JP2009287942A - Direction finder - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve time-difference direction finding by accurately determining a frequency gradient (group delay time) of a phase component of a cross spectrum by increasing an effective band width by autonomous frequency dispersion of received signals. <P>SOLUTION: A direction finder includes: first and second receiving antennas (1) arranged at a prescribed distance from each other for receiving incoming radio waves; a frequency dispersion part (3) for frequency dispersion of each frequency band of first and second received signals received by the first and second receiving antennas; a cross spectrum computing unit for computing the cross spectra of frequency-dispersed first and second received signals output from the frequency dispersion part; a time-difference detector (5) for detecting the time difference between incoming radio waves received by the first and second receiving antennas on the basis of a frequency gradient to the frequency of a phase in a frequency band having a high coherence level indicated by the cross spectra; and a direction computing unit (6) for computing the incoming direction of radio waves on the basis of the time difference and the prescribed distance. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an azimuth detection device for a radio wave source, and more particularly to an apparatus for detecting an azimuth using a difference in arrival time of radio waves. An azimuth detection device according to the present invention is incorporated in a radio wave detection device or the like mounted on an aircraft, a ship or the like, and detects the arrival direction of a radio wave source such as a radar device mounted on an opponent aircraft or a ship or the like with high accuracy Used.

  FIG. 14 shows a block diagram of a conventional azimuth detection apparatus using a difference in arrival time of general radio waves. About this structure, it is disclosed by Unexamined-Japanese-Patent No. 4-212081 (patent document 1), for example. In the figure, 1 is a receiving antenna for receiving radio waves, 2 is a high-frequency amplifier that amplifies a received signal, 12 is a detection amplifier that detects the amplified signal, 15 is a pulse rise detection circuit, 17 is a time counter, and 9 is a time counter. 17 is an azimuth calculation circuit that detects an arrival azimuth from the time difference data obtained in 17 and the interval between a pair of receiving antennas 1.

The operation principle of the above azimuth detection device is shown in FIG. Let d be the distance between the two receiving antennas 1. Since a path difference occurs according to the arrival direction θ, the arrival times of the radio waves arriving at the two receiving antennas are different. For this reason, arrival time difference (tau) arises between two receiving antennas. The arrival azimuth θ and the time difference τ have a relationship as shown in Expression (1).
sin θ = cτ / d (1)
Here, let c be the velocity of radio waves. The arrival direction of the radio wave can be calculated by measuring the time difference τ using the relationship of the formula (1).

  In the above azimuth detection apparatus, after the reception signal received by the reception antenna 1 is processed by the high frequency amplifier 3 and the detection amplifier 12, the rising timing is detected by the pulse rising detection circuit 15, and the detected timing and the receiving antenna 2 are detected. The time difference of the rise is measured by the time counter 17 from the timing detected in the same manner. The rise time difference is output to the azimuth calculation circuit 9 as the arrival time difference, and the arrival azimuth θ is calculated according to the equation (1).

  With this configuration, it is impossible to perform time difference measurement with an accuracy equal to or greater than the counter interval of the time counter 17, and there is a problem that the accuracy of time difference measurement does not improve beyond a certain level. Further, when the signal rises off, or between the reception signal processing channel of the reception antenna 1 and the channel of the reception antenna 2, there is a difference in the gain of the reception antennas 1 and 2 and the gain of the high-frequency amplifiers 12 and 13, When there is a difference in the amplitude of the received signal input to the pulse rising detection circuits 15 and 16, the timing at which the received signal waveform exceeds the amplitude threshold for detecting the rising timing, that is, the position of the signal waveform on the time axis is There is also a problem that time difference measurement errors are likely to occur due to differences between channels.

  Therefore, Japanese Patent Laid-Open No. 9-257902 (Patent Document 2) attempts to solve the above problem and obtain a highly accurate azimuth detection technique by using the phase of the cross spectrum. In this technology, for example, an incoming radio wave is received by two channels (ChA, ChB), and the received radio wave A and the frequency gradient of the cross spectrum phase component of the radio wave B including the arrival delay time with respect to A, that is, the group delay, The arrival delay time is calculated and the arrival direction of the radio wave is obtained. Hereinafter, an azimuth finding device according to this method is referred to as a conventional azimuth finding device.

Japanese Patent Laid-Open No. 4-212081 (FIG. 1) Japanese Patent Laid-Open No. 9-257902 (FIG. 1)

FIG. 16 is an explanatory diagram of the amplitude and phase of the cross spectrum. The calculation of the arrival delay time will be described with reference to this figure. In general, a received signal for calculating a cross spectrum is a signal including receiver noise. When a cross spectrum S ₀ ^D is calculated with a conventional azimuth detecting device in a signal including noise, the result is as shown in FIGS. 16 (a1) and 16 (a2). As apparent from FIGS. 16A1 and 16A2, in FIG. 3A2 showing the phase component φ ₀ ^D , a range in which the error is small and a frequency gradient can be obtained practically (this frequency range is represented by δf). Is) only the band portion of the cross spectrum S ₀ ^D and its periphery. When the fluctuation range of the phase component φ ₀ ^{D in} the range of δf is δφ ₀ ^D , the arrival time difference is calculated by (1 / 2π) × (δφ ₀ ^D / δf).

The range in which the frequency gradient can be obtained in practice also includes a minute error in the phase component. The error of the phase component and .delta..phi _n. The fluctuation range of the phase component φ ₀ ^D considering this error δφ _n is (δφ ₀ ^D + δφ _n ). The arrival time difference in this case is (1 / 2π) × ((δφ ₀ ^D + δφ _n ) / δf). Therefore, if δφ _n is constant, the error δφ _n / δf of the arrival time difference increases as δf decreases. For this reason, if the range for obtaining this practical frequency gradient is narrow, the frequency gradient obtained from this range has a large error (error of arrival time difference). As a result, the arrival delay time and the arrival direction of the radio wave to be finally obtained The error becomes larger. In particular, when the received signal is a radar pulse signal and the pulse width is long, the signal band is narrow and the cross spectrum bandwidth is also narrow, so that a large error is likely to occur. Therefore, in order to reduce the error between the arrival delay time to be finally obtained and the arrival direction of the radio wave, it is necessary to increase the range δf in which the practical frequency gradient can be obtained.

As described above, the wider the cross spectrum bandwidth, the larger the band portion, and the wider the range in which the frequency gradient can be obtained in practice. Therefore, in order to obtain the frequency gradient with high accuracy, it is necessary to increase the bandwidth of the cross spectrum. Note that the bandwidth of the cross spectrum ^SD matches the bandwidth of the incoming radio waves A and B (the incoming radio wave B includes the arrival delay time with respect to the incoming radio wave A, but the bandwidth is the same). .

  As described above, in the conventional azimuth detecting device, when the bandwidth of the incoming radio wave is small, the bandwidth of the cross spectrum becomes small, and the error of the frequency gradient of the phase component of the cross spectrum becomes large. There is a problem that the accuracy of the arrival delay time is deteriorated and the accuracy of the arrival direction finally obtained by the equation (1) is also deteriorated. Therefore, in the present invention, by increasing the bandwidth by the method described later, the range in which a practical frequency gradient can be obtained as shown in FIGS. 16 (b1) and (b2) is expanded, and the phase component frequency gradient is highly accurately determined. To be able to find the arrival direction of radio waves.

  As a method of increasing the signal band (frequency dispersion), a method of applying so-called chirp modulation (frequency sweep) used in a microscan receiver (Japanese Patent Laid-Open No. 2001-267971) or a chopper using a rectangular pulse signal Can be considered (Japanese Utility Model Laid-Open No. 6-49978). However, these methods use a periodic signal that is independent of the received signal and is asynchronous, and uses a time difference method using two receiving antennas. In order to be effective in the search device, it is necessary to shift the modulation timing by an amount corresponding to the arrival time difference between the two channels during modulation. For this reason, there is a problem that these methods cannot be used when measuring the arrival time difference. In addition, the method of finding the best modulation timing and increasing the signal bandwidth while shifting the modulation timing as appropriate is theoretically valid, but the signal processing time or circuit scale increases significantly, making implementation difficult. There was a problem of becoming.

  Further, as is clear from FIG. 16 (b2), the phase has the property of taking the same value every 2nπ (n is an integer) (that is, phase ambiguity is generated). Problem occurs. In the conventional azimuth detection device, in order to solve this problem, the number of reception antennas (the number of reception channels) is m sets m + 1 (m is a natural number). A new problem arises that increases.

  Therefore, in the present invention, by increasing the signal band by autonomous frequency dispersion using only the received signal, without using the arrival time difference itself whose value is unknown and with an appropriate circuit scale, The frequency gradient (group delay time) of the phase component of the cross spectrum is obtained with high accuracy. Further, in the present invention, the spectrum continuously spreads due to frequency dispersion, so that phase ambiguity is 2π at the maximum even if it exists between two consecutive points. Thus, the phase ambiguity is eliminated, and the average value of the frequency gradient is obtained by excluding the peak on the spike generated at the phase discontinuity point.

  The azimuth detecting device according to claim 1 is provided with first and second receiving antennas arranged at a predetermined distance apart to receive incoming radio waves, and first and second receiving antennas received by the first and second receiving antennas. A frequency dispersion unit for frequency-dispersing each frequency band of the received signal, a cross-spectrum computing unit for computing a cross spectrum of the first and second received signals dispersed from the frequency dispersion unit, and the cross A time difference detector for detecting a time difference between incoming radio waves received by the first and second receiving antennas from a frequency gradient with respect to a phase frequency in a frequency band having a high coherence level indicated by the spectrum, and based on the time difference and the predetermined distance And an azimuth calculator that calculates the arrival azimuth of the radio wave.

  The azimuth detecting apparatus according to claim 2 includes an A / D converter that converts the first and second received signals, which are analog signals, into digital signals, and processes after the frequency dispersion unit using the digital signals. Is performed.

  The azimuth detecting apparatus according to claim 3 is characterized in that the frequency dispersion unit is constituted by a clipping circuit or a limiting circuit.

  According to a fourth aspect of the present invention, in the azimuth detection device, the time difference detector eliminates phase ambiguity in which the phase varies by 2π between adjacent frequencies and becomes a phase discontinuity point in the frequency-phase characteristics of the cross spectrum. In order to achieve this, after phase connection is achieved by adding or subtracting 2π to the phase of the discontinuous point, the time difference between the incoming radio waves received by the first and second receiving antennas from the frequency gradient with respect to the phase frequency is calculated. It is characterized in that it is detected.

  In the azimuth detection apparatus according to claim 5, the time difference detector calculates a frequency gradient by frequency differentiation of the phase of the cross spectrum in the frequency-phase characteristic of the cross spectrum, and a frequency gradient exceeding a predetermined threshold value is calculated. A time difference between incoming radio waves received by the first and second receiving antennas is detected from an average frequency gradient calculated by averaging the remaining frequency gradients.

  In the azimuth detecting device according to the first to third aspects of the present invention, the obtained incoming radio wave is frequency-dispersed to widen the band, and then the cross spectrum band of the incoming radio wave is widened to accurately detect the phase component of the cross spectrum. By obtaining the frequency gradient, the calculation of the arrival time difference with high accuracy and the calculation of the arrival direction with high accuracy, which have not been realized with the conventional method, are realized.

  Further, in the azimuth detecting device according to the present invention according to claims 4 and 5, by utilizing differentiation or the like, the problem of phase ambiguity of phase 2nπ (where n is an integer) that has not been realized by the conventional method is solved. is doing.

Embodiment 1 FIG.
FIG. 1 shows a block diagram of an azimuth detection apparatus according to Embodiment 1 of the present invention. The present invention has two receiving systems ChA and ChB. 1 is a receiving antenna comprising two antennas for receiving radio waves, 2 is a high-frequency amplifier comprising two amplifiers corresponding to each antenna, 3 is a frequency dispersion unit for frequency-dispersing received radio waves, and 4 is data after frequency dispersion. An A / D conversion unit that performs A / D conversion to generate digital data, 5 is a time difference detector that calculates the arrival time difference from the obtained digital data, and 6 is a direction detector that calculates the arrival direction from the arrival time difference.

  Here, the time difference detector 5 includes an FFT unit that performs Fourier transform (frequency conversion) on the obtained digital data, a cross spectrum unit that obtains a cross spectrum from the data after Fourier transform, and a gradient that obtains a frequency gradient from the phase component of the cross spectrum. It comprises a calculation unit and a delay time calculation unit that calculates the arrival time difference from the calculated frequency gradient. The data handled by the receiving antenna 1, the high frequency amplifier 2, and the frequency dispersion unit 3 before the A / D conversion unit 4 is analog data. The A / D converter 4 converts analog data into digital data. Therefore, data handled after the time difference detector 5 is digital data.

  In the description of the above block diagram, the time difference detector 5 and the subsequent digital processing are performed. However, as shown in FIG. 2, the frequency dispersion unit 3 and the A / D conversion unit 4 are replaced, and the frequency dispersion unit 3 and the subsequent steps are replaced. The present invention can be implemented by a digital processing method. In FIG. 2, the configuration is the same as FIG. 1 except for the difference in the position of the A / D transformation portion 4.

FIG. 3 shows a processing flow of the direction finding apparatus according to the first embodiment. For each of the incoming radio waves A and B acquired by the receiving antenna and the high frequency amplifier, frequency dispersion is performed by the frequency dispersion unit (steps 1 and 2). Since the signal after frequency dispersion has various frequency components compared to before dispersion, the band is also wider than before dispersion. The signals after the dispersion of the incoming radio waves A and B are A _dis (t) and B _dis (t), respectively.

  Frequency dispersion is modulated using an independent / asynchronous modulation signal that is separate from the received signal, such as the chirp modulation method used in microscan modulators, etc., and the chopper modulation method using a rectangular pulse signal. The signal band is expanded by autonomous frequency dispersion using only the received signal. As a specific method of frequency dispersion, for example, in an analog circuit, clipping using a limiter (amplitude limiter) for an analog signal, hard limiting using a logic gate, and the like can be given.

The signals A _dis (t) and B _dis (t) after frequency dispersion are converted into digital data by the A / D converter (step 3). The converted data are _denoted as A ^D _dis (t) and B ^D _dis (t), respectively. The signals A ^D _dis (t) and B ^D _dis (t) after digital conversion are frequency-converted by FFT or the like (step 4). The signal after frequency conversion FA ^D _dis (f), and FB ^D _dis (f) (f is frequency).

Signal after frequency conversion FA ^D _dis (f), the FB ^D _dis (f), calculates the cross-spectrum ^S D (f) (Step 5). Since the calculated cross spectrum is calculated with respect to a signal whose bandwidth is expanded by frequency dispersion, the cross spectrum calculated without frequency dispersion (this cross spectrum is expressed as S ₀ ^D (f) and The bandwidth is wider than For this reason, the frequency gradient of the phase component can be determined with high accuracy without being affected by local phase error fluctuations, and a highly accurate arrival direction can be obtained. In addition, since the group delay does not change even if the frequency dispersion is performed, the cross spectrum when the frequency gradient of the cross spectrum S ₀ ^D (f) when the frequency dispersion is not performed is also used for the frequency gradient of the cross spectrum. The frequency gradient of S ^D (f) is the same.

The frequency gradient of the phase component of the cross spectrum is obtained by the following method (step 6). The first method eliminates the phase ambiguity by cross-phase phase connection using the fact that the phase ambiguity is 2π at the maximum because the spectrum continuously spreads due to frequency dispersion. In this method, the frequency gradient (group delay time) of the phase component of the spectrum is obtained. As shown in FIG. 4, specifically, the phase difference Δφ ^D (i) is calculated for the phase components φ ^D (i) and φ ^D (i + 1) at two adjacent points, and the change in the magnitude thereof is calculated. Thus, the phase ambiguity is solved by adding ± π correction to φ ^D (i + 1).

  The second method is a method of obtaining the frequency gradient (group delay time) of the phase component by differentiating the frequency of the phase component of the cross spectrum. Since the slope is obtained by differentiation, the phase frequency gradient can be obtained without being influenced by the phase ambiguity. However, when differentiation is simply performed, a spike-like peak occurs at a phase discontinuity point, or a range where the frequency gradient of the phase component other than the band cannot be obtained with high accuracy is included.

The peak at the phase discontinuity can be removed by the first method. As another method for removing the unnecessary range, in the present invention, a certain threshold value is set, and the data below the threshold value is excluded to obtain the average value of the frequency gradient excluding spike-like peaks. There is also a way to make it. The frequency gradient is calculated by performing the above two methods alone or in combination (after performing the first method, the second method is performed). The arrival time difference τ is calculated from the frequency gradient _Gφ of the obtained phase component by the time difference detector (step 7), and then the arrival direction θ is calculated by the direction detector (step 8). Hereinafter, specific examples of the frequency dispersion unit, which is a main part of the present invention, will be described with reference to examples.

Example 1.
The first embodiment is a method of implementing frequency dispersion by providing a limiter (amplitude limiter) as a frequency dispersion unit, adopting a clip circuit for the limiter, and clipping. FIG. 5 is a configuration block diagram of the azimuth detecting device according to the first embodiment. The basic configuration is the same as in FIG. This device includes a receiving antenna, a high frequency amplifier, a frequency dispersion unit (clip circuit), an A / D converter, and a computer for performing time difference detection (FFT, cross spectrum calculation, gradient calculation, delay time calculation) and direction detection. Become.

  The receiving antenna is a device that converts electromagnetic waves of radio waves emitted into the space into a high-frequency current, and may have any shape and configuration as long as it can fulfill this purpose. However, the required antenna shape and size differ depending on the frequency and polarization of radio waves. The high-frequency amplifier is a low-noise amplifier that selects and amplifies an input signal, and may have any shape and configuration as long as it can achieve this purpose. As shown in FIG. 6, the clip circuit is a circuit that cuts out a certain amplitude or more from the acquired high-frequency current, and is configured by a diode or the like, for example.

  An A / D converter is a device for converting a high-frequency current (analog signal) cut out by a clipping circuit into a digital signal so that it can be analyzed by a computer at a later stage. Any configuration is acceptable. The computer uses the converted digital signal to perform Fourier transform (FFT) calculation, cross spectrum calculation, frequency gradient calculation, arrival time difference calculation and arrival azimuth calculation, and finally the target radio wave source Output the direction of arrival of radio waves coming from.

Next, the operation will be described. The basic operation is the same as described with reference to FIG. The receiving antenna converts the electromagnetic wave of the incoming radio wave emitted into the space into a high frequency current, and the high frequency current is received by the receiver. In order to clarify that the function is related to time, the received radio waves A and B after reception are respectively A (t) and B (t) (t is time). A (t) and B (t) can be expressed as equations (2) and (3).
A (t) = AM (t) · sin {2πf _c t + φ _sys } (2)
B (t) = AM (t ) · sin {2πf c (t + τ) + φ sys} ··· (3)
However, AM (t) is an amplitude, f _c is a carrier frequency, and φ _sys is a phase unique to the system.

  For each of the incoming radio waves A (t) and B (t) after reception, clipping, which is a method of frequency dispersion, is performed using a clipping circuit. Signal waveforms before and after the clip circuit are as shown in FIG. By clipping by the clipping circuit, the incoming radio waves A (t) and B (t) after reception, which were originally sine waves, approach a rectangular wave, and the frequency is dispersed. An example of frequency dispersion corresponding to FIG. 6 is shown in FIG. As is clear from FIG. 7, a waveform having a single frequency before clipping becomes a waveform having a plurality of frequency components due to clipping (= a waveform in which frequency is dispersed).

Ａ_ｋ（ｔ）＝ＡＭ_ｋ（ｔ）・ｓｉｎ｛２π（ｆ_ｃ＋ｆ_ｋ）ｔ｝・・・（６）
Ｂ_ｋ（ｔ）＝ＡＭ_ｋ（ｔ）・ｓｉｎ｛２π（ｆ_ｃ＋ｆ_ｋ）ｔ
＋２π（ｆ_ｃ＋ｆ_ｋ）τ｝・・・（７）
ただし、ｋ＝−Ｌ、・・・、−１、０、１、・・・、Ｎ、ＡＭ_ｋ（ｔ）は各分散周波数成分の振幅、ｆ_ｋは分散周波数である。また、ｆ_０＝０である。 Since the signal after clipping has various frequencies compared to before the dispersion, the band is also wider than before the dispersion. Assuming that the signals after the dispersion of the incoming radio waves A and B are A _dis (t) and B _dis (t), they are expressed as equations (4) and (5).

A _k (t) = AM _k (t) · sin {2π (f _c + f _k ) t} (6)
B _k (t) = AM _k (t) · sin {2π (f _c + f _k ) t
+ 2π (f _c + f _k ) τ} (7)
Here, k = −L,..., −1, 0, 1,..., N, AM _k (t) is the amplitude of each dispersion frequency component, and f _k is the dispersion frequency. Further, f ₀ = 0.

The signals A _dis (t) and B _dis (t) after clipping are converted into digital data by an A / D converter. The converted data are _denoted as A ^D _dis (t) and B ^D _dis (t), respectively. In this description, the signal is converted into digital data after clipping. However, the signal may be converted into digital data before clipping, and the converted digital data may be clipped.

The signals A ^D _dis (t) and B ^D _dis (t) after digital conversion are frequency-converted by FFT or the like. Frequency conversion such as FFT is performed by a computer. The signal after frequency conversion FA ^D _dis (f), and FB ^D _dis (f) (f is frequency). Signal after frequency conversion FA ^D _dis (f), the FB ^D _dis (f), calculates the cross-spectrum ^S D (f) as in equation (8).
S ^D (f) = FA ^D _dis (f) · FB ^D _dis (f) ^* (8)
Where ^* is a conjugate symbol.

Since the cross spectrum calculated by the equation (8) is calculated for a signal whose band is expanded by frequency dispersion by clipping, the cross spectrum calculated without clipping (this cross spectrum is expressed as S _0. ^The band is wider than ^D (f). For this reason, as described above, the frequency gradient can be obtained with high accuracy without being influenced by local phase error fluctuations, and a highly accurate arrival direction can be obtained.

なお、周波数分散しても群遅延は変化しないため、クロススペクトルの位相成分の周波数勾配はＳ_０ ^Ｄ（ｆ）もＳ^Ｄ（ｆ）も同じである。また、クロススペクトルの計算は計算機により実施する。 A phase component φ ^D (f) is extracted from the obtained cross spectrum S ^D (f). S ^D (f) is a complex number, and its real part is Real (S ^D (f)), and its imaginary part is Image (S ^D
(F)), the amplitude component Amp ^D (f) and the phase component φ ^D (f) of S ^D (f)
) And (10).

Since the group delay does not change even if the frequency is dispersed, the frequency gradient of the phase component of the cross spectrum is the same for both S ₀ ^D (f) and S ^D (f). The calculation of the cross spectrum is performed by a computer.

As shown in FIG. 8, the frequency gradient G _φ of the phase component φ ^D (f) is obtained using the phase component φ ^D (f) of the cross spectrum obtained by the above equation (10). Further, for the obtained frequency gradient _Gφ of the phase component, the arrival time difference τ and the arrival direction θ are calculated by the delay time and direction calculation unit in the computer. The relationship of equation (11) is established between the frequency gradient G _φ of the phase component and the arrival time difference τ.
G _φ = 2πτ (11)
Therefore, if the arrival time difference is obtained from equation (11), the arrival direction can be obtained from equation (1).

Example 2
In the first embodiment, a clip circuit is used as the frequency dispersion unit, and frequency dispersion is realized by clipping. In the present embodiment, a hard limiter circuit unit is provided instead of the clip circuit as the frequency dispersion unit, and the frequency dispersion is realized by performing a hard limit.

  The signal waveforms before and after the hard limiter circuit are as shown in FIG. The hard limiter unit hard limits the amplitude of the received signal above a certain threshold (the threshold is determined in consideration of noise entering the device, etc.) to a constant value, for example, 1, and the incoming radio wave below a certain threshold. The component is converted to zero. The incoming radio wave component, which was originally a sine wave, approaches the rectangular wave as shown in FIG.

  An example of the hard limit of the incoming radio wave and the distribution of frequency components before and after the hard limit are shown in FIG. As is apparent from FIG. 10, what was a single frequency before the hard limit has a plurality of frequency components due to the hard limit (= frequency-distributed form).

  The following two methods can be considered as examples of the mounting method of the hard limiter circuit unit. As shown in FIG. 11, the first method is a method in which a hard limiter circuit unit is provided in the computer after the A / D converter and a hard limit is realized by calculation. As shown in FIG. 12, the second method is a method of realizing a hard limit by using a logic gate (1-bit A / D converter such as a CMOS gate) whose operation speed is a microwave band class as an analog device. When the method is used, the hard limiter unit and the A / D conversion unit can be combined into one circuit, and thus the size and weight can be reduced. In addition, compared with a general A / D converter, a 1-bit A / D converter can use a higher-speed one than a multi-bit converter, so that oversampling can be easily performed. Therefore, the subsequent processing can be performed with higher accuracy.

Example 3 FIG.
In the third embodiment, a phase connection method for eliminating the phase ambiguity of the phase components of the first and second embodiments will be described in more detail. Phase component of the two adjacent points phi ^D (i), for φ ^D (i + 1), calculates the phase difference △ φ ^D (i), by its size, expressions like ^{(12) φ D (i +} 1) Is corrected to solve the phase ambiguity and obtain the frequency gradient of the phase. An example is shown in FIG.
Δφ ^D (i) = φ ^D (i + 1) −φ ^D (i)
If Δφ ^D (i)> π Then φ ^D (i + 1) = φ ^D (i + 1) -2π
If Δφ ^D (i) <− π Then φ ^D (i + 1) = φ ^D (i + 1) + 2π
Otherwise φ ^D (i + 1) = φ ^D (i + 1) (12)

Example 4
In the fourth embodiment, the method for calculating the frequency gradient of the phase component of the first and second embodiments will be described in more detail. In addition, although this description describes the method of implementing Example 4 after Example 3, only Example 4 may be implemented independently. The flow of Example 4 is shown in FIG.

平均化は微小なノイズを除去するために実施する。 First, phase connection is performed by the method of the third embodiment, and as shown in the following formula (13), an average of adjacent _NP1 points is taken and averaged (smoothed) is performed. Phase component after averaging and phi ^D _{new new.}

Averaging is performed to remove minute noise.

微分により、図８のように周波数勾配は一定値になると考えられるため、容易に周波数勾配を求めることができる。しかしながら、単純に微分を実施した場合、位相の不連続点でスパイク状のピークが生じてしまったり、帯域以外の精度良く位相成分の周波数勾配を求めることのできない範囲を含むことになったりする。位相の不連続点でのピークについては、実施例３の方法により解決可能である。不要範囲の除去については以下の方法により解決する。 Next, as shown in the following formula (14), φ ^D _new (f) is differentiated with respect to the frequency to obtain a phase frequency gradient G _φ .

Since the frequency gradient is considered to be a constant value as shown in FIG. 8 by differentiation, the frequency gradient can be easily obtained. However, when differentiation is simply performed, a spike-like peak occurs at a phase discontinuity point, or a range where the frequency gradient of the phase component other than the band cannot be obtained with high accuracy is included. The peak at the phase discontinuity can be solved by the method of the third embodiment. The removal of the unnecessary range is solved by the following method.

First, in the band (crest) including the frequency at which the amplitude component of the cross spectrum is maximum, the amplitude width of the amplitude component, for example, the maximum value of −3 dB (3 dB down width) is obtained, and the frequency component falls within this frequency width. An average value _GφMAV of the frequency gradient of the phase component is _obtained . Next, data whose phase component frequency gradient value is G _φMAV × k _MAV or less and −G _φMAV × k _MAV or more is extracted from the phase component, and the average value of the data is obtained (k _MAV is an arbitrary value). The obtained average value is the gradient _{Gφ of the} phase component. Since the frequency gradient around the point having the maximum amplitude component is considered to be relatively accurate, the unnecessary range can be removed by this method.

1 is a configuration block diagram of an azimuth detection device according to Embodiment 1 of the present invention. It is a block diagram of the configuration of another azimuth detecting device according to Embodiment 1 of the present invention. It is an operation | movement flowchart of the direction finding apparatus in Embodiment 1 of this invention. It is explanatory drawing about the solution method of phase ambiguity. 1 is a configuration block diagram in Embodiment 1 of the present invention. It is explanatory drawing about operation | movement of the clip circuit in Example 1 of this invention. It is explanatory drawing about the change of the frequency by clipping in Example 1 of this invention. It is explanatory drawing about the calculation method of the frequency gradient of the phase component of a cross spectrum using the differentiation in Example 1 of this invention. It is explanatory drawing about the operation | movement of the hard limit in Example 2 of this invention. It is explanatory drawing about the change of the frequency by the hard limit in Example 2 of this invention. It is a block diagram of the configuration of an azimuth detection device including a hard limiter circuit unit in Embodiment 2 of the present invention. It is a block diagram of another azimuth detecting device including a hard limiter circuit unit in Embodiment 2 of the present invention. It is explanatory drawing about the method of calculating | requiring the frequency gradient of a phase component by the differentiation and the spike-shaped peak removal in Example 4 of this invention. It is a block diagram of a conventional azimuth detecting device. It is explanatory drawing about the principle of operation of the conventional direction finding device. It is explanatory drawing about the amplitude component and phase component of a cross spectrum.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Reception antenna, 2 High frequency amplifier, 3 Frequency dispersion | distribution part, 4 A / D converter, 5 Time difference detector, 6 Direction detector, 9 Direction calculation circuit, 12 Detection amplifier, 15 Pulse rise detection circuit, 17 Time counter

Claims (5)

The first and second receiving antennas that are arranged at a predetermined distance and receive incoming radio waves, and the frequency bands of the first and second received signals received by the first and second receiving antennas are respectively frequency-dispersed. A frequency disperser that performs a cross spectrum operation on the cross spectrum of the first and second frequency-dispersed received signals output from the frequency disperser, and a frequency band in which the level of coherence indicated by the cross spectrum is high. A time difference detector for detecting the time difference between the incoming radio waves received by the first and second receiving antennas from the frequency gradient with respect to the phase frequency in, and an azimuth calculation for calculating the arrival direction of the radio waves based on the time difference and the predetermined distance A direction finding device characterized by comprising a device. 2. An A / D converter for converting the first and second received signals, which are analog signals, into digital signals, and processing after the frequency dispersion unit is performed using the digital signals. The azimuth detecting device described. The azimuth detection apparatus according to claim 1, wherein the frequency dispersion unit includes a clipping circuit or a limiting circuit. In the frequency-phase characteristic of the cross spectrum, the time difference detector has 2π in the phase of the discontinuous point in order to eliminate the phase ambiguity that causes the phase to fluctuate by 2π between adjacent frequencies. A time difference between incoming radio waves received by the first and second receiving antennas is detected from a frequency gradient with respect to a phase frequency after phase connection is achieved by adding or subtracting. The orientation detection apparatus according to 1 or 2. The time difference detector calculates a frequency gradient from the frequency derivative of the phase of the cross spectrum in the frequency-phase characteristic of the cross spectrum, and calculates the average of the remaining frequency gradients excluding the frequency gradient exceeding a predetermined threshold. The azimuth detecting device according to claim 1 or 2, wherein a time difference between incoming radio waves received by the first and second receiving antennas is detected from an average frequency gradient.

Images