JP2007300337A - Mobile receiver - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To further improve reception characteristics of a digital terrestrial broadcasting receiver by compensating a multiple Doppler shift by combining adaptive directivity control and a diversity composition with each other. <P>SOLUTION: When respective incoming waves are extracted from a plurality of incoming waves affected by a Doppler shift caused in a mobile reception environment, an adaptive directivity calculating unit 340 having a directivity control unit 300 compensates the Doppler shift by controlling directivity adaptively by the incoming waves according to convergence conditions calculated by a convergence condition calculating unit 330 and performing AFC. Further, a diversity composing unit 400 performs transmission line estimation after fast Fourier transforming (FFT) the respective incoming waves with the Doppler shift compensated, and then performs maximum ratio composition so that a composite signal of the respective incoming waves has a maximum C/N. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a reception technique in the field of digital broadcasting and digital transmission that employs an OFDM (Orthogonal Frequency Division Multiplexing) system, and more particularly, multipath generated when receiving radio waves such as digital broadcasting and wireless LAN (Local Area Network). The present invention relates to a mobile receiver using adaptive array antenna technology and diversity combining technology for removing the effects of fading and interference waves.

  Non-Patent Documents 1 and 2 describe conventional adaptive array antenna technologies. These documents include a single carrier PSK (Phase Shift Keying) and multi-value QAM (Quadrature Amplitude Modulation) digital transmission schemes such as adaptive array antennas and transmission paths using MMSE algorithms, LMS algorithms, or RLS algorithms. The generator is described in detail.

  Non-patent Documents 3 and 4 and Patent Document 1 describe adaptive array antenna techniques for mobile reception when receiving an OFDM signal. Specifically, the technique of Non-Patent Document 3 pays attention to the fact that the first guard interval and the rear guard interval in the OFDM symbol are the same, and mobile reception is performed by separately receiving incoming waves using an MMSE adaptive array. It is possible to compensate for the degradation of the reception quality due to. The technique of Non-Patent Document 4 cancels the moving speed of a moving body by estimating a received signal at a point stationary with respect to the ground by a spatial interpolation process using an array antenna, and receives by moving reception. Quality degradation can be compensated. Further, the technique of Patent Document 1 enhances multipath resistance by dividing a received signal into a plurality of bands and performing synthesis based on maximum ratio combining for each divided band, and an adaptive array based on band division. The update rate of the directivity weight coefficient can be reduced, and the calculation load can be reduced.

  Non-patent documents 5 and 6 have been reported as mobile reception signal processing techniques other than adaptive array antennas. Specifically, the technique of Non-Patent Document 5 uses a method of combining diversity signals received by a plurality of antennas for each carrier, and collects the output of each receiving antenna and simultaneously reduces time fluctuation. By performing digital signal processing, it is possible to compensate for deterioration in reception quality due to mobile reception. In the technique of Non-Patent Document 6, fixed directional antennas are arranged before and after the moving direction of the moving body, and after estimating the Doppler frequency, the frequency deviation is compensated by AFC (Auto Frequency Control), and the above-mentioned non- This is combined with the diversity combining method disclosed in Patent Document 5, and thus it is possible to compensate for deterioration in reception quality due to mobile reception.

John Litva, Titus Kwok-Yeung Lo, “Digital Beam-forming in Wireless Communication”, Arttech House Publishers Simon Hykin, Hiroshi Suzuki, “Adaptive Filter Theory”, Science and Technology Publishing, 2000 Satoshi Hori, Nobuyoshi Kikuma, Naoki Inagaki, “Study on initialization of optimization algorithm in guard section MMSE adaptive array for OFDM”, Theory of Science, Vol. J-B88-B, no. 9 pp. 1821-1824 (2005-09) Nagai Norikazu, Takayanagi Eizo, Saito Masato, Okada Minoru, Yamamoto Heiichi, “A digital terrestrial broadcast receiver that compensates for Doppler spread by measuring the moving speed of an moving object using an array antenna”, Science theory, Vol. J-B88-B, no. 4 pp. 741-750 (2005-04) Ryota Yamazaki, Masayuki Takada, Hiroyuki Sumizumi, "Computer simulation evaluation of high-speed mobile reception characteristics in digital terrestrial broadcasting", 2005 Shingaku Sodai, B-5-129 (2005-09) Makoto Tsuruta, Hideo Kasami, Hidehiro Matsuoka, “A Study on Mobile Reception Characteristics of Terrestrial Digital Broadcasting on High-Speed Fading Transmission Lines”, 2005 Shingaku Sodai, B-1-265 JP 2005-117348 A

  In the aforementioned Non-Patent Documents 3 and 4 and Patent Document 1, it has already been proved that it is effective to apply adaptive directivity control by adaptive array to mobile reception. In Non-Patent Documents 5 and 6, it has already been proved that it is effective to apply diversity combining (Non-patent Document 5) or a combination of fixed directivity and diversity combining (Non-Patent Document 6) to mobile reception. ing. However, even if these techniques are used for a mobile receiver that moves at high speed, there is a problem that it is insufficient to improve the reception characteristics.

  Accordingly, the present invention has been made to solve the above-mentioned problems, and in a mobile receiver for digital terrestrial broadcasting, multiple Doppler shifts are compensated by combining adaptive directivity control and diversity combining, and reception characteristics are further improved. The goal is to improve.

  As means for solving the above problems, an array antenna unit configured by a plurality of antennas for receiving a plurality of incoming waves, and extracting one incoming wave from the received plurality of received waves by changing a directivity weighting factor. A directivity control unit that corrects a frequency shift due to the influence of the Doppler shift that occurs in a mobile reception environment with respect to the extracted incoming wave, and an FFT ( This is realized by a mobile receiver characterized by comprising a diversity combining unit that performs transmission channel estimation by performing Fast Fourier Transform), calculates a weighting factor, and performs in-phase combining after multiplying the signal after FFT. As a result, it is possible to combine a technique for adaptively controlling the directivity of an incoming wave to compensate for a Doppler shift and a technique for performing maximum ratio combining, and this combination can improve reception characteristics.

  In the mobile receiver according to the present invention, the directivity control unit inputs an orthogonal demodulation unit that performs orthogonal demodulation on a received incoming wave using a common local signal, and an output from the orthogonal demodulation unit. A Doppler frequency calculation unit that calculates a Doppler frequency associated with the influence of Doppler shift that occurs in the mobile reception environment, and a convergence condition calculation unit that calculates a convergence condition for updating a directivity weighting coefficient using the Doppler frequency. And an adaptive directivity calculation unit that updates a directivity weighting factor based on the calculated convergence condition, extracts one incoming wave from a plurality of incoming waves, and corrects a frequency shift with respect to the incoming wave. It is characterized by having.

  In the mobile reception device according to the present invention, the directivity control unit further includes a frequency error calculation unit that calculates a local oscillation frequency error of the common local signal used in the orthogonal demodulation unit, and the orthogonal demodulation unit The common local signal is corrected using the local oscillation frequency error calculated by the frequency error calculating unit, and orthogonal demodulation is performed using the corrected common local signal.

  In the mobile reception device according to the present invention, the Doppler frequency calculation unit calculates a Doppler frequency in the 0 degree direction using a directivity weighting factor in the 0 degree direction with respect to the traveling direction of the mobile reception device. A 180 degree direction Doppler frequency is calculated using a 180 degree direction directivity weight coefficient, and the convergence condition calculation unit uses the calculated Doppler frequency to calculate a convergence condition for updating the directivity weight coefficient. It is characterized by calculating.

  In the mobile receiver according to the present invention, the convergence condition calculating unit scans the frequency axis with an extraction filter in a band of a Doppler frequency range corresponding to the half-value width of the directivity pattern generated by the array antenna unit. The power peak is selected and the next power peak is selected except for the Doppler frequency range centered on the peak, and the shift value corresponding to the frequency shift due to the Doppler shift is converged. It is calculated as a condition.

  In the mobile receiver according to the present invention, the convergence condition calculation unit performs a process of selecting a power peak using an extraction filter, obtains a shift value in descending order of power, and determines one of the obtained shift values. A convergence condition is calculated in which one peak wave is a desired wave and the rest are unnecessary waves.

  As described above, according to the mobile receiver of the present invention, the multiple Doppler shift compensation technology of the directivity control unit and the maximum ratio synthesis technology of the diversity combining unit are combined to compensate for the multiple Doppler shift and to improve the reception characteristics. Further improvements can be made. This enables high-speed mobile reception of terrestrial digital broadcasting.

The best mode for carrying out the present invention will be described below in detail with reference to the drawings.
First, the system configuration of the mobile receiver according to the embodiment of the present invention will be described with reference to FIG. The mobile receiver 1 includes an array antenna unit 100, an RF (Radio Frequency) -IF (Intermediate Frequency) frequency conversion unit 200, a directivity control unit 300 that compensates for multiple Doppler shifts, and a diversity combining unit 400 that performs maximum ratio combining. These four processing units are included.

  The array antenna unit 100 includes A antennas arranged in various shapes (linear array, rectangular array, circular array, etc.), and receives a plurality of incoming waves. The RF-IF frequency conversion unit 200 inputs A reception signals via the array antenna unit 100, and converts the radio frequency (RF) into an intermediate frequency (IF) for each of the reception signals.

  The directivity control unit 300 receives the A signals converted by the RF-IF frequency conversion unit 200 and outputs L signals in response to the incoming waves. In this case, by changing the directivity weight coefficient, one incoming wave is extracted from a plurality of incoming waves, AFC is performed on the incoming wave, and the Doppler shift is compensated so that the Doppler frequency becomes 0 Hz. The directivity control unit 300 changes the directivity weighting coefficient from moment to moment according to a change in the state of the transmission path.

  Further, diversity combining section 400 receives L signals from directivity control section 300, performs transmission path estimation for each input signal after FFT, and C / N (Carrier to Noise ratio) of the combined signal of each input. The maximum ratio synthesis is performed so that is maximized.

  Here, the relationship between the OFDM modulation signal and the guard interval is shown in FIG. The OFDM modulation signal is composed of a guard interval and an effective symbol interval. As a guard interval, a signal having the same waveform as the end of the effective symbol interval is added to the beginning of the effective symbol interval, thereby preventing intersymbol interference caused by waves arriving with a delay time within the guard interval length. Degradation is suppressed.

  Hereinafter, the original guard interval is referred to as HGI (Head Guard Interval), and the last guard interval of the effective symbol interval is referred to as TGI (Tail Guard Interval).

[Directivity control unit (Example 1)]
Hereinafter, the directivity control unit 300 illustrated in FIG. 1 will be described in detail. This directivity control unit 300 operates the directivity weight coefficient calculation unit 342 described later so that the difference between these two section signals is minimized by utilizing the fact that HGI and TGI are originally the same signal. Let If the incoming wave that is synchronized is the desired wave, only the desired wave is received, and the two signals are completely the same in a state where no unnecessary wave is received. Therefore, by determining the directivity weighting coefficient so that the difference between these two section signals is minimized, the combined signal output from the directivity control unit 300 can be a signal of only a desired wave.

  The configuration of the first example (Example 1) of the directivity control unit 300 is shown in FIG. The directivity control unit 300 includes an orthogonal demodulation unit 310, a Doppler frequency calculation unit 320, a convergence condition calculation unit 330, and an adaptive directivity calculation unit 340. The orthogonal demodulation unit 310 receives A signals from the RF-IF frequency conversion unit 200 and performs orthogonal demodulation on each signal. The Doppler frequency calculation unit 320 calculates the Doppler frequency using the directivity pattern in the 0 degree direction and the directivity pattern in the 180 degree direction with respect to the traveling direction of the mobile reception device 1 that is a moving body. The convergence condition calculation unit 330 uses the Doppler frequency calculated by the Doppler frequency calculation unit 320 to calculate a convergence condition in which one wave is a desired wave and the rest is an unnecessary wave. The adaptive directivity calculation unit 340 calculates an optimal directivity weighting coefficient in accordance with the convergence condition calculated by the convergence condition calculation unit 330, extracts one incoming wave from a plurality of incoming waves, and the frequency shift of the incoming wave. To compensate for the Doppler shift. As a result, the directivity control unit 300 outputs a signal compensated for Doppler shift (a signal of only a desired wave) to the diversity combining unit 400.

(Quadrature demodulator)
The orthogonal demodulator 310 includes a common local signal generator 311, a distributor 312, a BPF (Band Path Filter) 313-1 to A, a demodulator 314-1 to A, and an LPF (Low Path Filter) 315-1 to A. ing. The quadrature demodulator 310 receives each signal converted by the RF-IF frequency converter 200. These signals pass through the BPF 313-1 to A, and are demodulated by the demodulation units 314-1 to A using the common local signal generated by the common local signal generation unit 311 and distributed by the distribution unit 312. . Thereby, orthogonal demodulation section 310 can obtain complex baseband IQ signals (hereinafter referred to as baseband IQ signals) corresponding to the number of antennas (A) of array antenna sections 100 installed. The baseband IQ signal is band-limited by the LPF 315-1 to A-A and A / D converted (not shown). The common local signal generated by the common local signal generator 311 may be a signal independent of the system clock of the mobile reception device 1.

  Then, the A baseband IQ signals are sent to the Doppler frequency calculation unit 320, the convergence condition calculation unit 330, and the adaptive directivity calculation unit 340. Further, in the Doppler frequency calculation unit 320, a difference component calculation unit 325 described later calculates a Doppler frequency ((a) in FIG. 3), and in the convergence condition calculation unit 330, a convergence condition calculation unit 334 described later calculates adaptive directivity. The convergence condition used by the unit 340 is calculated, and the adaptive directivity calculation unit 340 adaptively controls the directivity based on the convergence condition to perform AFC, and then sends a signal to the diversity combining unit 400.

[Doppler frequency calculator]
The Doppler frequency calculation unit 320 includes a 0-degree direction directivity weight coefficient unit 321-1, a 180-degree direction directivity weight coefficient unit 321-2, a complex multiplication unit 322-1-1 to 2-A, and an array synthesis unit 323-1. , 2, Doppler frequency estimation units 324-1 and 324-2, and difference component calculation unit 325. In this Doppler frequency calculation section 320, the 0 degree direction directivity weight coefficient section 321-1 and the 180 degree direction directivity weight coefficient section 321-2 are each in the 0 degree direction with respect to the traveling direction of the mobile reception device 1 which is a moving body. Then, a directivity pattern (directivity weighting coefficient) is generated in the direction of 180 degrees and the incoming wave is separated and received, and then the Doppler frequency estimation unit 324 shown in FIG. 4 estimates the Doppler frequency. The Doppler frequency includes a frequency shift caused by a local oscillation frequency error.

  Complex multipliers 322-1 to 32-1 to A perform complex multiplication on the 0-degree directional weighting factor generated by 0-degree directional weighting factor unit 321-1 and the baseband IQ signal, respectively, and array synthesizer 323. -1 combines these signals. Similarly, complex multipliers 322-2-1 to 322-1 to A multiply the 180-degree directional directivity weighting coefficient generated by 180-degree directional directivity weighting coefficient unit 321-2 and the baseband IQ signal, respectively, and perform array synthesis. The unit 323-2 combines these signals.

  As shown in FIG. 4, the Doppler frequency estimation units 324-1 and 324-2 have effective symbol length delay units 324-11-1 and 2, multiplication units 324-12-1 and 2, moving average units 324-13-1, 2 and Doppler frequency calculation unit 324-14. When the Doppler frequency estimation units 324-1 and 324-2 receive the baseband IQ signals synthesized by the array synthesis units 323-1 and 323-2, the effective symbol length delay units 324-11-1 The effective symbol length delay unit 324-11-2 delays the Q-axis data of the baseband IQ signal by the effective symbol period. Multiplying section 324-12-1 multiplies the I-axis data and the delayed I-axis data, and moving average section 324-13-1 uses the multiplied signal to determine the I-axis data and the effective symbol period. For the correlation with the I-axis data delayed by a certain amount, the moving average of the guard period is calculated, and the moving average Sii (i) is output to the Doppler frequency calculation unit 324-14. Similarly, the multiplication unit 324-12-2 multiplies the I-axis data and the delayed Q-axis data, and the moving average unit 324-13-2 determines the I-axis data and the effective symbol based on the signal after the multiplication. For the correlation with the Q-axis data delayed by the period, the moving average of the guard period is calculated, and the moving average Siq (i) is output to the Doppler frequency calculation unit 324-14.

このようにして、ドップラー周波数推定部３２４−１が０度方向の（＋）ドップラー周波数を推定し、ドップラー周波数推定部３２４−２が１８０度方向の（−）ドップラー周波数を推定する。そして、差分成分算出部３２５は、ドップラー周波数推定部３２４−１により推定された（＋）ドップラー周波数、及びドップラー周波数推定部３２４−２により推定された（−）ドップラー周波数を入力し、（±）ドップラー周波数の差分をとって絶対値を算出する。これにより、収束条件算出部３３０が収束条件を算出するために使用するドップラー周波数（図３の（ａ））が決定される。 The Doppler frequency calculation unit 324-14 calculates the Doppler frequency for each i symbol (1 OFDM symbol) using the following equation.

In this way, the Doppler frequency estimation unit 324-1 estimates the (+) Doppler frequency in the 0 degree direction, and the Doppler frequency estimation unit 324-2 estimates the (−) Doppler frequency in the 180 degree direction. Then, the difference component calculation unit 325 inputs (+) Doppler frequency estimated by the Doppler frequency estimation unit 324-1 and (−) Doppler frequency estimated by the Doppler frequency estimation unit 324-2, and (±) The absolute value is calculated by taking the difference in Doppler frequency. Thereby, the Doppler frequency ((a) of FIG. 3) used for the convergence condition calculation unit 330 to calculate the convergence condition is determined.

[Convergence condition calculation unit]
The convergence condition calculation unit 330 includes an omnidirectional weight coefficient unit 331, a complex multiplication unit 332-1 to A, an array synthesis unit 333, and a convergence condition calculation unit 334. The convergence condition calculation unit 330 performs a predetermined filtering process on the combined signal generated by the complex operation of the omnidirectional weighting factor and the baseband IQ signal, calculates the convergence condition, and converts the convergence condition into diversity. The data is output to the synthesis unit 400. This will be specifically described below.

  The complex multipliers 332-1 to 332 -A complex multiply the omnidirectional weight coefficient from the omnidirectional weight coefficient unit 331 and the baseband IQ signal from the orthogonal demodulator 310, respectively. Is synthesized. The convergence condition calculation unit 334 receives the combined signal from the array combining unit 333 and the Doppler frequency from the Doppler frequency calculation unit 320, and performs the filtering process shown in FIG. FIG. 5 is a diagram for explaining filter processing by the convergence condition calculation unit 334. The convergence condition calculation unit 334 uses a narrowband power sensor having a bandwidth W [Hz] of the Doppler frequency corresponding to the half-value width of the directivity pattern generated by the array antenna unit 100 to perform a frequency interval between fd1 and fd2. Record power and offset frequency while scanning. fd1 is the (+) Doppler frequency estimated by the Doppler frequency estimation unit 324-1, and fd2 is the (−) Doppler frequency estimated by the Doppler frequency estimation unit 324-2. Specifically, when the convergence condition calculation unit 334 selects the first power peak in the scan from fd1 to fd2, the bandwidth W [Hz] around the frequency of the first peak. The next second peak is searched for except. In this way, the power of the peak signal and its offset frequency are recorded.

ｆｄａは、θをθ１からθ２まで変えたときに得られる最小のドップラー周波数であり、ｆｄｂは、θをθ１からθ２まで変えたときに得られる最大のドップラー周波数である。すなわち、収束条件計算部３３４は、ドップラー周波数算出部３２０から入力したドップラー周波数からＦｄを得て、前述の（２）式により、指向性パターンの半値幅に相当するドップラー周波数の帯域幅Ｗ［Ｈｚ］を算出する。 Here, the relationship between the half-value width of the directivity pattern and the bandwidth of the Doppler frequency will be described. FIG. 6 is a diagram illustrating a relationship between the traveling direction of the mobile reception device 1 that is a moving body and the directivity of the antenna. The traveling direction is 0 degree, and the central angle of the antenna directivity is θ. θ is set to 0 to 360 degrees clockwise. When θ = 0 degrees, the maximum Doppler frequency is Fd, and when θ = 180 degrees, the maximum Doppler frequency is −Fd. Here, the larger absolute value of fd1 and fd2 obtained by the Doppler frequency estimation units 324-1 and 324-2 is defined as Fd. The circle shown in FIG. 6 is a circle at a level that becomes the half-value width of the antenna directivity, and the angles θ1 and θ2 for obtaining the half-value width can be obtained from the intersection of the directivity pattern and this circle. The bandwidth W [Hz] of the Doppler frequency corresponding to the half width of the directivity pattern can be obtained by the following equation.

fda is the minimum Doppler frequency obtained when θ is changed from θ1 to θ2, and fdb is the maximum Doppler frequency obtained when θ is changed from θ1 to θ2. That is, the convergence condition calculation unit 334 obtains Fd from the Doppler frequency input from the Doppler frequency calculation unit 320, and calculates the bandwidth W [Hz] of the Doppler frequency corresponding to the half-value width of the directivity pattern according to the above equation (2). ] Is calculated.

The convergence condition calculation unit 334 records the peak signal power and its offset frequency, and then sequentially sets the frequency offset value by the number of inputs (L) of the adaptive directivity calculation unit 340 in descending order of power. Extraction is performed, and a convergence condition is determined in which one of the L waves is a desired wave and the remaining (L-1) waves are unnecessary waves. Taking FIG. 5 as an example, in the scan between fd1 and fd2, after recording the power and offset frequency at each peak, _L signals P _{1 to} P _L are extracted from the one with the highest power, and P ₁ can peak directivity in "fd1 direction convergence condition for the unnecessary wave of the desired wave and _P 2 to P _L, the remaining _{(P 2, ···, P L} ) as a null directivity pattern Create ". Similarly, convergence conditions are created in which P _{2 to} P _L are desired waves and the remaining waves are unnecessary waves. In this way, the convergence condition is created for the number of inputs (L) of the adaptive directivity calculation unit 340.

  When the convergence condition calculation unit 330 calculates L convergence conditions, the adaptive directivity calculation unit 340 then updates the weighting factor based on each convergence condition, and receives and receives a plurality of incoming waves separately. Doppler shift compensation is performed by AFC.

[Adaptive directivity calculation unit]
The adaptive directivity calculation unit 340 includes an HGI extraction unit 341-1-1 to LA, a directivity weight coefficient calculation unit 342-1 to L, a complex multiplication unit 343-1-1 to LA, and an array synthesis unit 344. -1 to L, TGI extraction units 345-1 to 345-1 and AFC circuits 346-1 to 346-1. The HGI extraction units 341-1-1 to LA perform a HGI extraction process on each baseband IQ signal from the orthogonal demodulation unit 310, that is, a process of extracting HGI from the OFDM modulated signal. More specifically, a moving average for the guard interval between the signal in FIG. 2 and the signal obtained by delaying the signal in FIG. 2 by the effective symbol length is obtained to detect a correlation output peak (a symbol boundary becomes a peak). The guard interval length is extracted from the effective symbol interval length based on the peak. The HGI extraction units 341-1-1 to LA output the baseband IQ signals X _n (t) from which the HGIs are extracted to the directivity weight coefficient calculation units 342-1 to 342 -L, respectively.

On the other hand, the TGI extraction units 345-1 to 345 -L perform TGI extraction processing on the weighted synthesized signals obtained by the complex multiplication units 343-1-1 to LA and the array synthesis units 344-1 to 344 -L. That is, processing for extracting the TGI from the OFDM modulated signal is performed. Specifically, as in the HGI extraction process, the correlation output peak is detected, and the guard interval length is extracted from the effective symbol interval length based on the peak. The TGI extraction units 345-1 to 345-1 output the weighted composite signals y _n (t) obtained by extracting the TGIs to the directivity weight coefficient calculation units 342-1 to 342 -L, respectively.

アレーの合成出力は以下のようになる。

但し、上添字Ｔ，Ｈはそれぞれ転置、共役転置を表す。また、下添字ｎ，ｋはそれぞれダイバーシティ合成部４００への入力番号（詳細については後述する）、キャリア番号を表す。 If the vector notation of the guard interval signal and the weighting coefficient is defined as follows,

The combined output of the array is as follows:

However, the superscripts T and H represent transposition and conjugate transposition, respectively. Further, the subscripts n and k represent an input number (details will be described later) and a carrier number to the diversity combining unit 400, respectively.

但し、Ｅ［・］は期待値演算である。最適重み係数ベクトルは、

で与えられ、相関行列Ｒ_ｎｘｘ及び相関ベクトルｒ_ｎｘｒは次式で表される。

指向性重み係数算出部３４２−１〜Ｌは、収束条件算出部３３０から入力した収束条件を満たすように、（７）式を繰り返して計算する。具体的には、指向性重み係数算出部３４２−１〜Ｌは、入力信号の１シンボル区間のサンプル値を１０００とすると、ＨＧＩ及びＴＧＩを用いたＭＭＳＥアダプティブアルゴリズムにより、（７）式を繰り返して１０００回の計算を行い指向性重み係数を更新する。この処理により、（６）式のＥ（平均自乗誤差の期待値）が小さくなり、１シンボル区間の計算が終了した後には、所望波にメインビームを向け、不要波にヌルを向ける指向性パターンを作成することが可能な最適な指向性重み係数を得ることができる。例えば、図７に示すように、経験上得られる閾値を知っていれば、繰り返し回数Ｒで指向性重み係数の更新を停止することができる。そして、複素乗算部３４３−１−１〜Ｌ−Ａは、指向性重み係数算出部３４２−１〜Ｌから最適な指向性重み係数を入力し、当該最適な重み係数と直交復調部３１０からのベースバンドＩＱ信号とを複素乗算し、アレー合成部３４４−１〜Ｌは、複素乗算した信号を合成し、合成信号をＡＦＣ回路３４６−１〜Ｌにそれぞれ出力する。 Assuming that the error signal in the above two sections is e _n (t), the evaluation function to be minimized is expressed by the following equation. Reference numbers _r n (t) is a common practice to substitute _y n (t).

However, E [•] is an expected value calculation. The optimal weight coefficient vector is

The correlation matrix R _nxx and the correlation vector r _nxr are expressed by the following equations.

The directivity weight coefficient calculators 342-1 to 342-1 to L repeatedly calculate equation (7) so that the convergence condition input from the convergence condition calculator 330 is satisfied. Specifically, the directivity weight coefficient calculators 342-1 to 342-1 to L repeat equation (7) using the MMSE adaptive algorithm using HGI and TGI, assuming that the sample value of one symbol interval of the input signal is 1000. The directivity weighting coefficient is updated by performing 1000 calculations. By this processing, E (expected value of mean square error) in equation (6) is reduced, and after the calculation of one symbol section is completed, the directivity pattern in which the main beam is directed to the desired wave and null is directed to the unnecessary wave. Can be obtained. For example, as shown in FIG. 7, if the threshold value obtained from experience is known, the update of the directivity weight coefficient can be stopped at the number of repetitions R. The complex multipliers 343-1-1 to LA receive the optimal directional weight coefficients from the directivity weight coefficient calculators 342-1 to 342 -L, and The baseband IQ signal is subjected to complex multiplication, and array synthesis sections 344-1 to 344 -L synthesize the complex multiplied signals and output the synthesized signals to AFC circuits 346-1 to 346 -L, respectively.

  With the above processing, each incoming wave can be separated from a plurality of incoming waves. That is, the signals input to the AFC circuits 346-1 to 346-1 L become the signals of the respective incoming waves separated from the plurality of incoming waves. In addition, since each incoming wave can be separated as the beam directivity becomes sharper, the number of antennas is better.

  Next, the AFC circuits 346-1 to 346-1 L respectively input the separated incoming wave signals, and restore the frequency shift caused by the Doppler shift. The detailed processing of AFC is also described in Non-Patent Document 6 described above.

FIG. 8 is a diagram for explaining compensation for Doppler shift by AFC. In FIG. 8, the multiple Doppler shift is a combination of the incoming waves P _{1 to} P _n having their respective arrival angles and Doppler shifts. The complex multipliers 343-1-1 to LA and the array synthesizers 344-1 to 34 -L extract only the P _n wave by complex multiplication of the input signal X _n and the directivity weight coefficient W _n . After the P _n wave is extracted, the AFC circuits 346-1 to 346-1 -L offset the frequency shift (+ Fd) due to the Doppler shift to 0 Hz by AFC (offset the offset frequency of the extracted P _n wave to 0). ). By this process, the Doppler shift can be compensated, and by performing the same process from P _{1 to} P _n , the multiple Doppler shift can be compensated.

  As described above, according to the directivity control unit 300 of the first embodiment provided in the mobile reception device 1, the adaptive directivity calculation unit 340 minimizes the difference between the interval signals between the HGI and the TGI of the OFDM modulated signal. As described above, the directivity weight coefficient is determined to separate the incoming wave signal (the signal of only the desired wave), and the offset is set so that the offset frequency of the signal becomes zero. Thereby, the Doppler shift can be compensated.

但し、ｌはブランチ番号、ｉはシンボル番号、ｋはキャリア番号、Ｌはブランチ総数、Ｈ_ｌ ^＊は伝送路特性Ｈ_ｌの共役複素数を表す。
そして、同相合成部４０７は、複素乗算部４０６により複素乗算された信号につき、各ブランチの信号を同相合成する。そして、デマッピング部４０８は、この同相合成された信号をデマッピングする。 [Diversity synthesis department]
Next, the diversity combining unit 400 that performs diversity combining for each carrier will be described. FIG. 9 is a diagram illustrating a configuration of a diversity combining unit 400 including L branches. The diversity combining unit 400 includes an FFT unit 401, an SP extracting unit 402, a symbol filter 403, a carrier filter 404, a weight coefficient calculating unit 405, and a complex multiplying unit 406 for each branch, and further includes an in-phase combining unit 407 and a demapping unit. 408. The FFT unit 401 performs FFT on the OFDM signal, the SP extraction unit 402 extracts the SP, and the symbol filter 403 performs time direction interpolation processing (symbol filter). Then, the carrier filter 404 performs an interpolation process (carrier filter) in the frequency direction, and obtains a transmission path characteristic H ₁ for each carrier. The weighting factor calculation unit 405 calculates the weighting factor (W _l ) by the following formula. Complex multiplier 406, FFT signal and the weighting coefficient _{(W l)} complex multiplication by the FFT unit 401. Such a series of processing is performed for each branch.

Here, l is a branch number, i is a symbol number, k is a carrier number, L is the total number of branches, and H _l ^* is a conjugate complex number of the transmission path characteristic H _l .
Then, the in-phase synthesis unit 407 performs in-phase synthesis of the signals of each branch for the signal that is complex-multiplied by the complex multiplication unit 406. Then, the demapping unit 408 demaps the in-phase synthesized signal.

  Under the influence of multipath fading, mobile reception characteristics and multipath tolerance differ depending on the interpolation processing of the symbol filter 403. In the following, two types of transmission path estimation schemes of 4-symbol equalization and 1-symbol equalization will be described for the interpolation processing by the symbol filter 403. As shown in FIG. 10A, 4-symbol equalization is a method of linear interpolation using SPs that are 4 symbols apart in time. According to this method, it is easily affected by time variation. However, since transmission path estimation is performed using SPs interpolated for every three carriers in the frequency direction, it is difficult to be affected by a multipath wave having a long delay time in a multipath environment. On the other hand, as shown in FIG. 10B, the 1-symbol equalization is a method in which only interpolation in the frequency direction is performed without performing interpolation in the time direction. According to this method, it is strong against the influence of time fluctuation (not easily affected). However, since transmission path estimation is performed using only the SP for every 12 carriers in the frequency direction, it is easily affected by a multipath wave having a long delay time in a multipath environment.

  In this manner, the symbol filter 403 performs 4-symbol equalization for a long delay time multipath, and 1-symbol equalization for a short delay time multipath. It is desirable to do. This is described in JP-A-2005-45664.

  As described above, according to the diversity combining unit 400 included in the mobile reception device 1, by performing transmission path estimation after FFT for each input signal and calculating a weighting factor, the C / N ( The maximum ratio synthesis is performed so that (Carrier to Noise ratio) is maximized. Thus, by combining the multiple Doppler shift compensation technique by the directivity control unit 300 and the maximum ratio combining technique by the diversity combining unit 400, the multiple Doppler shift can be compensated and the reception characteristics can be further improved. It becomes possible.

[Directivity control unit (Example 2)]
Hereinafter, a second example (Example 2) of the directivity control unit 300 illustrated in FIG. 1 will be described in detail with reference to FIG. Comparing the directivity control unit 300 of the second embodiment with the directivity control unit 300 of the first embodiment shown in FIG. 3, the directivity control unit 300 of the second embodiment and the directivity control unit 300 of the first embodiment are , The orthogonal demodulator 310, the Doppler frequency calculator 320, the convergence condition calculator 330, and the adaptive directivity calculator 340. However, the directivity control unit 300 according to the second embodiment is different in that a frequency error calculation unit 350 is provided in addition to these configurations. That is, the directivity control unit 300 according to the second embodiment illustrated in FIG. 11 performs a quadrature demodulation unit 310 that performs quadrature demodulation on the input A signals, and 0 in the traveling direction of the mobile reception device 1 that is a moving body. One wave out of L signals output to the diversity combining unit 400 using the Doppler frequency and the Doppler frequency calculating unit 320 that calculates the Doppler frequency using the directivity pattern in the direction of 180 degrees and the directivity pattern in the direction of 180 degrees. , A convergence condition calculation unit 330 that calculates a convergence condition in which the desired wave is a desired wave and the remaining is an unnecessary wave, calculates an optimal directivity weighting coefficient according to the convergence condition, restores the frequency shift of the incoming wave, and compensates for the Doppler shift Difference between transmission frequency from adaptive directivity calculation unit 340 and transmission station and local local oscillation frequency (frequency deviation caused by local oscillation frequency error) Calculated and provided with a frequency error calculating unit 350 for correcting the common local signal.

  Referring to FIG. 11, frequency error calculation section 350 receives a baseband IQ signal (received signal of one antenna of array antenna section 100) that has been quadrature demodulated by quadrature demodulation section 310, and a transmission station in the input signal. Is calculated as a correction value and output to the common local signal generator 311 of the quadrature demodulator 310 as a correction value.

  FIG. 12 is a diagram showing the configuration of the frequency error calculation unit 350 shown in FIG. The frequency error calculation unit 350 includes effective symbol length delay units 351-1 and 352, multiplication units 352-1 and 352, moving average units 353-1 and 353-1, a common local frequency calculation unit 354, and a subtraction unit 355. A correction value for compensating for a local oscillation frequency error of the common local signal generated by the common local signal generation unit 311 of the demodulation unit 310 is generated.

  The effective symbol length delay unit 351-1 delays the I-axis data in the input baseband IQ signal by the effective symbol period, and the effective symbol length delay unit 351-2 delays the Q in the input baseband IQ signal. Delay axis data by valid symbol period. The multiplier 352-1 multiplies the I-axis data by the delayed I-axis data, and the moving average unit 353-1 delays the I-axis data and the effective symbol period by the multiplied signal. For the correlation with the I-axis data, the moving average of the guard period is calculated, and the moving average Sii (i) is output to the common local frequency calculator 354. Similarly, the multiplier 352-2 multiplies the I-axis data and the delayed Q-axis data, and the moving average unit 353-2 delays the I-axis data and the effective symbol period by the multiplied signal. For the correlation with the Q-axis data, the moving average of the guard period is calculated, and the moving average Siq (i) is output to the common local frequency calculator 354.

そして、減算部３５５は、その共通ローカル信号の周波数（ｆ１）と送信周波数（ｆ０）との差分をとり、共通ローカル信号の補正値（＝ｆ０−ｆ１）を決定する。直交復調部３１０の共通ローカル信号発生部３１１は、周波数誤差算出部３５０から補正値を入力し、共通ローカル信号を補正する。 The common local frequency calculation unit 354 calculates the frequency (f1) of the common local signal for each i symbol (1 OFDM symbol) using the following equation.

Then, the subtraction unit 355 takes the difference between the frequency (f1) of the common local signal and the transmission frequency (f0), and determines the correction value (= f0−f1) of the common local signal. The common local signal generator 311 of the orthogonal demodulator 310 receives the correction value from the frequency error calculator 350 and corrects the common local signal.

  As described above, according to the directivity control unit 300 of the second embodiment provided in the mobile reception device 1, the frequency error calculation unit 350 corrects the correction value for compensating for the frequency shift due to the local oscillation frequency error of the common local signal. The orthogonal demodulator 310 generates a common local signal corrected by the correction value, performs orthogonal transform, and outputs a baseband IQ signal. As a result, the reception characteristics can be further improved as compared with the directivity control unit 300 of the first embodiment.

1 is a system configuration diagram showing a mobile reception device according to an embodiment of the present invention. It is a figure explaining the relationship between an OFDM modulation signal and a guard area. It is a figure which shows the structure (Example 1) of a directivity control part. It is a figure which shows the structure of a Doppler frequency estimation part. It is a figure explaining an extraction filter process. It is a figure explaining the relationship between a Doppler frequency and a bandwidth. It is a figure for demonstrating the convergence process of a directivity weighting coefficient. It is a figure explaining the mode of Doppler shift compensation by AFC. It is a figure which shows the structure of a diversity synthetic | combination part. It is a figure explaining two types of transmission path estimation methods. It is a figure which shows the structure (Example 2) of a directivity control part. It is a figure which shows the structure of a frequency error calculation part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Mobile receiver 100 Array antenna part 200 RF-IF frequency conversion part 300 Directivity control part 310 Orthogonal demodulation part 311 Common local signal generation part 312 Distribution part 313 BPF
314 Demodulator 315 LPF
320 Doppler frequency calculation unit 321 Directivity weight coefficient unit 322 Complex multiplication unit 323 Array synthesis unit 324 Doppler frequency estimation unit 324-11 Effective symbol length delay unit 324-12 Multiplication unit 324-13 Moving average unit 324-14 Doppler frequency calculation unit 325 Difference component calculation unit 330 Convergence condition calculation unit 331 Nondirectional weight coefficient unit 332 Complex multiplication unit 333 Array synthesis unit 334 Convergence condition calculation unit 340 Adaptive directivity calculation unit 341 HGI extraction unit 342 Directivity weight coefficient calculation unit 343 Complex multiplication Unit 344 array combining unit 345 TGI extracting unit 346 AFC circuit 350 frequency error calculating unit 351 effective symbol length delay unit 352 multiplying unit 353 moving average unit 354 common local frequency calculating unit 355 subtracting unit 400 diversity combining unit 401 FFT unit 402 SP extracting unit 4 03 Symbol filter 404 Carrier filter 405 Weight coefficient calculation unit 406 Complex multiplication unit 407 In-phase synthesis unit 408 Demapping unit

Claims (6)

An array antenna unit configured with a plurality of antennas and receiving a plurality of incoming waves;
One arriving wave is extracted from a plurality of received arriving waves by changing the directivity weighting coefficient, and frequency deviation due to the influence of Doppler shift occurring in the mobile reception environment is corrected with respect to the extracted arriving waves. A directivity control unit,
A diversity combining unit that performs FFT on each incoming wave in which the frequency deviation is corrected, performs transmission path estimation, calculates a weighting coefficient, and multiplies the signal after FFT to perform in-phase combining; A mobile reception device characterized. The mobile receiver according to claim 1, wherein
The directivity control unit is
An orthogonal demodulation unit that performs orthogonal demodulation using a common local signal on the received incoming wave;
A Doppler frequency calculation unit that inputs an output from the quadrature demodulation unit and calculates a Doppler frequency associated with an influence of a Doppler shift that occurs in the mobile reception environment;
Using the Doppler frequency, a convergence condition calculation unit for calculating a convergence condition for updating the directivity weighting factor;
An adaptive directivity calculation unit that updates a directivity weighting factor based on the calculated convergence condition, extracts one incoming wave from a plurality of incoming waves, and corrects a frequency shift with respect to the incoming wave; A mobile receiver characterized by the above. The mobile receiver according to claim 2, wherein
The directivity control unit further includes a frequency error calculation unit that calculates a local oscillation frequency error of the common local signal used by the orthogonal demodulation unit,
The mobile receiver according to claim 1, wherein the orthogonal demodulator corrects a common local signal using the local oscillation frequency error calculated by the frequency error calculator, and performs orthogonal demodulation using the corrected common local signal. The mobile receiver according to claim 2 or 3,
The Doppler frequency calculation unit calculates a Doppler frequency in the 0 degree direction using a directivity weight coefficient in the 0 degree direction with respect to the traveling direction of the mobile receiver, and uses a directivity weight coefficient in the 180 degree direction with respect to the traveling direction. To calculate the Doppler frequency in the 180 degree direction,
The mobile reception apparatus, wherein the convergence condition calculation unit calculates a convergence condition for updating a directivity weighting factor using the calculated Doppler frequency. In the mobile receiver as described in any one of Claim 2 to 4,
The convergence condition calculation unit selects a power peak while scanning the frequency axis with an extraction filter in a band of a Doppler frequency range corresponding to the half-value width of the directivity pattern generated by the array antenna unit. A shift value corresponding to a frequency shift due to the influence of the Doppler shift is calculated as a convergence condition. Receiver device. The mobile receiver according to claim 5, wherein
The convergence condition calculation unit performs a process of selecting a power peak using an extraction filter, obtains a shift value in descending order of power, and uses the peak wave of one of the obtained shift values as a desired wave. A mobile reception apparatus that calculates a convergence condition for an unnecessary wave.

Images