2) + PC (2) x PC
(3) + ... + PC (N−2) × PC (N−1) (multiplier side is a conjugate complex), two vectors are created, and these 2
Arctan calculation vector phase rotator 1
Input to 04 to calculate Arctan. Arctan
Using the angle A and the angle B obtained by the calculation, the phase correction information estimator 105 estimates the linear equation as θ = A × f + B. According to this method, the weight of the phase of the pilot signal, which has become unreliable due to the reduction of the amplitude due to multipath fading, can be reduced, and therefore the reliability of the estimated value of the phase correction amount obtained from the pilot signal It is possible to improve.

The phase correction information estimator 105 uses, as a method of estimating the phase correction amount of the transmission / reception signal vector using the pilot signal vector, other than linear approximation between the pilot signal vectors described above, an arbitrary pilot signal vector A method of using the phase correction amount as it is may be used.

Although the Arctan calculation process and the vector phase rotation process are performed by one Arctan calculation vector phase rotator 104 here, a separate Arctan calculator (phase calculator) that performs each process is separately provided. And a phase rotator may be used.

(Second Embodiment) With respect to the overall configuration of an OFDM communication apparatus according to the second embodiment of the present invention,
Since this is the same as the case of the above embodiment, FIG. 1 is also referred to in this embodiment. In the OFDM communication apparatus according to the second embodiment, as the method of estimating the phase correction information, the phase correction information obtained from the OFDM symbol to be demodulated and the OFDM demodulated in a certain section in the past are used.
A weighted average value with the phase correction information obtained from the symbol is used.

FIG. 5 is a functional block diagram showing the configuration of the phase shift amount calculation circuit in the OFDM communication apparatus according to the second embodiment. In FIG. 5, the phase shift amount calculation circuit includes a subtractor 801, a first averaging circuit 802, a delay device 803, a carrier number generator 804, and a multiplier 80.
5, a cumulative adder 806, and a second averaging circuit 807.
A third averaging circuit 901 and a first multistage delay device 90.
2, the fourth averaging circuit 903, and the second multistage delay device 9
04 and. In FIG. 5, portions having the same functions as those of the phase shift amount calculation circuit 703 according to the first embodiment are designated by the same reference numerals and description thereof will be omitted.

The first multi-stage delay device 902 delays the inclination value of the phase shift straight line obtained from the past several symbols and inputs it to the third averaging circuit 901. Third averaging circuit 9
01 calculates a weighted average value of the slope values of the delayed straight line, and outputs the result as the slope value of the phase shift amount estimation straight line of the demodulation target symbol. As a weighting method,
By increasing the weight value for the slope value obtained from the pilot signal adjacent to the OFDM symbol currently being demodulated, the influence of large noise in the past can be reduced.

The second multistage delay device 904 delays the intercept value of the phase shift straight line obtained from the past several symbols and inputs it to the fourth averaging circuit 903. Fourth averaging circuit 9
03 calculates a weighted average value of the intercept values of the delayed straight line, and outputs the result as the intercept value of the phase shift amount estimation straight line of the demodulation symmetrical symbol. The weight value of the intercept value is the same as the slope value.

That is, in estimating the phase shift amount estimating line, the phase shift amount calculating circuit calculates the average value of the slope value and the intercept value of the phase shift amount estimating line obtained from several past symbols as the demodulation target symbol. It is used as the slope value and intercept value of the phase shift amount estimation line.

As described above, according to the second embodiment, the phase correction amount for several symbols obtained from past (before demodulation target) OFDM symbols and the phase correction amount obtained from the OFDM symbol to be demodulated Since the total phase correction value is determined by taking the average of and, the burst noise is present in the communication in time.
Even if the reliability of the phase correction amount of the FDM symbol is significantly reduced, it is possible to perform the phase correction processing with high reliability.

(Third Embodiment) FIG. 6 shows a third embodiment of the present invention.
3 is a functional block diagram showing an overall configuration of an OFDM communication device 3 according to the embodiment of FIG. In FIG. 6, the OFDM communication device 3 includes an antenna 101, a receiving circuit 102, and an FFT.
The circuit 103, the control circuit 1006, and the switch circuit 10
02, 1003, 1004, memory 1001, Ar
It includes a ctan calculation vector phase rotator 104, a phase correction information estimator 1005, an adder 106, and a demodulation error corrector 107. In FIG. 6, parts having the same functions as those of the OFDM communication device 1 according to the first embodiment are designated by the same reference numerals, and description thereof will be omitted. Also,
The phase correction information estimator 1005 has the same function as the phase correction information estimator according to the second embodiment.

The OFDM communication apparatus 3 calculates the total phase correction value from the total phase correction value obtained from the demodulation target OFDM symbol and from any OFDM symbol before and after the demodulation target. A weighted average value with the total phase correction value is used.

The control circuit 1006 outputs the pilot signal vector from the FFT circuit 103 at the timing
The switch circuit 1002 is connected and the switch circuit 1003 is disconnected. The control circuit 1006, at the timing when the data vector is output from the FFT circuit 103,
The switch circuit 1002 is disconnected, the switch circuit 1003 is connected, and the data vector is stored in the memory 1001. The control circuit 1006 connects the switch circuit 1004 to the memory 1001 after an appropriate delay time has elapsed.
The data vector stored in (1) is sent to the Arctan calculation vector phase rotator 104. This switch switching operation is performed based on preset timing.

By the switch switching operation of the control circuit 1006 as described above, the pilot signal vector for the data vector included in the OFDM symbol is input to the Arctan calculation vector phase rotator 104 in advance. Therefore, the phase correction information estimator 1
005 calculates in advance the phase correction information of the data vector to be demodulated (the phase shift estimated value due to multipath fading and the phase shift estimated value due to a factor other than multipath fading), and the phase correction information is calculated before the demodulation target. And a weighted average with the total phase correction value obtained from any OFDM symbol after the demodulation target.

As described above, in the third embodiment, in estimating the phase correction information, the total phase correction value is calculated using the phase correction information obtained by the pilot signals included in the OFDM symbols after the demodulation target. Therefore, in addition to the effect of the second embodiment, it is possible to more accurately follow the total phase correction value for the phase shift amount that changes with time, and as a result, the phase correction with higher reliability can be achieved. It becomes possible to perform processing.

The control circuit 1006 has the memory 1001.
Timing of sending / receiving the transmission / reception signal vector stored in the Arctan calculation vector phase rotator 104, and the first multistage delay device 902 and the second multistage delay device 90
By adjusting the delay time by 4, it is possible to adjust the time range of the OFDM symbol used to estimate the phase correction information.

The configuration of the third embodiment may be added to the configuration of the first embodiment. In this case, the packet signal vector included in the OFDM symbol of demodulation symmetry precedes and is input to the Arctan calculation vector phase rotator 104, and the phase correction information estimator 105
However, the phase correction information is calculated in advance. As a result, the adder 106 inputs the latest total phase correction value relating to the data vector to be demodulated to the Arctan calculation vector phase rotator 104, so that it is possible to perform highly reliable phase correction processing. .

(Fourth Embodiment) FIG. 7 shows a fourth embodiment of the present invention.
3 is a functional block diagram showing an overall configuration of an OFDM communication device 4 according to the embodiment of FIG. In FIG. 7, the OFDM communication device 4 includes an antenna 101, a receiving circuit 102, and an FFT.
The circuit 103, the control circuit 1006, and the switch circuit 10
02, 1003, 1004, memory 1001, Ar
a ctan calculation vector phase rotator 104, a phase correction information estimator 1202, a received signal state discriminator 1201,
It includes an adder 106 and a demodulation error corrector 107. In FIG. 7, the OFDM communication device 3 according to the third embodiment
Portions having the same functions as those of the above are given the same reference numerals, and description thereof will be omitted.

In addition to the function of the OFDM communication device 3 according to the third embodiment, the OFDM communication device 4 is based on the state discrimination of the received signal and the weight value for calculating the slope value and intercept value of the phase shift amount estimation line. The function to change is added.

FIG. 8 is a functional block diagram showing the structure of the phase correction information estimator 1202. In FIG. 8, parts having the same functions as those of the phase correction information estimator 105 according to the first embodiment are designated by the same reference numerals, and description thereof will be omitted. In FIG. 8, the phase correction information estimator 120
Reference numeral 2 denotes a multipath fading phase shift amount storage unit 701.
A subtractor 702, a phase shift amount calculation circuit 1301,
It includes switch circuits 704 and 705, a control circuit 706, a transmission line phase error estimation unit 707, and a pilot signal phase shift amount calculation unit 708.

FIG. 9 is a functional block diagram showing the structure of the phase shift amount calculation circuit 1301. In FIG. 9, the phase shift amount calculation circuit 1301 includes a subtractor 801, a first averaging circuit 802, a delay device 803, a carrier number generator 804, a multiplier 805, a cumulative adder 806, Two
Averaging circuit 807, third averaging circuit 901, first multi-stage delay device 902, fourth averaging circuit 903,
It has a second multistage delay device 904 and a weight value generator 1302. In FIG. 9, portions having the same functions as those of the phase shift amount arithmetic circuit according to the second embodiment (see FIG. 5) are designated by the same reference numerals, and description thereof will be omitted.

The received signal state discriminator 1201 detects whether or not the state related to the reliability of the received signal has changed, such as whether or not the power of the received signal from the receiving circuit 102 has weakened, and If it has changed, a control signal for changing the weight value to the weight value generator 1302,
The OFDM symbol number, which is the symmetry of weight value change, is sent.

FIG. 10 is a functional block diagram showing the structure of the received signal state discriminator 1201. In FIG. 10, the received signal state discriminator 1201 is a square circuit 1501, 15
02, an adder 1503, and a threshold value generation circuit 1504
And a comparator 1505. Square circuit 1501,15
By adding the output of 02 in the adder 1503,
Calculate the power of the received signal. Comparator 1505 compares the power of the received signal given from adder 1503 with the threshold value generated by threshold value generation circuit 1504, and if the power of the received signal is below the threshold value, the reception signal is being received. The phase correction information estimator 12 judges that the reliability of the signal is low.
A control signal for changing the weight value and the OFDM symbol number which is the target thereof are sent to the weight value generator 1302 in 02.

The weight value generator 1302 generates a small weight value for the notified OFDM symbol number according to the control signal for changing the weight value from the received signal state discriminator 1201, and the third averaging circuit. 901 and the fourth averaging circuit 903.

The third averaging circuit 901 and the fourth averaging circuit 903 use the weight value generator 1 to calculate the phase weight value of the pilot signal vector in the vicinity of the OFDM symbol number.
It is determined based on the weight value input from 302, and the weighted average of the parameters (slope value and intercept value) of the phase shift amount estimation line is calculated.

As described above, in the fourth embodiment, in estimating the slope value and the intercept value of the phase shift amount estimation straight line,
Since a weight value for a low reliability is suppressed from a plurality of slope values and intercept values obtained from a series of pilot signal vectors, highly reliable phase correction information can be extracted. Therefore, it is possible to provide the OFDM communication device capable of improving the accuracy of the phase correction information and improving the effect of the phase correction processing.

In the fourth embodiment, the weight of the parameter of the phase shift amount estimation line is changed. However, by setting the individual weight for each phase shift amount of the pilot signal vector, the reliability can be improved. The phase correction amount may be estimated by extracting the phase shift amount obtained from the highly reliable pilot signal.

(Fifth Embodiment) FIG. 11 is a functional block diagram showing the structure of an OFDM communication apparatus 5 according to a fifth embodiment of the present invention. 11, the OFDM communication device 5 includes an antenna 101, a receiving circuit 102, and an FFT.
The circuit 103, the control circuit 507, and the switch circuit 50
2, 503, 504, memory 505, and subtractor 506
, Arctan calculation vector phase rotator 104, phase correction information estimator 105a, adder 106, demodulation error corrector 107, and re-encoder 501. Figure 1
1, the OFDM communication device 1 according to the first embodiment
Portions having the same functions as those of the above are given the same reference numerals, and description thereof will be omitted.

The OFDM communication apparatus 5 is characterized by updating the phase shift estimation value due to multipath fading to the latest value.

The control circuit 507 includes a switch circuit 502,
It controls the opening and closing of 503 and 504, stores the phase of the vector in the memory 505, and controls the operation of the subtractor 506. The opening and closing of the switch is performed according to a predetermined rule. In FIG. 11, the arrow from the control circuit 507 to the switch circuits 503 and 504 is omitted.

Control circuit 507 causes memory 505 to periodically store the phase of the vector in the received OFDM symbol (for example, the phase of the pilot signal vector) at predetermined intervals. The re-encoder 501 re-encodes the vector obtained when the symbol whose phase is stored in the memory 505 is input to the demodulation error corrector 107.
As a result, the re-encoder 501 generates an OFDM symbol for collating the vector phase stored in the memory 505.

When re-encoding is performed by the re-encoder 501, the control circuit 507 closes the switch circuit 503 and switches the switch circuit 502 to the subtractor 506 side. As a result, the Arctan calculation vector phase rotator 104 calculates the phase of the OFDM symbol by the Arctan calculation and outputs it.

In response to this, the control circuit 507 causes the subtracter 506 to determine the phase of the OFDM symbol and the memory 50.
The difference with the phase stored in 5 is calculated. The calculated phase difference is input to the phase correction information estimator 105a. This phase difference is stored in the multipath fading phase shift amount storage unit of the phase correction information estimator 105a. At this time, unlike the case of the first embodiment,
The phase correction information estimator 105a functions so that the phase difference is directly stored in the multipath fading phase shift amount storage section without passing through the transmission path phase error estimation section. The calculated phase difference indicates the estimated phase shift value due to the latest multipath fading.

As described above, in the fifth embodiment, the phase shift estimated value due to multipath fading is periodically recalculated, so that it can be always updated to the latest value. Therefore, the effect of the phase correction processing can be enhanced.

The phase stored in the memory may be a phase other than the pilot signal vector.

Hereinafter, OFDs according to the first to fifth embodiments will be described.
An embodiment for implementing the M communication device by software will be described. As hardware of the OFDM communication device when the above functions are implemented by software, a semiconductor for digital signal processing (Digital Signal) capable of executing a program described below is used.
Processor: DSP) may be used, or may be realized by a CPU that reads and executes the program from a recording medium that stores the program described below. Further, the OFDM communication device in this case is provided with a memory for storing various data being processed. Since such a hardware configuration is known, a detailed description thereof will be omitted, and the operation of the OFDM communication apparatus when the program is executed will be described below with reference to a flowchart.

(Sixth Embodiment) FIG. 12 is a flowchart showing the operation of the OFDM communication apparatus according to the sixth embodiment of the present invention. Hereinafter, referring to FIG. 12, the sixth
The operation of the OFDM communication apparatus according to the embodiment will be described.

First, the OFDM communication device receives an OFDM signal (step S101). Next, the OFDM communication device performs a FFT operation on the received OFDM signal to obtain a subcarrier signal vector assigned to each subcarrier (step S10).
2).

Next, the OFDM communication apparatus calculates Arctan of the known signal vector and the pilot signal vector, and obtains their respective phases (step S103).
Next, the OFDM communication device subtracts the known phase of the known signal vector stored in the memory in advance from the phase of the received known signal vector to calculate a phase shift estimated value K due to multipath fading and stores it in the memory. It is stored (step S104).

Next, the OFDM communication apparatus subtracts the known phase of the pilot signal vector stored in the memory in advance from the phase of the received pilot signal vector to calculate the phase shift amount L of the pilot signal vector ( Step S105). Next, the OFDM communication apparatus calculates a value L-K by subtracting the estimated phase shift value K due to multipath fading calculated in step S104 from the phase shift amount L of the pilot signal vector calculated in step S105 (step S106). ). As a result, the phase shift amount is calculated by removing the estimated value of the phase shift due to multipath fading from the pilot signal phase shift amount.

Next, the OFDM communication apparatus carries out step S1.
Based on the phase shift amount calculated in 06, the average slope value and the average intercept value of the phase shift estimation line are obtained, the formula of the phase shift estimation line is obtained, and each OFDM symbol number is used. An estimated value of phase shift due to factors other than multipath fading is obtained (step S107). For a more detailed method of calculating the average slope value and the average intercept value, see the phase shift amount calculation circuit 70 in the first embodiment.
Since it has been described in Section 3, detailed description is omitted here.

Next, the OFDM communication device sums the phase shift estimation amount K due to multipath fading stored in the memory and the phase shift estimation value due to factors other than multipath fading to calculate the total phase correction value. (Step S108).

Next, the OFDM communication apparatus carries out step S1.
Based on the total phase correction value obtained in 08, the phases of all input signal vectors are corrected (step S109). afterwards,
The OFDM communication device performs error correction and demodulation processing based on all phase-corrected input signal vectors (step S110), and ends the processing.

When the OFDM communication apparatus executes the program for performing such processing, the OFDM communication apparatus having the same effect as the OFDM communication apparatus according to the first embodiment is provided.

(Seventh Embodiment) FIG. 13 is a flowchart showing the operation of the OFDM communication apparatus according to the seventh embodiment of the present invention. In FIG. 13, the same step numbers are used for operations similar to those in the OFDM communication apparatus according to the sixth embodiment, and description thereof will be omitted. Hereinafter, the operation of the OFDM communication apparatus according to the seventh embodiment will be described with reference to FIG.

After steps S101 to S106, the OFD
The M communication device obtains an average slope value and an average intercept value of the phase shift amount estimation line, and stores them for a certain period in the memory (step S201). Next, the OFDM communication apparatus calculates the average gradient value from the past to the present before the certain period by using the average gradient value for the certain period stored in the memory (step S202). Next, OF
The DM communication device uses the average intercept value for a certain period stored in the memory to calculate the average intercept value from the past to the present before the certain period (step S20).
3). Next, the OFDM communication apparatus obtains the equation of the phase shift amount estimation line using the average slope value and the intercept value from the past to the present before a certain period, and estimates the phase shift value due to factors other than multipath fading. Is calculated (step S204).

After that, the OFDM communication device sums the phase shift estimation value due to multipath fading stored in the memory and the phase shift estimation value due to factors other than multipath fading to obtain a total phase correction value, Is corrected and demodulation processing is performed (steps S108 to S11).
0), the process ends.

When the OFDM communication apparatus executes the program for performing such processing, the O according to the second embodiment
An OFDM communication device having the same effect as the FDM communication device is provided.

(Eighth Embodiment) FIG. 14 is a flowchart showing the operation of the OFDM communication apparatus according to the eighth embodiment of the present invention. In FIG. 14, the same step numbers are used for operations similar to those in the OFDM communication apparatuses according to the sixth and seventh embodiments, and description thereof will be omitted. Hereinafter, the operation of the OFDM communication apparatus according to the eighth embodiment will be described with reference to FIG.

After receiving the input signal and performing the FFT operation (steps S101 and S102), the OFDM communication apparatus stores only the data vector in the current input signal in the memory and precedes the pilot signal vector with the data signal vector. (Step S301).

Next, the OFDM communication apparatus calculates the phase shift estimation amount due to multipath fading, and subtracts the phase shift estimation value due to multipath fading from the phase shift amount of the preceding pilot signal to calculate the phase shift amount. (Steps S103 to S106).

Next, the average slope value and the average intercept value are obtained from the preceding pilot signal vector and stored in the memory (step S302).

Next, the OFDM communication apparatus obtains the total phase correction amount (steps S202 to S204, S108).
Next, the OFDM communication device reads the data vector stored in the memory, rotates the vector based on the total phase correction amount, and corrects the phase (step S30).
3) Perform demodulation error correction processing (step S11)
0), the process ends.

When the OFDM communication apparatus executes the program for performing such processing, the OFDM communication apparatus having the same effect as the OFDM communication apparatus according to the third embodiment is provided.

(Ninth Embodiment) FIG. 15 is a flowchart showing the operation of the OFDM communication apparatus according to the ninth embodiment of the present invention. In FIG. 15, the same step numbers are used for operations similar to those in the OFDM communication apparatus according to the eighth embodiment, and description thereof will be omitted. The operation of the OFDM communication apparatus according to the ninth embodiment will be described below with reference to FIG.

After receiving the OFDM signal (step S10)
1), the OFDM communication device detects the reception status from the signal reception level and calculates the weight value C according to the reception status (step S401).

Next, the OFDM communication apparatus, after the FFT calculation (step S102), calculates the phase shift amount of the packet signal by subtracting the phase shift estimated value due to multipath fading and the phase shift estimated value due to multipath fading, An average value of the slope value and the intercept value of the phase shift estimation line is calculated (steps S301, S103 to S1).
06, S302).

Next, the OFDM communication apparatus uses the weight value C to calculate the weighted average of the slope values from the past to the present (step S402). Next, the OFDM communication device uses the weight value C to calculate the weighted average of the intercept values from the past to the present (step S403).

After that, the OFDM communication apparatus obtains a phase shift estimated value due to factors other than multipath fading,
The total phase correction value is obtained, the phase of the data vector is corrected, and demodulation processing is performed (steps S204, S303, S1.
01), the processing is ended.

When the OFDM communication apparatus executes the program for performing such processing, the OFDM communication apparatus having the same effect as the OFDM communication apparatus according to the fourth embodiment is provided.

(Tenth Embodiment) FIG. 16 is a flowchart showing the operation of the OFDM communication apparatus according to the tenth embodiment of the present invention. In FIG. 16, the same step numbers are used for operations similar to those in the OFDM communication apparatus according to the sixth embodiment, and description thereof will be omitted. The operation of the OFDM communication apparatus according to the tenth embodiment will be described below with reference to FIG.

After receiving the OFDM signal, the FFT calculation processing, and the phase calculation (steps S101 to S103), the OFDM signal
The communication device determines whether or not a predetermined timing has arrived (step S501).

When the predetermined timing has come, O
The FDM communication device stores the phase of the pilot signal vector at the timing in the memory (step S50).
2), the operation proceeds to step S104 and thereafter. On the other hand, when the predetermined timing has not come, the OFDM communication apparatus proceeds to the operation of step S104 and thereafter.

In steps S104 to S109, O
The FDM communication device calculates the total phase correction value. Then O
The FDM communication device determines whether or not a predetermined timing has arrived (step S503). The timing here is a timing different from the timing in step S501, and is the timing for recalculating the phase shift estimated value due to multipath fading. If not, the OFDM communication device performs demodulation error correction processing (step S110).

On the other hand, if it has arrived, the OFDM communication apparatus performs demodulation error correction processing (step S110: on the drawing,
The processing arrow is omitted) and the data is re-encoded (step S504). Then OFDM
The communication device obtains the phase of the pilot signal vector in the re-encoded data by Arctan calculation, compares it with the phase of the pilot signal vector stored in the memory, and recalculates the phase shift estimation amount K due to multipath fading. (Step S505). After that, OF
The DM communication device updates the estimated phase shift amount K due to multipath fading stored in the memory, and returns to the operation of step S101.

When the OFDM communication apparatus executes the program for performing such processing, the OFDM communication apparatus having the same effect as the OFDM communication apparatus according to the fifth embodiment is provided.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing a configuration of an OFDM communication device 1 according to a first embodiment of the present invention.

FIG. 2 is a functional block diagram showing a part of an internal configuration of an Arctan calculation vector phase rotator 104 in the OFDM communication apparatus 1.

FIG. 3 is a functional block diagram showing a configuration of a phase correction information estimator 105.

FIG. 4 is a functional block diagram showing an internal configuration of a phase shift amount calculation circuit 703.

FIG. 5 is a functional block diagram showing a configuration of a phase shift amount calculation circuit in an OFDM communication apparatus according to a second embodiment of the present invention.

FIG. 6 is a functional block diagram showing an overall configuration of an OFDM communication device 3 according to a third embodiment of the present invention.

FIG. 7 is a functional block diagram showing an overall configuration of an OFDM communication device 4 according to a fourth embodiment of the present invention.

FIG. 8 is a functional block diagram showing a configuration of a phase correction information estimator 1202.

9 is a functional block diagram showing the configuration of a phase shift amount calculation circuit 1301. FIG.

FIG. 10 is a functional block diagram showing the configuration of a received signal state discriminator 1201.

FIG. 11 is a functional block diagram showing a configuration of an OFDM communication device 5 according to a fifth embodiment of the present invention.

FIG. 12 is a flowchart showing an operation of the OFDM communication device according to the sixth embodiment of the present invention.

FIG. 13 is a flowchart showing an operation of the OFDM communication device according to the seventh embodiment of the present invention.

FIG. 14 is a flowchart showing an operation of the OFDM communication apparatus according to the eighth embodiment of the present invention.

FIG. 15 is a flowchart showing an operation of the OFDM communication device according to the ninth embodiment of the present invention.

FIG. 16 is a flowchart showing an operation of the OFDM communication device according to the tenth embodiment of the present invention.

FIG. 17 is a diagram for explaining a phase error due to second to fourth factors.

FIG. 18 is a block diagram showing a configuration of a conventional OFDM communication apparatus.

FIG. 19 is a diagram showing a structure of a communication packet.

FIG. 20 is a diagram showing a distribution of a phase shift amount of a signal when viewed in the frequency axis direction.

[Explanation of symbols]

1, 3, 4, 5 OFDM communication apparatus 101 antenna 102 receiving circuit 103 FFT circuit 104 Arctan calculation vector phase rotator 105, 1005, 1202, 105a phase correction information estimator 106, 1503 adder 107 demodulation error corrector 604 code selection Unit 605 Angle change constant table unit 606, 607 Bit shift unit 608, 609, 610 Sign addition unit 611, 612, 613 Adder / subtractor 701 Multipath fading phase shift amount storage unit 702, 506 Subtractor 703, 1301 Phase shift amount calculation circuit 704, 705, 1003 to 1004, 502 to 504
Switch circuits 706, 1006, 505 Control circuit 707 Transmission line phase error estimation unit 708 Pilot signal phase shift amount calculation unit 801 Subtractor 802 First averaging circuit 803 Delay unit 804 Carrier number generator 805 Multiplier 806 Cumulative adder 807 Second averaging circuit 808 Adder 901 Third averaging circuit 902 First multi-stage delay device 903 Fourth averaging circuit 904 Second multi-stage delay device 1001 Memory 1201 Received signal state discriminator 1302 Weight value generator 1501 and 1502 Square circuit 1504 Threshold generation circuit 1505 Comparator 501 Re-encoder 208 Phase error estimation symbol 209 due to transmission path Pilot signal

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Masahiro Ishii             1006 Kadoma, Kadoma-shi, Osaka Matsushita Electric             Sangyo Co., Ltd. F-term (reference) 5K022 DD01 DD33

Claims (29)

[Claims] 1. The received OFDM signal is subjected to a discrete Fourier transform (FFT).
an OFDM communication device for correcting the phase of the vector signal obtained by performing the processing, and a phase calculation unit that calculates the phase of the vector signal, and an OFDM communication device that is included in the OFDM signal calculated by the phase calculation unit. By comparing each with the known phase for the phase of each vector signal for multiple known signals
For each of the multiple phase shift factors that exist, a phase correction information estimation unit that outputs each phase shift estimation value and a plurality of the phase shift estimation values that the phase correction information estimation unit outputs are summed, and total phase correction is performed. And a vector phase rotation unit that rotates the phase of the vector signal by the total phase correction value output by the summation unit and outputs a vector signal whose phase has been corrected.
DM communication device. 2. The phase correction information estimator is configured to estimate a phase shift obtained from a known signal included in an OFDM symbol to be demodulated and a phase obtained from a known signal included in an OFDM symbol other than the demodulated object. The OFDM communication device according to claim 1, wherein the respective phase shift estimation values are output by obtaining a weighted average with the shift estimation value. 3. The phase correction information estimator increases the weight value for an estimated phase shift value obtained from an OFDM symbol adjacent to an OFDM symbol currently being demodulated, and obtains a weighted average. The OFDM communication device according to claim 2, wherein 4. The phase shift estimation value obtained from a known signal included in an OFDM symbol other than the demodulation target is an OFD for a certain section before the OFDM symbol to be demodulation target.
The OFDM communication apparatus according to claim 3, wherein the OFDM communication apparatus is a phase shift estimation value obtained from M symbols. 5. The phase shift estimation value obtained from a known signal included in an OFDM symbol other than the demodulation target is a phase shift estimation value obtained from all OFDM symbols before the OFDM symbol to be demodulated. The OFDM communication device according to claim 3, characterized in that. 6. The phase shift estimation value obtained from a known signal included in an OFDM symbol other than the demodulation target is a phase shift estimation value obtained from an OFDM symbol after the OFDM symbol to be demodulated. The OFDM communication device according to claim 3, wherein 7. A phase shift estimated value obtained from a known signal included in an OFDM symbol other than the demodulation target is a phase shift estimated value obtained from an OFDM symbol before the OFDM symbol to be demodulated and a demodulation target. OFD
4. The O according to claim 3, which is an estimated value of phase shift obtained from OFDM symbols after M symbols.
FDM communication device. 8. A reception state discrimination unit for discriminating a reception state of the received OFDM signal is further provided, wherein the reception state discrimination unit is in accordance with a reception state of an OFDM symbol included in the received OFDM signal, The OFDM communication apparatus according to claim 2, wherein a weight value of a weighted average by the phase correction information estimation unit is changed. 9. A re-encoding unit that periodically re-encodes the demodulated OFDM symbol, wherein the phase calculation unit includes an OFD encoded by the re-encoding unit.
The phase of the vector for M symbols is calculated, and the phase correction information estimation unit has already calculated the phase of the vector for the known signal included in the re-encoded OFDM symbol calculated by the phase calculation unit. The OFDM communication device according to claim 1, wherein the estimated phase shift value is corrected. 10. The OFDM communication apparatus according to claim 1, wherein the phase calculator calculates only the phase of the vector signal for the known signal. 11. The OFDM communication apparatus according to claim 1, wherein the phase correction information estimation unit uses a polarity inverter for comparison with the known phase. 12. The phase shift factor is classified into a factor due to multipath fading and a factor other than multipath fading, and a known symbol for phase error estimation due to the multipath fading is included in the OFDM signal. And a plurality of known pilot signals included in each data symbol are included, and the phase correction information estimation unit subtracts the known phase from the phase of the vector for the known symbol for phase error estimation. According to the first phase shift estimated value calculation unit for calculating the phase shift estimated value due to the multipath fading, the known phase is subtracted from the phase of the vector for the pilot signal, and the phase shift estimated value due to the multipath fading is further calculated. By subtracting, factors other than multipath fading can be Based on the pilot signal phase shift amount calculation unit that calculates the phase shift amount of the pilot signal and the phase shift amount calculated by the pilot signal phase shift amount calculation unit, estimate the phase shift due to a factor other than the multipath fading. A second phase shift estimated value calculation unit that calculates a straight line and calculates a phase shift estimated value due to factors other than multipath fading of each OFDM symbol, and the summing unit includes a phase shift estimated value due to the multipath fading. The OFDM communication apparatus according to claim 1, wherein the total phase correction value is calculated by adding together a phase shift estimation value due to a factor other than the multipath fading. 13. The second phase shift estimated value calculation unit includes:
The weighted average is calculated by obtaining the weighted average of the linear parameter obtained from the pilot signal included in the OFDM symbol to be demodulated and the linear parameter obtained from the pilot signal included in the OFDM symbol other than the demodulated object. 13. The OFDM according to claim 12, characterized in that a straight line is obtained, and an estimated value of phase shift due to factors other than the multipath fading is calculated.
Communication device. 14. The second phase shift estimated value calculation unit comprises:
The OFD according to claim 13, wherein a weighted average is calculated by increasing a weight value for the parameter obtained from a pilot signal adjacent to a pilot signal included in an OFDM symbol currently being demodulated.
M communication device. 15. The second phase shift estimated value calculation unit comprises:
As a parameter of the straight line obtained from a pilot signal included in an OFDM symbol other than a demodulation target, a parameter obtained from a pilot signal included in an OFDM symbol before the OFDM symbol to be a demodulation target is stored to obtain a weighted average. The OFDM communication device according to claim 14, wherein the OFDM communication device is determined. 16. A vector storage unit for temporarily storing a vector of an OFDM symbol to be demodulated, and a vector of the pilot signal of the vector of the OFDM symbol stored in the vector storage unit is preceded. And a control unit for supplying the pilot signal phase shift amount calculation unit to the pilot signal phase shift amount calculation unit, wherein the pilot signal phase shift amount calculation unit calculates the phase shift amount of the vector for the pilot signal given earlier and calculates the second phase. The second estimated phase shift value calculation unit provides the shift estimated value calculation unit with an O
As a parameter of the straight line obtained from the pilot signal included in the FDM symbol, based on the phase shift amount of the preceding pilot signal given from the pilot signal phase shift amount calculation unit, O to be demodulated.
A parameter obtained from the OFDM symbols after the FDM symbol is stored, and a weighted average is obtained using the parameter and the parameter obtained from the pilot signal included in the OFDM symbol before the OFDM symbol to be demodulated. 16. The OFDM communication device according to claim 15, characterized in that 17. A vector storage unit for temporarily storing a vector for an OFDM symbol to be demodulated, and a vector of a pilot signal among the vectors for the OFDM symbol stored in the vector storage unit. And a control unit for supplying the pilot signal phase shift amount calculation unit to the pilot signal phase shift amount calculation unit, wherein the pilot signal phase shift amount calculation unit calculates the phase shift amount of the vector for the pilot signal given earlier and calculates the second phase. The second estimated phase shift value calculation unit provides the shift estimated value calculation unit with an O
As a parameter of the straight line obtained from the pilot signal included in the FDM symbol, based on the phase shift amount of the preceding pilot signal given from the pilot signal phase shift amount calculation unit, O to be demodulated.
15. The OFDM communication apparatus according to claim 14, wherein the weighted average is obtained by storing parameters obtained from the OFDM symbols after the FDM symbol. 18. A reception state determination unit that determines a reception state of the received OFDM signal, wherein the reception state determination unit is in accordance with a reception state of an OFDM symbol included in the received OFDM signal. 14. The O according to claim 13, wherein a weight value of a weighted average by the second phase shift estimated value calculation unit is changed.
FDM communication device. 19. A re-encoding unit that periodically re-encodes the demodulated OFDM symbol, wherein the phase calculation unit includes an OFD encoded by the re-encoding unit.
The phase of the vector for M symbols is calculated, and the pilot signal phase shift calculation unit has already calculated the phase of the vector for the pilot signal included in the re-encoded OFDM symbol calculated by the phase calculation unit. Subtract the phase of the pilot signal vector corresponding to the OFDM symbol, and based on the result,
The OFDM according to claim 12, wherein the estimated value of the phase shift due to the multipath fading is modified.
Communication device. 20. The received OFDM signal is subjected to a Discrete Fourier Transform (FFT).
An OFDM communication apparatus for correcting the phase of a vector signal obtained by performing the Sform) processing, the phase calculation unit calculating the phase of the vector signal, and the pilot included in the OFDM symbol obtained by the phase calculation unit. Based on the phase of the vector for the signal, after obtaining the phase shift estimation line due to factors other than multipath fading, the phase shift estimated value calculation unit for obtaining the phase shift estimated value for each OFDM symbol, the phase shift estimated value A phase shift estimation unit that rotates the phase of the vector signal by the estimated phase shift value due to factors other than multipath fading calculated by the calculation unit, and outputs a vector signal whose phase has been corrected. The value calculation unit is an OFDM to be demodulated.
The phase shift estimation straight line parameter obtained from the pilot signal included in the symbol and the phase shift estimation straight line parameter obtained from the pilot signal included in the OFDM symbol other than the demodulation target are obtained by obtaining a weighted average to obtain the phase. An OFDM communication device, characterized in that a deviation estimation value is obtained. 21. The phase shift estimation value calculation unit obtains from a pilot signal included in an OFDM symbol before demodulation as a parameter of a phase shift estimation line obtained from a pilot signal included in an OFDM symbol other than a demodulation target. 21. The OFDM communication apparatus according to claim 20, wherein a weighted average with respect to past parameters is obtained by storing parameters of the estimated straight line. 22. A vector storage unit for temporarily storing a vector of an OFDM symbol to be demodulated, and a vector of the pilot signal among the vectors of the OFDM symbols stored in the vector storage unit is preceded. And a control unit for giving the phase shift estimated value calculation unit, wherein the phase shift estimated value calculation unit obtains and stores the parameter of the estimated straight line based on the pilot signal given in advance. 21. The OFDM communication apparatus according to claim 20, wherein a weighted average with the parameters of the estimated straight line after the demodulation target is obtained. 23. The received OFDM signal is subjected to a Discrete Fourier Transform (FFT).
An OFDM communication apparatus for correcting the phase of a vector signal obtained by performing the Sform) processing, comprising: a phase calculation unit that calculates the phase of the vector signal; Based on the phase of the vector for the known symbol for phase error estimation due to path fading, a phase deviation estimated value calculation unit for obtaining a phase deviation estimated value due to multipath fading, and multipath fading calculated by the phase deviation estimated value calculation unit. A vector phase rotation unit that rotates the phase of the vector signal by the estimated phase shift value and outputs a phase-corrected vector signal, and a re-encoding unit that periodically re-encodes the demodulated OFDM symbol. And the phase calculation unit includes an OFD encoded by the re-encoding unit.
The phase of the vector for M symbols is calculated, and the phase shift estimated value calculation unit has already calculated the phase of the vector for the pilot signal included in the re-encoded OFDM symbol calculated by the phase calculation unit. An OFDM communication apparatus, wherein the phase of a vector of a pilot signal corresponding to the OFDM symbol is subtracted, and the estimated value of the phase shift due to the multipath fading is corrected based on the result. 24. The received OFDM signal is subjected to a discrete Fourier transform (FFT).
sform) processing is performed to correct the phase of the vector signal, the method comprising: calculating the phase of the vector signal, each of a plurality of known signals included in the calculated OFDM signal. For each phase of the vector signal, by comparing each with a known phase, a step of calculating each phase shift estimated value for each of a plurality of factors of phase shift, and a plurality of calculated phase shift estimated values An OFDM communication method comprising: a step of setting a summed value as a total phase correction value; and a step of rotating a phase of the vector signal by the total phase correction value to obtain a vector signal in which the phase is corrected. 25. The received OFDM signal is subjected to a Discrete Fourier Transform (FFT).
A method for correcting the phase of a vector signal obtained by processing the vector signal based on the phase of the vector for the pilot signal included in the calculated OFDM symbol. Then, after obtaining the phase shift estimation line due to factors other than multipath fading, the step of calculating the phase shift estimated value for each OFDM symbol, and the calculated phase shift estimated value due to factors other than multipath fading, Rotating the phase of the vector signal to obtain a vector signal whose phase has been corrected, in the step of calculating the phase shift estimated value, the phase obtained from the pilot signal included in the OFDM symbol to be demodulated. Parameter of deviation estimation line and OF other than demodulation target An OFDM communication method, wherein the phase shift estimation value is obtained by obtaining a weighted average with a parameter of a phase shift estimation line obtained from a pilot signal included in a DM symbol. 26. The received OFDM signal is subjected to a Discrete Fourier Transform (FFT).
A method for correcting the phase of a vector signal obtained by performing a process for calculating the phase of the vector signal, the method for estimating a phase error due to multipath fading included in the calculated OFDM symbol. Calculating a phase shift estimated value due to multipath fading based on the phase of the vector for the known symbol; rotating the phase of the vector signal by the calculated phase shift estimated value due to the multipath fading, Obtaining a corrected vector signal, periodically re-encoding the demodulated OFDM symbol, obtaining a vector phase for the re-encoded OFDM symbol, and re-encoding The pilot signal included in the OFDM symbol From the stomach to the vector of the phase, by subtracting the phase of the vector of the corresponding pilot signal to the OFDM symbols which have already been calculated, based on the results,
Updating the phase shift estimation value due to the multipath fading. 27. The received OFDM signal is subjected to a Discrete Fourier Transform (FFT).
A program executed by an OFDM communication device for correcting the phase of a vector signal obtained by performing a process, the step of calculating the phase of the vector signal, and a plurality of steps included in the calculated OFDM signal. For each phase of each vector signal of the known signal, by comparing each with the known phase, for each factor of a plurality of phase shifts existing, a step of calculating each phase shift estimation value, A step of setting a value obtained by adding the estimated phase shift values as a total phase correction value; and a step of rotating the phase of the vector signal by the total phase correction value to obtain a vector signal in which the phase is corrected. Communication program. 28. The received OFDM signal is subjected to a Discrete Fourier Transform (FFT).
Sform) is a program executed by an OFDM communication apparatus for correcting the phase of a vector signal obtained by performing a process, the step of calculating the phase of the vector signal, the pilot signal included in the calculated OFDM symbol, Based on the phase of the vector, after calculating the phase shift estimation line due to factors other than multipath fading, calculating the phase shift estimation value for each OFDM symbol, and calculating the phase shift due to factors other than multipath fading. Rotating the phase of the vector signal by the estimated value and obtaining a vector signal with the phase corrected, in the step of calculating the phase shift estimated value, the pilot signal included in the OFDM symbol to be demodulated Of the phase shift estimation straight line obtained from An OFDM communication program, wherein the phase shift estimation value is obtained by obtaining a weighted average of a meter and a parameter of a phase shift estimation line obtained from a pilot signal included in an OFDM symbol other than a demodulation target. 29. The received OFDM signal is subjected to a Discrete Fourier Transform (FFT).
A program executed by an OFDM communication apparatus for correcting the phase of a vector signal obtained by performing a process of calculating the phase of the vector signal, the multipath fading included in the calculated OFDM symbol. Based on the phase of the vector for the known symbol for phase error estimation, a step of calculating a phase shift estimation value due to multipath fading, and a phase shift of the vector signal by the calculated phase shift estimation value due to multipath fading. Rotating, obtaining a phase-corrected vector signal, periodically re-encoding the demodulated OFDM symbol, obtaining a vector phase for the re-encoded OFDM symbol, Recoded OFDM symbol From the vector phase of the pilot signal included in, already subtracted phase vectors of the pilot signal corresponding to the OFDM symbols which are calculated, based on the results,
Updating the phase shift estimation value due to the multipath fading.