JP4287308B2 - Frequency offset estimation method and apparatus, and receiving apparatus using the same - Google Patents

Description

  The present invention relates to a frequency offset estimation technique, and more particularly to a frequency offset estimation method and apparatus for estimating a frequency offset without using the periodicity of a signal sequence pattern, and a receiving apparatus using the same.

  In wireless communication, effective use of limited frequency resources is generally desired. One of the technologies for effectively using frequency resources is the adaptive array antenna technology. In the adaptive array antenna technology, the antenna directivity pattern is formed by controlling the amplitude and phase of signals transmitted and received by a plurality of antennas. That is, an apparatus equipped with an adaptive array antenna changes the amplitude and phase of signals received by a plurality of antennas, adds the changed plurality of received signals, respectively, and changes the amplitude and phase (hereinafter referred to as the amount of change). A signal equivalent to a signal received by an antenna having a directivity pattern corresponding to “weight” is received.

  In the adaptive array antenna technology, an example of a process for calculating a weight is a method based on a minimum mean square error (MMSE) method. In the MMSE method, a Wiener solution is known as a condition for giving an optimum weight value, and a recurrence formula with a smaller amount of calculation than directly solving the Wiener solution is also known. As the recurrence formula, for example, an adaptive algorithm such as an RLS (Recursive Least Squares) algorithm or an LMS (Least Mean Squares) algorithm is used. On the other hand, for the purpose of increasing the data transmission speed and improving the transmission quality, there are cases where data is subjected to multicarrier modulation and a multicarrier signal is transmitted.

There is a MIMO (Multiple Input Multiple Output) system as a technique for increasing the data transmission speed using the adaptive array antenna technology. In the MIMO system, each of the transmission apparatus and the reception apparatus includes a plurality of antennas, and separate signals are transmitted from the plurality of antennas included in the transmission apparatus, and the reception apparatus separates these signals using adaptive array antenna technology. Receive. That is, for the communication between the transmission device and the reception device, channels up to the maximum number of antennas are set to improve the data transmission rate. Furthermore, if a multicarrier signal is transmitted in such a MIMO system, the data transmission rate is further increased. On the other hand, when the received signal is demodulated, the receiving device needs to estimate the transmission path characteristics and weight. However, the inventor has so far disclosed a technique for estimating these with high accuracy (for example, Patent Documents). 1).
JP 2003-124857 A

  In general, transmission path characteristics are estimated based on a training signal included in a transmission signal. According to the transmission path characteristic estimation technique described in Non-Patent Document 1, in order to estimate the transmission path characteristics with high accuracy despite the increase in the number of antennas of the transmission apparatus and the reception apparatus, training is performed. It is desirable that the signal pattern is not periodic. On the other hand, the receiving apparatus also estimates the frequency offset included in the received signal. Generally, the frequency offset is estimated using the periodicity of the pattern of the training signal. In terms of channel utilization efficiency, the training signal section should be short, and it is desirable to be able to estimate the frequency offset based on a training signal that is not a periodic pattern.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a frequency offset estimation method and apparatus for estimating a frequency offset based on a training signal that is not a periodic pattern, and a receiving apparatus using the same. It is in.

One embodiment of the present invention is a frequency offset estimation apparatus. This apparatus includes a receiving unit that sequentially receives a sequence of transmitted known signals as a received signal via a transmission channel, and a transmission channel characteristic that estimates transmission channel characteristics based on the received signal and the sequence of known signals that are sequentially received. An estimation unit, a generation unit that generates a plurality of replica signals corresponding to sequentially received signals from a sequence of estimated transmission path characteristics and known signals, one of the sequentially received signals, and the sequentially received reception A derivation unit for deriving a phase error between the replica signal corresponding to one of the signals and deriving a plurality of phase errors over at least a part of a sequence of the known signal; and based on the derived plurality of phase errors. And a frequency offset estimator for estimating the frequency offset included in the sequentially received reception signal.
With the above device, the phase error between the replica signal generated from the received signal and the received signal is derived, and the frequency offset is estimated from the variation in the derived phase error. Therefore, there is no predetermined periodicity in the sequence of known signals. Can also estimate the frequency offset.

  The receiving unit sequentially receives a plurality of known signal sequences respectively transmitted from a plurality of transmitting antennas as a plurality of received signals by a plurality of receiving antennas, and the patterns of the plurality of known signal sequences are different from each other, The transmission path characteristic estimator is configured for a plurality of combinations of each of a plurality of transmission antennas and each of a plurality of reception antennas based on a sequence of a plurality of reception signals sequentially received and a plurality of known signals having different patterns. And generating a plurality of replica signals corresponding to each of a plurality of receiving antennas from a sequence of a plurality of known signals having different patterns from the estimated plurality of transmission path characteristics. The derivation unit derives a plurality of phase errors corresponding to each of the plurality of reception antennas, and the frequency offset estimation unit calculates the frequency offset. DOO or the estimated corresponding to each of the plurality of receiving antennas.

  The “known signal sequence pattern” is a combination of known signal sequence values in a predetermined period. Furthermore, the length of the predetermined period for “the patterns of a plurality of known signal sequences are different” may be arbitrary. That is, it may be different in a short cycle or may be different in a long cycle.

  The derivation unit may divide one value of the received signal received sequentially by the value of the replica signal corresponding to one of the received signals received sequentially, and derive the phase component of the division result as a phase error. . The deriving unit multiplies one value of the sequentially received signal by the complex conjugate value of the replica signal corresponding to one of the sequentially received signals, and derives the phase component of the multiplication result as a phase error. Also good. The frequency offset estimation unit may estimate the frequency offset included in the received signal sequentially received from the plurality of derived phase errors by the least square method. The frequency offset estimator selects predetermined two of the plurality of derived phase errors, and derives a difference in phase error between the selected two by subtraction, so that it is included in the sequentially received reception signal. The frequency offset may be estimated. The frequency offset estimation unit may derive one frequency offset by statistical processing from the frequency offset estimated corresponding to each of the plurality of receiving antennas.

  Another aspect of the present invention is a receiving device. This apparatus includes a receiving unit that sequentially receives a sequence of transmitted known signals as a received signal via a transmission channel, and a transmission channel characteristic that estimates transmission channel characteristics based on the received signal and the sequence of known signals that are sequentially received. An estimation unit, a generation unit that generates a plurality of replica signals corresponding to sequentially received signals from a sequence of estimated transmission path characteristics and known signals, one of the sequentially received signals, and the sequentially received reception A derivation unit for deriving a phase error between the replica signal corresponding to one of the signals and deriving a plurality of phase errors over at least a part of a sequence of the known signal; and based on the derived plurality of phase errors. A frequency offset estimator for estimating the frequency offset included in the received signal received sequentially, and the received signal received sequentially by the estimated frequency offset. Comprises a correcting unit for correcting a frequency offset contained in the signal to be received by the receiver followed, and a processing unit for processing the corrected signal frequency offset.

  Yet another aspect of the present invention is a frequency offset estimation method. This method sequentially receives a transmitted sequence of known signals as a received signal via a transmission path, estimates transmission path characteristics based on the received signal and the sequence of known signals, and estimates the estimated transmission path characteristics. And a plurality of replica signals respectively corresponding to the received signals sequentially received from the sequence of known signals, and at least a part of the sequence of known signals between the received signals and the plurality of replica signals sequentially received. A plurality of phase errors are respectively derived, and a frequency offset included in the received reception signal is estimated based on the derived plurality of phase errors.

  Yet another embodiment of the present invention is also a frequency offset estimation method. The method includes a step of sequentially receiving a sequence of transmitted known signals as a received signal via a transmission path, a step of estimating transmission path characteristics based on the sequence of received signals and a sequence of known signals, and an estimation Generating a plurality of replica signals each corresponding to a sequentially received signal, one of the sequentially received signals, and one of the sequentially received signals from the transmission path characteristics and the known signal sequence A step of deriving a phase error with respect to the replica signal, further deriving a plurality of phase errors over at least a part of a section of the known signal sequence, and a received signal sequentially received based on the plurality of derived phase errors; And estimating a frequency offset included in.

  In the receiving step, a plurality of known signal sequences respectively transmitted from a plurality of transmitting antennas are sequentially received as a plurality of received signals by a plurality of receiving antennas, and the patterns of the plurality of known signal sequences are different from each other. The step of estimating the transmission path characteristics is based on a combination of each of a plurality of transmission antennas and each of a plurality of reception antennas based on a sequence of a plurality of reception signals sequentially received and a plurality of known signals having different patterns. The step of estimating and generating a plurality of transmission path characteristics respectively includes a plurality of replica signals corresponding to each of a plurality of receiving antennas from a sequence of a plurality of known signals having different patterns from the estimated transmission path characteristics. Generating and deriving a plurality of phase errors corresponding to each of a plurality of receiving antennas. , Estimating the frequency offset may be estimated to correspond to the respective frequency offsets of the plurality of receiving antennas.

  In the deriving step, one value of the sequentially received signal is divided by the value of the replica signal corresponding to one of the sequentially received signals, and the phase component of the division result is derived as a phase error. Good. The deriving step multiplies one value of the received signal sequentially received by the complex conjugate value of the replica signal corresponding to one of the sequentially received signals, and derives the phase component of the multiplication result as a phase error. May be. The step of estimating the frequency offset may estimate the frequency offset included in the received signal sequentially received from the plurality of derived phase errors by the least square method. The step of estimating the frequency offset includes a predetermined two of the derived phase errors, and includes the difference in the phase error between the selected two by subtraction, thereby including in the sequentially received signal. The estimated frequency offset may be estimated. In the step of estimating the frequency offset, one frequency offset may be derived by statistical processing from the frequency offset estimated corresponding to each of the plurality of receiving antennas.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.

  According to the present invention, the frequency offset can be estimated based on a training signal that is not a periodic pattern.

Example 1
Before describing the present invention in detail, an outline will be described. Embodiment 1 of the present invention relates to a MIMO system configured by a transmission apparatus having a plurality of antennas and a reception apparatus having a plurality of antennas. In addition, the transmission apparatus according to the present embodiment transmits a signal by a multicarrier, specifically, OFDM (Orthogonal Frequency Division Multiplexing) modulation, and the transmitted signal forms a burst signal. A preamble signal is arranged at the head portion of the burst signal, and a receiving apparatus that receives the signal executes transmission path characteristic estimation, frequency offset estimation, and reception weight vector calculation based on the preamble signal.

  The receiving apparatus according to the present embodiment estimates transmission path characteristics from a received signal and a preamble signal. The receiving device calculates a reception weight vector from the estimated transmission path characteristics, generates a replica signal from the estimated transmission path characteristics and the preamble signal, and calculates a phase error between the received signal and the replica signal. Further, the frequency offset is estimated from the time change of the phase error between the received signal and the replica signal. Therefore, the receiving apparatus can execute transmission path characteristic estimation, frequency offset estimation, and reception weight vector calculation from the same preamble signal. Further, the receiving apparatus demodulates the received data signal following the preamble signal based on the estimated transmission path characteristic, the estimated frequency offset, and the reception weight vector.

  FIG. 1 illustrates a spectrum of a multicarrier signal according to the first embodiment. FIG. 1 shows a spectrum of a signal in a wireless LAN (Local Area Network) compliant with the IEEE 802.11a standard as a wireless system to which the OFDM modulation method is applied. The MIMO system of the present embodiment will be described based on the IEEE802.11a standard for simplicity of explanation, but is not limited to the IEEE802.11a standard. One of a plurality of carriers in the OFDM system is generally called a subcarrier, but here, one subcarrier is designated by a “subcarrier number”.

  In the IEEE802.11a standard, the size of the Fourier transform is 64 (hereinafter, one FFT (Fourier Transform) point is referred to as “FFT point”), so that signals are arranged from subcarrier numbers “0” to “63”. Although possible, as shown in the figure, signals are arranged on 52 subcarriers from subcarrier numbers “−6” to “31” and “33” to “58”. The subcarrier number “32” is set to null in order to reduce the influence of the DC component in the baseband signal. The signal arranged in the subcarrier is modulated by BPSK, QSPK, 16QAM, and 64QAM.

  FIG. 2 illustrates a configuration of the communication system 100 according to the first embodiment. The communication system 100 includes a transmission device 10 and a reception device 12. Further, the transmission device 10 includes a first transmission antenna 14 a and an L-th transmission antenna 14 l that are collectively referred to as a transmission antenna 14, and the reception device 12 is a first reception antenna 16 a that is generally referred to as a reception antenna 16. , An M-th receiving antenna 16m. As described above, the number of transmitting antennas 14 is L, and the number of receiving antennas 16 is M.

  The transmitting apparatus 10 transmits different signals from the first transmitting antenna 14a to the Lth transmitting antenna 14l. The receiving device 12 receives a signal transmitted from the first transmitting antenna 14a to the second transmitting antenna 14l by the first receiving antenna 16a to the Mth receiving antenna 16m. Furthermore, the receiving device 12 separates the received signals by adaptive array signal processing, and independently demodulates the signals transmitted from the first transmitting antenna 14a from the Lth transmitting antenna 14l. Here, the transmission path characteristic between the first transmission antenna 14a and the first reception antenna 16a is h11, the transmission path characteristic between the first transmission antenna 14a and the Mth reception antenna 16m is hm1, If the transmission path characteristic between the L transmitting antenna 14l and the first receiving antenna 16a is h11, and the transmission path characteristic between the L transmitting antenna 14l and the Mth receiving antenna 16m is hm1, the receiving apparatus 12 enables adaptively demodulating signals transmitted from the first transmitting antenna 14a to the Mth transmitting antenna 14m by enabling only transmission paths such as h11 and hm1, for example, by adaptive array signal processing. To work.

  FIG. 3 shows the configuration of the burst format according to the first embodiment, which corresponds to a speech channel of the IEEE 802.11a standard. In the OFDM modulation system, generally, the total of the Fourier transform size and the number of symbols in the guard interval is used as one unit. This single unit is called an OFDM symbol in this embodiment. In the IEEE802.11 standard, the size of the Fourier transform is 64 as described above, and the number of FFT points in the guard interval is 16, so the OFDM symbol corresponds to 80 FFT points.

  In the burst signal, “preamble” of “4 OFDM symbol”, “signal” of “1 OFDM symbol”, and “data” of an arbitrary length are arranged from the head. The preamble is a known signal transmitted for AGC setting, timing synchronization, carrier reproduction, and the like in the receiving apparatus 12. The signal is a control signal, and the data is information to be transmitted from the transmission device 10 to the reception device 12. Further, as shown in the figure, the “preamble” of the “4OFDM symbol” is separated into “STS (Short Training Sequence)” of “2OFDM Symbol” and “LTS (Long Training Sequence)” of “2OFDM Symbol”.

  The STS is composed of ten signal units “t1” to “t10”, and one unit “t1” or the like has 16 FFT points, and has a predetermined periodicity by “t1” to “t10”. Is formed. As described above, the STS has a unit of 16 FFT points in the time domain, but in the frequency domain, 12 subcarriers among the 64 subcarriers shown in FIG. 1 are used. The STS is particularly used for AGC setting, timing synchronization, and initial frequency offset estimation. Here, the estimation of the initial frequency offset is the estimation of the frequency offset whose estimation accuracy is somewhat rough.

  On the other hand, the LTS is composed of two signal units “T1” and “T2” and a guard interval “GI2” twice as long as the above-described guard interval, and one unit “T1” is 64 FFT. It is a point, and “GI2” is a 32 FFT point. The LTS is used particularly for estimation of transmission path characteristics, estimation of a highly accurate frequency offset, and calculation of a reception weight vector. In the IEEE802.11a standard, “T1” and “T2” of LTS have the same signal pattern, that is, have periodicity between them, but in this embodiment, “T1” of LTS "And" T2 "have different signal patterns, i.e., no periodicity between them. Hereinafter, the LTS is simply referred to as a preamble signal or a training signal, and the LTS may indicate a combination of “T1” and “T2”.

  FIG. 4 shows the configuration of the transmission apparatus 10. The transmitter 10 includes a data separation unit 20, a first modulation unit 22a, a second modulation unit 22b, an L-th modulation unit 22l, which are collectively referred to as a modulation unit 22, and a first radio unit 24a, a second, which are collectively referred to as a radio unit 24. The radio unit 24b, the L-th radio unit 24l, the control unit 26, and the first modulation unit 22a include an error correction unit 28, an interleave unit 30, a preamble addition unit 32, an IFFT unit 34, a GI unit 36, and an orthogonal modulation unit 38. The first radio unit 24a includes a frequency conversion unit 40 and an amplification unit 42.

  The data separator 20 separates data to be transmitted according to the number of antennas. The error correction unit 28 performs encoding for error correction on the data. Here, it is assumed that convolutional encoding is performed, and the encoding rate is selected from predetermined values. The interleave unit 30 interleaves the convolutionally encoded data. The preamble adding unit 32 adds STS and LTS to the head of the burst signal. Therefore, the preamble adding unit 32 stores STS and LTS, respectively, and at least the LTS has different patterns corresponding to the first to Lth modulation units 22l to 22l.

  The IFFT unit 34 performs IFFT (Inverse Fast Fourier Transform) in units of FFT points, and converts a frequency domain signal using a plurality of subcarrier carriers into a time domain signal. The GI unit 36 adds a guard interval to the time domain data. As shown in FIG. 3, guard intervals added to the LTS and the data are different. The quadrature modulation unit 38 performs quadrature modulation on time domain data. The frequency converter 40 converts the orthogonally modulated signal into a radio frequency signal. The amplifying unit 42 is a power amplifier that amplifies a radio frequency signal. Finally, separate signals are transmitted from the plurality of transmitting antennas 14. The control unit 26 controls the timing of the transmission device 10 and the like. In the present embodiment, it is assumed that the directivity of each of the transmitting antennas 14 is omnidirectional, and the transmitting apparatus 10 does not perform adaptive array signal processing.

  FIG. 5 shows the configuration of the receiving device 12. The receiving device 12 includes a first radio unit 50a, a second radio unit 50b, an M-th radio unit 50m, which are collectively referred to as a radio unit 50, a first processing unit 52a, a second processing unit 52b, which are collectively referred to as a processing unit 52, An M processor 52m, a first demodulator 54a, a second demodulator 54b, an L demodulator 54l, a data combiner 56, and a controller 58, which are collectively referred to as a demodulator 54, are included. Further, as signals, a first radio reception signal 200a, a second radio reception signal 200b, an M-th radio reception signal 200m, and a first baseband reception signal 202a, which are collectively referred to as a radio reception signal 200, A second baseband received signal 202b, an Mth baseband received signal 202m, a first synthesized signal 204a collectively referred to as a synthesized signal 204, a second synthesized signal 204b, and an Lth synthesized signal 204l are included.

  The radio unit 50 performs frequency conversion processing, amplification processing, AD conversion processing, and the like between the radio frequency radio reception signal 200 and the baseband baseband reception signal 202. Here, since a wireless LAN based on the IEEE802.11a standard is assumed as the communication system 100, the wireless frequency of the wireless reception signal 200 corresponds to the 5 GHz band. In addition, the initial frequency offset is estimated and corrected. The processing unit 52 performs adaptive array signal processing on the baseband received signal 202 and outputs a composite signal 204 corresponding to the transmitted plurality of signals. The demodulator 54 demodulates the composite signal 204. Furthermore, deinterleaving and decoding are also executed. The data combiner 56 combines the signals output from the demodulator 54 corresponding to the data separator 20 in FIG. The control unit 58 controls the timing of the receiving device 12 and the like.

  FIG. 6 shows the configuration of the first radio unit 50a. The first radio unit 50a includes an LNA unit 60, a frequency conversion unit 62, a quadrature detection unit 64, an AGC 66, an AD conversion unit 68, and an initial frequency offset correction unit 70.

  The LNA unit 60 amplifies the first radio reception signal 200a. The frequency conversion unit 62 performs frequency conversion between a radio frequency 5 GHz band and an intermediate frequency on a signal to be processed. The quadrature detection unit 64 performs quadrature detection on the intermediate frequency signal to generate a baseband analog signal. The AGC 66 automatically controls the gain in order to make the amplitude of the baseband analog signal within the dynamic range of the AD converter 68. In the initial setting of the AGC 66, the STS of the received signal is used, and control is performed so that the strength of the STS approaches a predetermined value. The AD converter 68 converts a baseband analog signal into a digital signal.

  The initial frequency offset correction unit 70 estimates an initial frequency offset in the STS section, and further corrects the LTS, signal, and data by the estimated initial frequency offset. The initial frequency offset estimation method may be any method. For example, a phase error between signals corresponding to “t9” and “t10” in the STS is calculated, and a time corresponding to 16 FFT points is calculated. Divide the phase error calculated in. Further, statistical processing such as averaging may be added to such a result. The signal whose initial frequency offset is corrected is output as the first baseband received signal 202a.

  FIG. 7 shows a configuration of the first processing unit 52a. The first processing unit 52a includes a transmission path characteristic estimation unit 80, a frequency offset estimation unit 82, a first FFT 84a, a second FFT 84b, an MFFT 84m, collectively referred to as FFT 84, a reception weight calculation unit 86, and a first multiplication collectively referred to as a multiplication unit 88. Unit 88a, second multiplier 88b, Mth multiplier 88m, first delay unit 90a, second delay unit 90b, Mth delay unit 90m, and first unit collectively referred to as delay unit 90 Unit 92a, second multiplier 92b, Mth multiplier 92m, and eleventh multiplier 94aa, twelfth multiplier 94ab, first N multiplier 94an, 21st multiplier 94ba, and 22nd multiplier, collectively referred to as multiplier 94. 94bb, 2nd N multiplier 94bn, M1 multiplier 94ma, M2 multiplier 94mb, MN multiplier 94mn, first adder 96a collectively referred to as adder 96, second adder 96b, it includes a first N addition unit 96n. Further, the signal includes a transmission path characteristic signal 206, a first frequency offset signal 210a, which is collectively referred to as a frequency offset signal 210, a second frequency offset signal 210b, and an Mth frequency offset signal 210m.

  The baseband received signal 202 forms a burst signal as shown in FIG. 3 and is received sequentially. Here, each of the signals transmitted from the transmitting antenna 14 includes different LTSs, and the baseband received signal 202 is a combined signal.

The transmission path characteristic estimation unit 80 stores in advance LTSs of different patterns respectively transmitted from the transmitting antennas 14, and based on these LTSs and the baseband received signal 202, the transmission path characteristics in the LTS section are calculated. presume. As shown in FIG. 2, the transmission path characteristic estimation unit 80 estimates transmission path characteristics for each combination of L transmitting antennas 14 and M receiving antennas 16. That is, L × M types of transmission path characteristics are estimated. Here, an example of a transmission path characteristic estimation method in the transmission path characteristic estimation unit 80 will be described. If the number of samples in one LTS section is S, the total number of samples in all LTS sections is 2S. Since oversampling is not performed, S is 64. If a vector composed of the mth baseband received signal 202 in the LTS section is zm, zm is expressed as follows.

ここで、マルチパスの遅延時間をｌ_ｋとし、ｋ＝０を先行波として考えるので、?０＝０としている。さらに、厳密に遅延時間は、送信用アンテナ１４と受信用アンテナ１６に応じて異なるが、ベースバンドにおけるサンプル値を考える限りにおいて、送信用アンテナ１４と受信用アンテナ１６によらないと考えれられる。したがって、遅延時間は?_{ｍ，ｋ，ｌ}と書くべきであるが、ここでは、単に?_ｋと示す。また、ＬＴＳ区間において送信用アンテナ１４から送信されるＬＴＳの系列は、受信装置１２で既知である。ｍ番目の受信用アンテナ１６に到来したｌ番目の送信用アンテナ１４からのｋ番目のマルチパス波の複素振幅をｈ_{ｍ，ｌ，ｋ}とすれば、評価関数Ｊ_ｍは次のように示される。 Here, x _m (t) represents the m-th baseband received signal 202 at time t. If the LTS transmitted from the l-th transmitting antenna 14 at time t is d _l (t), the 2S-dimensional signal composed of the k-th multipath wave of the signal transmitted from the l-th transmitting antenna 14 is used. The vector d _{l, k} is shown as follows:

Here, the delay time of multi-path and l _k, so think about the k = 0 as the preceding wave, is set to? 0 = 0. Furthermore, although the delay time strictly differs depending on the transmission antenna 14 and the reception antenna 16, it is considered that the delay time does not depend on the transmission antenna 14 and the reception antenna 16 as long as the sample values in the baseband are considered. Therefore, the delay time should be written as? _{M, k, l} , but here it is simply indicated as? _K. In addition, the LTS sequence transmitted from the transmission antenna 14 in the LTS section is known by the receiving device 12. If the complex amplitude of the k-th multipath wave from the l-th transmitting antenna 14 arriving at the m-th receiving antenna 16 is hm _{, l, k} , the evaluation function J _m is expressed as follows. .

ここで、Ｋは想定する最も遅延の大きなマルチパスの番号を示す。表記を簡潔にするために、ｄ_ｌ，ｋを列要素とする２Ｓ行Ｌ（Ｋ＋１）列の行列Ｑとｈ_{ｍ，ｌ，ｋ}からなるＬ（Ｋ＋１）次列ベクトルｂ_ｍを用いてＪ_ｍを表すと以下のようになる。

ただし、Ｑとｂ_ｍは、具体的には次のように示される。

Here, K indicates the number of the multipath with the longest delay assumed. For simplicity of _{notation, d l,} 2S row L (K + 1) matrix of columns Q and _{h m, l,} consisting of _k L (K + 1) to _k and column elements _J using the following column vector _{b m m} Is expressed as follows.

However, Q and b _m are specifically shown as follows.

さらに、Ｊ_ｍを最小にするｂ_ｍは次のように示される。

ここで、Ｈはエルミート共役を示し、ｂ_ｍの各要素がｍ番目の受信用アンテナ１６に入力される各マルチパス波の複素振幅、すなわち伝送路特性となる。なお、Ｑは、ＬＴＳのシンボルにより決まる行列なのでＬ（Ｋ＋１）行２Ｓ列の行列（Ｑ^ＨＱ）^−１Ｑ^Ｈは事前に計算され、伝送路特性推定部８０に記憶されているものとする。最終的に、伝送路特性推定部８０は、伝送路特性ｂ_ｍを伝送路特性信号２０６として出力する。

Further, b _m that minimizes the J _m is shown as follows.

Here, H indicates Hermitian conjugate, and each element of b _m is a complex amplitude of each multipath wave input to the m-th receiving antenna 16, that is, transmission path characteristics. Since Q is a matrix determined by LTS symbols, a matrix (Q ^H Q) ⁻¹ Q ^H of L (K + 1) rows and 2S columns is calculated in advance and stored in the transmission path characteristic estimation unit 80. . Finally, the transmission path characteristic estimation unit 80 outputs the transmission path characteristic b _m as the transmission path characteristic signal 206.

The frequency offset estimation unit 82 estimates the frequency offset based on the transmission path characteristic signal 206. A method for estimating the frequency offset will be described later, but the phase component of the frequency offset estimated at the end of the LTS is inverted before being output as the frequency offset signal 210. Further, the frequency offset signal 210 generates a phase signal to be rotated in the direction opposite to the estimated frequency offset in units of the FFT point time interval T _p by feedback control by the delay unit 90 and the multiplication unit 92.

  Multiplier 88 corrects the frequency offset included in baseband received signal 202 with a phase reflecting the frequency offset output from multiplier 92. The FFT 84 performs FFT processing on the baseband received signal 202 whose frequency offset is corrected, and outputs a frequency domain signal. The number of subcarriers of the signal subjected to the FFT processing by the FFT 84 is 52 as described above, but here it is generalized as N.

さらに、すべての送信用アンテナ１４を考慮して、各列要素がｗ_ｌ（ｎ）であるＭ行Ｌ列の行列をＷ（ｎ）とすると、Ｗ（ｎ）は次のように示される。

The reception weight calculation unit 86 estimates a reception weight vector based on the transmission path characteristic signal 206. The reception weight vector w _l (n) for the l-th transmitting antenna 14 in the n-th subcarrier is expressed as follows.

Further, in consideration of all the transmitting antennas 14, if W (n) is a matrix of M rows and L columns where each column element is w _l (n), W (n) is expressed as follows.

ここで、相関行列Ｒｘｘ（ｎ）とＶ（ｎ）は、次のように示される。

なお、ｘ（ｔ，ｎ）は時刻ｔでｎ番目のサブキャリアでの受信信号を示し、ｄ_ｌ（ｔ，ｎ）は時刻ｔでｌ番目の送信用アンテナ１４から送信されるトレーニング信号を示し、Ｅはアンサンブル平均を示す。 Further, the optimal solution W _opt (n) of W (n) based on the MMSE criterion is expressed as follows.

Here, the correlation matrices Rxx (n) and V (n) are expressed as follows.

Note that x (t, n) represents a received signal on the nth subcarrier at time t, and d _l (t, n) represents a training signal transmitted from the lth transmitting antenna 14 at time t. , E represents the ensemble average.

アレー応答ベクトルを使用すれば、相関行列Ｒｘｘ（ｎ）は次のように示される。 For ease of explanation, the first 16 values of the aforementioned _{hm, l, k} are hm _{, l, 0} , hm _{, l, 1} , ..., hm _{, l, 15} , respectively. It is assumed that an FFT of 64 samples whose elements after that are all 0 is a _{m, l} (n) (n = 0, 1,..., 63). Furthermore, using a _{m, l} (n), an M-dimensional column vector a _l (n) is defined as follows, which is an array response vector from the l-th transmitting antenna 14.

Using an array response vector, the correlation matrix Rxx (n) can be expressed as:

ここで、σ^２は熱雑音電力であり、Ｉは単位行列である。さらに、相関ベクトルｖ_ｌ（ｎ）は、次のように示される。

従って、ＭＭＳＥ基準の受信ウエイトベクトルｗ_{ｌ，ｏｐｔ}（ｎ）は次式で示される。

ひとつのｌの値、例えばｌ＝１に対応した受信ウエイトベクトルｗ_{１，ｏｐｔ}（ｎ）が受信ウエイト計算部８６で計算される。また、それ以外のｌの値、例えばｌ＝２に対応した受信ウエイトベクトルｗ_{２，ｏｐｔ}（ｎ）が図５の第２処理部５２ｂで計算される。

Here, σ ² is thermal noise power, and I is a unit matrix. Further, the correlation vector v _l (n) is expressed as follows.

Accordingly, the reception weight vector w _{l, opt} (n) based on the MMSE is expressed by the following equation.

A reception weight vector w _{1, opt} (n) corresponding to one value of l, for example, l = 1 is calculated by the reception weight calculation unit 86. In addition, the second processing unit 52b shown in FIG. 5 calculates other reception weight vectors w _{2, opt} (n) corresponding to l, for example, l = 2.

The multiplier 94 multiplies the signal that has been subjected to the FFT processing by the FFT 84 and the reception weight vector w _{1, opt} (n). The addition unit 96 adds the multiplication results of the multiplication unit 94 to the same n value in the reception weight vector. Adder 96 outputs the addition results for all subcarriers N. Here, the N addition results are collectively referred to as the first combined signal 204a.

  This configuration can be realized in terms of hardware by a CPU, memory, or other LSI of an arbitrary computer, and in terms of software, it is realized by a program having a reservation management function loaded in memory. The functional block realized by those cooperation is drawn. Accordingly, those skilled in the art will understand that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof.

  FIG. 8 shows the configuration of the frequency offset estimation unit 82. The frequency offset estimation unit 82 includes a phase difference calculation unit 110 and an offset calculation unit 112. The signal includes a first phase error signal 208a, a second phase error signal 208b, and an Mth phase error signal 208n, which are collectively referred to as a phase error signal 208.

  The phase difference calculation unit 110 receives the transmission line characteristic signal 206 and generates a replica signal, which will be described later, from the LTS time domain signal pattern stored in advance and the estimated transmission line characteristic. Furthermore, the phase difference calculation unit 110 calculates the phase difference between the baseband received signal 202 and the corresponding replica signal in units of the receiving antenna 16. The calculated phase difference is output as a phase error signal 208.

Here, FIG. 9 shows a configuration of the phase difference calculation unit 110. The phase difference calculation unit 110 includes a replica generation unit 114 and a division unit 116. Here, the baseband received signal 202 and the phase error signal 208 are shown together as one signal, but this is a display for convenience, and actually, the first baseband received signal 202a and the first phase error are as shown in FIG. It is configured by a signal such as the signal 208a, and the configuration of the phase difference calculation unit 110 also corresponds to it. Furthermore, the baseband received signal 202 and the phase error signal 208 in the figure are signals corresponding to the mth receiving antenna 16. As described above, the replica generation unit 114 generates a replica signal. In other words, if the training signal in the time domain corresponding to each of the l-th transmitting antennas 14 is c _l (t), the replica signal x ′ _m (t) is determined from the transmission path characteristics input by the transmission path characteristics signal 206. Is calculated as follows:

なお、前述のごとくｈ_{ｍ，ｌ，ｋ}は、先頭の１６個の値のみを有効にしている。レプリカ信号ｘ’_ｍ（ｔ）は、ＬＴＳ区間のうちの「Ｔ１」と「Ｔ２」にわたって計算される。

As described above, only the first 16 values are valid for hm _{, l, k} . The replica signal x ′ _m (t) is calculated over “T1” and “T2” in the LTS interval.

The division unit 116 receives the replica signal generated by the replica generation unit 114 and the baseband reception signal 202, and divides the value of the baseband reception signal 202 by the value of the replica signal corresponding to the baseband reception signal 202. Furthermore, the division unit 116 sets the phase component of the division result as a phase error. That is, the phase error y _m (t) in units of the receiving antenna 16 is expressed as follows. The phase error is output as the phase error signal 208.

Returning to FIG. The offset calculation unit 112 estimates the frequency offset included in the baseband reception signal 202 based on the phase error signal 208, that is, the plurality of phase errors y _m (t) derived by the division unit 116. Here, the offset calculator 112 estimates the frequency offset included in the baseband received signal 202 from the plurality of phase errors y _m (t) by the least square method. That is, a phase error fluctuation line y = at + b is assumed from t = 0 to t = 127 in the section from “T1” to “T2” of the LTS, and each of the phase error y _m (t) and the error of the straight line is assumed. The constants a and b are derived so that is minimized. Therefore, the distance l _t between the phase error fluctuation line and the phase error y _m (t) is _expressed as follows.

ここで、図１０は、時間と位相誤差の関係を示し、前述の位相誤差ｙ_ｍ（ｔ）、ｙ、ｌ_ｔは図示のごとく示される。さらに、ｌ_ｔ ^２の総和Ｅは以下のとおりに示される。ここで、Ｅを最小にするような定数ａとｂが導出される。

Here, FIG. 10 shows the relationship between time and phase error, and the aforementioned phase errors y _m (t), y, and l _t are shown as shown. Further, the total E of l _t ² is shown as follows. Here, constants a and b that minimize E are derived.

  Here, the constant a corresponds to the frequency offset, but the constant a is in phase. For example, it is shown as 5 °. The offset calculator 112 inverts the sign of the constant a to derive a phase in the direction opposite to the frequency offset. In the above example, it is indicated as -5 °. Further, the derived phase is converted into a complex number having values of the in-phase component and the quadrature component, and these are output as the frequency offset signal 210. The frequency offset is derived for each receiving antenna 16.

  FIG. 11 is a flowchart showing the procedure of the reception process. The receiving device 12 estimates the timing in the STS section of the received burst signal, the initial frequency offset correction unit 70 estimates the initial frequency offset (S10), and further corrects the estimated frequency offset to receive baseband. The signal 202 is output. In the LTS section of the burst signal, the transmission line characteristic estimation unit 80 estimates the transmission line characteristic (S12). Further, based on the estimated transmission path characteristics, the frequency offset estimation unit 82 estimates the frequency offset (S14), and the reception weight calculation unit 86 calculates the reception weight vector (S16). In step 16, the reception weight calculation unit 86 may calculate the reception weight vector based on the corrected transmission line characteristic after correcting the transmission line characteristic using the frequency offset estimated by the frequency offset estimation unit 82. In the burst signal data section, the multiplier 88 corrects the frequency offset included in the baseband received signal 202, and the multiplier 94 receives data while performing weighting with the reception weight vector (S18).

  According to the embodiment of the present invention, the replica signal is generated based on the transmission path characteristic estimated from the received signal and the known signal, and the frequency offset is estimated from the fluctuation of the phase error between the received signal and the replica signal. The frequency offset can be estimated even if the signal does not have a predetermined periodicity. In addition, since the frequency offset can be estimated even if the training signal does not have a predetermined periodicity, the frequency offset estimation and the reception weight vector can be performed even when the reception weight vector is calculated from a training signal without the predetermined periodicity. The training signal for the calculation can be shared, and the use efficiency of the burst signal can be improved. In addition, since it is possible to cope with an increase in the number of data sequences transmitted in parallel by a plurality of antennas in a MIMO system, it is possible to improve the data transmission speed and transmission quality in the MIMO system. Further, since the frequency offset is estimated by the least square method from the phase error variation of the received signal and the replica signal, the frequency offset can be estimated with high accuracy.

(Example 2)
A second embodiment of the present invention relates to a receiving apparatus in a MIMO system that estimates a frequency offset included in a received signal, similarly to the first embodiment of the present invention. Since the receiving apparatus according to the present embodiment calculates the phase error between the received signal and the replica signal and estimates the frequency offset from the time change of the phase error between the received signal and the replica signal, as in the first embodiment. The training signal does not require a predetermined periodicity. However, in the second embodiment, the estimation of the frequency offset based on the time change of the phase error is performed by obtaining the difference between two phase errors separated by a predetermined time, not by the least square method. Further, here, the phase error is indicated by a real number. Therefore, since the difference between the two phase errors is derived not by complex multiplication but by subtraction of real numbers, the influence of noise can be suppressed. Further, the frequency offset can be estimated more easily than the least square method.

  The receiving device 12 according to the second embodiment is the same as the receiving device 12 of FIG. 5 according to the first embodiment, and the first processing unit 52a according to the second embodiment is the same as the first processing unit 52a of FIG. Since the frequency offset estimation unit 82 according to the second embodiment is the same as the frequency offset estimation unit 82 of FIG. 8 according to the first embodiment, the description thereof is omitted.

  FIG. 12 illustrates a configuration of the offset calculation unit 112 according to the second embodiment. The offset calculation unit 112 includes a delay unit 122, a complex number conversion unit 124, an addition unit 126, an addition unit 128, and an addition unit 128. Here, the phase error signal 208 and the frequency offset signal 210 are collectively shown as one signal, but this is a display for convenience, and actually the first phase error signal 208a and the first frequency offset signal 210a as shown in FIG. It is assumed that the configuration of the offset calculation unit 112 also corresponds to that. Further, the phase error signal 208 and the frequency offset signal 210 here are signals corresponding to the mth receiving antenna 16.

The delay unit 122 delays the phase error signal 208. The adder 126 subtracts the phase error signal 208 by the value delayed by the delay unit 122 to obtain a difference in the phase error signal 208 corresponding to the delay time in the delay unit 122. Here, if the delay time in the delay unit 122 is T, the subtraction result Δy _m (t) in the adder 126 is expressed as follows.

The adding unit 128 adds the subtraction result of the adding unit 126 over a predetermined period of the LTS, and the complex number conversion unit 124 has the following frequency when T = 64, that is, when T is set to 64 FFT points. An offset Δf is derived. Note that the number m of the receiving antenna 16 is omitted.

  Here, 3.2 μs is a period of 64 FFT points in the IEEE 802.11b standard. The complex number conversion unit 124 inverts the sign of Δf to derive a phase in the direction opposite to the frequency offset. Further, the derived phase is converted into a complex number having values of the in-phase component and the quadrature component, and these are output as the frequency offset signal 210.

  Further, a receiving apparatus 12 according to another aspect (hereinafter referred to as “modification”) will be described separately from the second embodiment described above. The receiving device 12 according to the modified example is the same as the receiving device 12 of FIG. 5 according to the first embodiment, and the first processing unit 52a according to the modified example is the same as the first processing unit 52a of FIG. Since the frequency offset estimation unit 82 according to the modification is the same as the frequency offset estimation unit 82 of FIG.

  FIG. 13 shows a configuration of the phase difference calculation unit 110 according to the modification. The phase difference calculation unit 110 includes a replica generation unit 114, a complex conjugate unit 118, and a multiplication unit 120. Here, the baseband received signal 202 and the phase error signal 208 are shown together as one signal, but this is a display for convenience, and actually, the first baseband received signal 202a and the first phase error are as shown in FIG. It is configured by a signal such as the signal 208a, and the configuration of the phase difference calculation unit 110 also corresponds to it. Furthermore, it is assumed that the baseband reception signal 202 and the phase error signal 208 here are signals corresponding to the m-th reception antenna 16. The replica generation unit 114 generates a replica signal similarly to the replica generation unit 114 of FIG.

変形例のオフセット計算部１１２は、位相誤差ｙ_ｍ（ｔ）にもとづいて、以下のとおりに計算して周波数オフセットΔｆを導出する。なお、受信用アンテナ１６の番号ｍは省略した。

The complex conjugate unit 118 derives a complex conjugate value of the value of the replica signal generated by the replica generation unit 114. Multiplier 120 multiplies the value of baseband received signal 202 by the complex conjugate value of the replica signal corresponding to baseband received signal 202. The multiplication result is shown as follows and output as the phase error signal 208.

Based on the phase error y _m (t), the offset calculation unit 112 of the modification calculates as follows to derive the frequency offset Δf. Note that the number m of the receiving antenna 16 is omitted.

FIG. 14 shows a simulation result of the reception characteristics of the second embodiment. The simulation shows Eb / N0 vs. BER characteristics when a frequency offset of 50 kHz is given. Here, the circles indicate the characteristics when no frequency offset is given (hereinafter referred to as “reference characteristics”). If the simulation results when the frequency offset is given approach the circle, the frequency offset is corrected with high accuracy. It can be said that it is made. Also, Δ marks indicate characteristics when the frequency offset estimation of the second embodiment is obtained by complex multiplication instead of subtraction of the real number of the phase error, that is, when derived as follows (hereinafter referred to as “complex number characteristics”). Called).

  □ indicates the frequency offset based on the first embodiment, ● indicates the frequency offset based on the modification, ◇ indicates the frequency offset based on the second embodiment It is a characteristic. According to the simulation result of FIG. 14, it can be said that the characteristics of the first embodiment, the characteristics of the second embodiment, and the characteristics of the modified example are close to the reference characteristics, and the frequency offset is estimated and corrected with high accuracy. On the other hand, the complex number characteristic is 2 to 3 dB worse than the reference characteristic.

In the case of complex characteristics, the denominator of division is the replica signal x ′ _m (t) as shown in Equation 22. Therefore, when the absolute value of the replica signal is small and the Gaussian noise is large, the division result of x _m (t) and x ′ _m (t) is more greatly affected by the Gaussian noise. That is, Gaussian noise is emphasized and the characteristics deteriorate. On the other hand, when the real number of the phase error is subtracted as in the second embodiment, the characteristics are not deteriorated because the Gaussian noise is not emphasized.

  According to the embodiment of the present invention, the replica signal is generated based on the channel characteristic estimated from the received signal and the known signal, and the frequency offset is estimated from the difference in phase error between the received signal and the replica signal. The frequency offset can be estimated even if the signal does not have a predetermined periodicity. In addition, since the frequency offset can be estimated even if the training signal does not have a predetermined periodicity, the frequency offset estimation and the reception weight vector can be performed even when the reception weight vector is calculated from a training signal without the predetermined periodicity. The training signal for the calculation can be shared, and the use efficiency of the burst signal can be improved. In addition, since it is possible to cope with an increase in the number of data sequences transmitted in parallel by a plurality of antennas in a MIMO system, it is possible to improve the data transmission speed and transmission quality in the MIMO system. In addition, since the frequency offset is estimated by subtracting the real number of the two phase errors, the influence of noise can be suppressed and the frequency offset can be estimated with high accuracy.

  The present invention has been described based on the embodiments. This embodiment is an exemplification, and it will be understood by those skilled in the art that various modifications can be made to combinations of the respective constituent elements and processing processes, and such modifications are also within the scope of the present invention. is there.

  In the first and second embodiments of the present invention, the transmission device 10, the reception device 12, the processing unit 52, the transmission path characteristic estimation unit 80, and the frequency offset estimation unit 82 are configured to correspond to the MIMO system. However, the present invention is not limited to this, and may be configured to be applied to a normal communication system that is not a MIMO system, for example, a wireless LAN of the IEEE802.11a standard. In that case, it is only necessary to demodulate one series of signals instead of the L series of signals as in the first and second embodiments. According to this modification, the present invention can be applied to various communication systems. That is, it is only necessary that the received signal includes a frequency offset.

  In the first and second embodiments of the present invention, the transmission device 10, the reception device 12, and the processing unit 52 are configured to support multicarrier signals such as OFDM. However, the present invention is not limited thereto, and may be configured to be applied to a communication system that performs communication using a single carrier signal, for example, a wireless LAN of the IEEE802.11b standard, a simple mobile phone system, a mobile phone system, or a third generation mobile phone system. Good. According to this modification, the present invention can be applied to various communication systems. That is, it is only necessary that the received signal includes a frequency offset.

  In Embodiments 1 and 2 of the present invention, M frequency offsets corresponding to the mth receiving antenna 16 are estimated. However, the present invention is not limited to this. For example, one frequency offset may be estimated. In that case, statistical processing such as averaging may be performed on the estimated M frequency offsets to calculate one frequency offset, or one of the estimated M frequency offsets may be selected, or One frequency offset may be estimated from the beginning by an arbitrary method. According to this modification, when the frequency oscillators connected to the M receiving antennas 16 are common, it is possible to prevent deterioration of characteristics without estimating M types of frequency offsets. Further, if the M types of frequency offsets are statistically processed, the influence of noise can be further suppressed, so that the characteristics can be improved. That is, it may be estimated according to the configuration of the receiving device 12.

  A configuration in which the first and second embodiments of the present invention are arbitrarily combined is also effective. According to this modification, the effect obtained by combining the first and second embodiments can be obtained.

It is a figure which shows the spectrum of the multicarrier signal which concerns on Example 1. FIG. 1 is a diagram illustrating a configuration of a communication system according to a first embodiment. It is a figure which shows the structure of the burst format which concerns on Example 1. FIG. It is a figure which shows the structure of the transmitter of FIG. It is a figure which shows the structure of the receiver of FIG. It is a figure which shows the structure of the 1st radio | wireless part of FIG. It is a figure which shows the structure of the 1st process part of FIG. It is a figure which shows the structure of the frequency offset estimation part of FIG. It is a figure which shows the structure of the phase difference calculation part of FIG. FIG. 9 shows the relationship between time and phase error in the offset calculator of FIG. It is a flowchart which shows the procedure of the reception process of FIG. FIG. 10 is a diagram illustrating a configuration of an offset calculation unit according to the second embodiment. FIG. 10 is a diagram illustrating a configuration of a phase difference calculation unit according to another aspect of the second embodiment. It is a figure which shows the simulation result of the receiving characteristic of Example 1 and Example 2. FIG.

Explanation of symbols

  10 transmitting apparatus, 12 receiving apparatus, 14 transmitting antenna, 16 receiving antenna, 20 data separating section, 22 modulating section, 24 radio section, 26 controlling section, 28 error correcting section, 30 interleave section, 32 preamble adding section, 34 IFFT unit, 36 GI unit, 38 quadrature modulation unit, 40 frequency conversion unit, 42 amplification unit, 50 radio unit, 52 processing unit, 54 demodulation unit, 56 data combination unit, 58 control unit, 60 LNA unit, 62 frequency conversion unit 64 quadrature detection unit, 66 AGC, 68 AD conversion unit, 70 initial frequency offset correction unit, 80 transmission path characteristic estimation unit, 82 frequency offset estimation unit, 84 FFT, 86 reception weight calculation unit, 88 multiplication unit, 90 delay unit , 92 multiplier, 94 multiplier, 96 add Arithmetic unit, 100 communication system, 110 phase difference calculation unit, 112 offset calculation unit, 114 replica generation unit, 116 division unit, 118 complex conjugate unit, 120 multiplication unit, 122 delay unit, 124 complex number conversion unit, 126 multiplication unit, 128 Adder, 200 radio reception signal, 202 baseband reception signal, 204 composite signal, 206 transmission path characteristic signal, 208 phase error signal, 210 frequency offset signal.

Claims (9)

A receiving unit for sequentially receiving a sequence of transmitted known signals as a received signal via a transmission line;
A transmission path characteristic estimation unit that estimates transmission path characteristics based on the sequence of the received signal and the known signal received sequentially;
A generating unit that generates a plurality of replica signals respectively corresponding to the sequentially received reception signals from the estimated transmission path characteristics and the known signal sequence;
A phase error between one of the sequentially received signals and a replica signal corresponding to one of the sequentially received signals is derived, and a plurality of phases over at least a section of the series of known signals. A derivation unit for deriving an error;
A frequency offset estimator for estimating a frequency offset included in the sequentially received signal based on the plurality of derived phase errors;
A frequency offset estimation apparatus comprising:   The derivation unit divides one value of the sequentially received signal by a replica signal value corresponding to one of the sequentially received signals, and derives a phase component of the division result as the phase error. The frequency offset estimation apparatus according to claim 1.   The derivation unit multiplies one value of the sequentially received signal by a complex conjugate value of a replica signal value corresponding to one of the sequentially received signals, and uses a phase component of the multiplication result as the phase error. The frequency offset estimation apparatus according to claim 1, wherein the frequency offset estimation apparatus is derived.   4. The frequency offset estimation unit estimates a frequency offset included in the sequentially received signal from the plurality of derived phase errors by a least square method. 5. Frequency offset estimation device.   The frequency offset estimator selects predetermined two of the plurality of derived phase errors, and derives a difference in phase error between the selected two by subtraction, thereby obtaining the sequentially received reception signal. The frequency offset estimation apparatus according to claim 1, wherein the included frequency offset is estimated. The receiving unit sequentially receives a plurality of known signal sequences respectively transmitted from a plurality of transmitting antennas as a plurality of received signals by a plurality of receiving antennas, and the plurality of known signal sequence patterns are different from each other. And
The transmission path characteristic estimation unit is configured to combine each of the plurality of transmission antennas and each of the plurality of reception antennas based on a sequence of the plurality of reception signals sequentially received and a plurality of known signals having different patterns. For each, estimate multiple transmission line characteristics,
The generating unit generates a plurality of replica signals corresponding to each of the plurality of receiving antennas from a sequence of a plurality of known signals with different estimated channel characteristics and the pattern,
The deriving unit derives the plurality of phase errors corresponding to each of the plurality of receiving antennas,
The frequency offset estimation apparatus according to claim 1, wherein the frequency offset estimation unit estimates the frequency offset corresponding to each of the plurality of reception antennas.   The frequency offset estimation apparatus according to claim 6, wherein the frequency offset estimation unit derives one frequency offset by statistical processing from the frequency offset estimated corresponding to each of the plurality of reception antennas. A receiving unit for sequentially receiving a sequence of transmitted known signals as a received signal via a transmission line;
A transmission path characteristic estimation unit that estimates transmission path characteristics based on the sequence of the received signal and the known signal received sequentially;
A generating unit that generates a plurality of replica signals respectively corresponding to the sequentially received reception signals from the estimated transmission path characteristics and the known signal sequence;
A phase error between one of the sequentially received signals and a replica signal corresponding to one of the sequentially received signals is derived, and a plurality of phases over at least a section of the series of known signals. A derivation unit for deriving an error;
A frequency offset estimator for estimating a frequency offset included in the sequentially received signal based on the plurality of derived phase errors;
A correction unit that corrects a frequency offset included in a signal to be received by the reception unit following the reception signal sequentially received by the estimated frequency offset;
A processing unit for processing the signal corrected for the frequency offset;
A receiving apparatus comprising:   A sequence of transmitted known signals is sequentially received as a received signal via a transmission path, a transmission path characteristic is estimated based on the received signal and the sequence of the known signal, and the estimated transmission path characteristic A plurality of replica signals respectively corresponding to the sequentially received reception signals are generated from the known signal series, and the received reception signals and a plurality of replicas are received at least over a part of the known signal series. A frequency offset estimation method comprising: deriving a plurality of phase errors with respect to a signal and estimating a frequency offset included in the received reception signal based on the derived plurality of phase errors.

Images