JPH0532710B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a receiving sound beam synthesis method for a device for observing the cross-sectional structure of an object using pulsed ultrasonic waves, particularly a high-speed ultrasonic tomographic imaging device for observing even the motion of an object. Regarding.

[Conventional technology]

In order to obtain ultrasonic tomographic images at high speed, instead of mechanically scanning a fixed directional probe, we give the signals of the arrayed ultrasonic elements a delay time according to a certain distribution, change the element selection, Alternatively, a configuration is adopted in which the directivity can be changed by changing the delay time distribution. An example of this is described in Japanese Patent Application Laid-Open No. 49-43480.

The configuration for obtaining desired directivity using delay time distribution will be explained in detail below with reference to the drawings. A transmitted sound wave a(t) shown in FIG. 1(i) is emitted to the target.

Here, a(t)=A(t)sinωt...(1) (A(t)=1(0≦t≦τ ₀ ) 0 (other)). Corresponding to the target object, the above ω(=2πfc:fc
is the frequency), τ ₀ is set. For example, in sonar, fc≒100kHz, τ ₀ ≒100μs,
In medical applications, the values are approximately 5MHz and 2μs, respectively, and vary widely. This transmitted sound wave is reflected by an object and enters the array receiving element group d _p to _{d o-1} as shown in FIG. Therefore, as shown in FIG. 1, element outputs a _p to a _o-1 having a time difference τ ₂ corresponding to the direction in which the object exists are obtained. Here, τ ₁ is the round-trip sound wave propagation time to the object.

FIG. 1 shows the configuration of a receive beam combining system that is currently widely used. Here DL _p is the received signal a _p
is an analog delay circuit that delays the
Performs signal delay of D _p . Here, D _p (n-p)·τ ₄ is obtained, and by changing this basic delay time τ ₄ , the reception beam direction is changed. This delay circuit
Since the received signal a _p is a _p =a(t-τ ₁ -Pτ ₂ ), the output b _p of DL _p is b _p =a(t-τ ₁ -Pτ ₂ -D _p ) =a{t-τ ₁ +P(τ ₄ −τ ₂ )−nτ ₄ }, and if the sound wave incidence direction and reception direction match (τ ₂ =τ ₄ ), b _p =a(t−τ ₁ −nτ ₄ ), and all outputs have the same waveform. After such delay time matching, an adder S obtains a received signal C in the direction of interest. This target azimuth signal C becomes C= _o-1 〓 ^p=0 b _p =na (t-τ ₁ −nτ ₄ ) and is obtained as a large output. The above is about the signal arriving from the target direction, but for the signal from the undesired direction (τ ₄ −τ ₂ =Δ), b _p =a{t−τ ₁ +PΔ−nτ ₄ }, so the adder output As shown in FIG. 1, the signals C cancel each other out, resulting in a suppressed output.

As can be understood from the above explanation of the basic operation, in the case of the conventional method, the delay time accuracy of the delay circuit DL is required to be the accuracy of the carrier wave period (τ ₃ ), and is τ ₃ /2.
If the accuracy is within a certain range, the target azimuth signal will also be degraded, and a delay time accuracy of approximately τ ₃ /4 is usually required, making the configuration extremely difficult.

The configuration shown in FIG. 2 is an improvement on this. Multiplying the received signal a _p and the reference signals e _sp and e _cp is performed by multipliers M _sp and M _cp constituted by balanced modulators. The internal configurations of the multipliers M _sp and M _cp are the same and are given different symbols for convenience of explanation. This reference signal is a delayed sine wave with a center frequency ω, and the distribution of delay times between channels varies depending on what kind of directivity reception is performed, that is, convergence reception, which will be described later. To simplify the explanation, let us explain the case of obtaining directivity corresponding to a plane wave from an oblique direction as shown in Fig. 1. These reference signals are e _sp = sin {ω (t-Pτ ₄ )} ……( 2) e _cp = cos {ω(t-Pτ ₄₎ } ...(3) They are delayed signals with a phase shift of 90° from each other. That is, there is a phase difference of basic delay time τ ₄ between adjacent element channels. A reference signal for each channel having such a delay distribution can be created without using an analog delay line. Its configuration is shown in Figure 3a. Third
A rectangular wave having a period τ ₃ (τ ₃ =2π/ω) shown in FIG. 3B is applied as data to the shift register SHR, and the contents of SHR are shifted by a clock having a period τ ₄ . With this configuration, a waveform f _p with a period of τ ₃ is obtained, which is delayed by Pτ ₄ from the Pth bit of the shift register. Furthermore, by delaying this f _p by a digital delay circuit DD _p using a monostable multivibrator having a delay time of τ ₃ /4, a signal g _p having a phase difference of 90° from f _p is obtained as shown in Figure 3b. is obtained. By shaping these f _p , g _p by resonance filters E _sp , F _cp with center frequency ω, e _sp , e _cp
is obtained. Digital delay circuits DD ₀ , DD ₁ , . . . , resonant filters are similarly connected to the outputs f ₀ , f ₁ , . . . of each bit of the shift register SHR.
F _s0 , F _s1 ... and F _c0 , F _c1 , ... are provided, and reference signals e _s0 , e _s1 , ... and e _c0 , e _c1 , ... having a phase difference of τ ₄ between adjacent channels are provided. ...is obtained.

The outputs h _sc and h _cp of the multipliers M _sp and M _cp are respectively h _sp = a _p・e _sp = A(τ) sin(ωτ)・sin {ω(t−Pτ ₄ )} = A(τ)/ 2 [cos {ω(P(τ ₄ − _{τ 2} ) − τ ₁ ) = −cos {ω(2t − τ ₁ −P(τ ₂ + τ ₄ )}) = A(τ)/2 [cosφ _p −cos (2ωt−φ _p ′)}……
(4) and h _CP = a _P・e _P = A(τ) sin(ωτ)・cos {ω(t−Pτ
₄ )} = A (τ) / 2 {sinφ _P + sin (2ωτ−φ _P ′)}
...(5) Here, φ _P = ω{P (τ ₄ − τ ₂ ) − τ ₁ } φ _P ′=ω{P (τ ₄ + τ ₂ ) + τ ₁ } τ=t−τ ₁ −Pτ ₂ = τ(t). In h _CP and h _SP , A _(t) is the envelope component of the received waveform and has a frequency component sufficiently lower than sin (2ωτ) and cos (2ωτ). For this reason, low-pass filters L _SP and L _CP reduce the 2ω frequency component.
It is possible to separate and extract only the frequency component of A ₍ 〓 ₎ . Such filter outputs i _SP(t) and i _CP(t) are h _{SP and} i CP(t), respectively.
Only the first term on the right side of h _CP becomes: i _SP(t) = A(τ)/2cosφ _P ……(6) i _CP(t) = A(τ)/2sinφ _P ……(7). A waveform like this is converted into an analog delay circuit.
Delayed by _DSP and _DCP . These D _P and D _P have the same configuration, and the variable delay time (n-p) τ ₅
is a delay line that gives Here, τ ₅ is a value related to the delay time setting values of the analog signal delay units D _P and D _P , and each channel delay time setting value is used for receiving a plane wave from an infinite distance, which was used for simplicity in the example. In this case, the delay time setting value for each element is given as (n-p)τ ₅ for element number p. Here, τ ₅ is a set value, and an error of Δp occurs in actual operation. For this reason
The outputs j _SP and j _CP of DS _P and DC _P are j _SP = A(τ _P ′)/2cosφ _P ……(8) j _CP = A(τ _P ′)/2sinφ _P ……(9) τ _P ′ =τ{t-(n-p) _τ5 } =t- _τ1 +P( _τ5 - _τ2 ) _-nτ5 . Such signals j _SP and j _CP are added by adders S _S and S _C that obtain the sum of n signals, respectively. The adder outputs k _S and k _C are respectively ks= _o-1 〓 ^p=0 j _SP =1/2 _o-1 〓 ^p=0 A(τ _P ′) cosφ _P ……(10) kc= _{o- 1} 〓 ^p=0 j _CP = 1/2 _o-1 〓 ^p=0 A(τ _P ′) sinφ _P ...(11). If we consider the case where a sound wave is incident from the target direction, τ ₂ = τ ₄ = τ ₅ , so ks = nA (t-τ ₁ - nτ ₄ )/2 cos (-ωτ ₁ ) ...(12) kc = The signal becomes nA(t-τ ₁ −nτ ₄ )/2sin(−ωτ ₁ ) (13), resulting in a large output. This signal is squared by squarers T _S and T _C , added by S _B ,
An output signal C is obtained by square rooting using a square rooter R. From this configuration, the output C for the target direction signal is C=√ ² _S + ² _C = nA (t-τ ₁ − nτ ₄ )/2 ...(14), regardless of the distance to the target object (τ ₁ maximum output will be obtained (regardless of On the other hand, for sound waves from directions other than the target, τ ₄ −τ ₂ = Δ,
If τ ₅ −τ ₂ =Δ′, then k _S =1/2 _o-1 〓 〓 ^p=0 A{t−τ ₁ +PΔ′−nτ ₅ }・cos{ω(pΔ−τ ₁ )
} k _C = 1/2 _o-1 〓 〓 ^p=0 A{t−τ ₁ +PΔ′−nτ ₅ }・sin{ω(pΔ−τ ₁ )
}, and when (n-1)ωΔ becomes 2π or more, k _S , k _C
becomes a small value, resulting in a suppressed output. This will be explained regarding _kS . Note that the same applies to _kC .

k _S = 1/2 _o-1 〓 〓 ^p=0 A(t-τ ₁ +PΔ'-nτ ₅ )・cos{ω(p・Δ-τ
₁ )}, and here, if (n-1)ωΔ>2π, then p
Corresponding to the change in from 0 to n-1, ωp・Δ
changes from 0 to 2π. For this reason, cos{ω(p・
The value of Δ−τ ₁ )} increases by 1 in response to this change in ωp・Δ.
k _S which is the result of adding up all of these changes for a period of time is a small value because the positive and negative values are averaged. Here, since A(t-τ ₁ +PΔ′-nτ ₅ ) is normally Δ=Δ′, pΔ′pΔ<nΔ2π/ωτ ₃ ≪τ ₀ , and t=τ ₁ where the reflected signal is obtained. There is almost no change near the time. This (n-1)ωΔ=
The direction corresponding to 2π becomes the first zero point of the directional characteristic, and the same directional characteristic as the conventional system is realized.

That is, as described above, A(t-τ ₁ +pΔ′-nτ ₅ ) does not change near the center of the reflected wave with respect to changes in Δ and Δ′. Therefore, this value is
Then, k _S 1/2 _o-1 〓 ^p=0 Bcos {ω(pΔ−τ ₁ )}. Such a change in k _S with respect to Δ takes the well-known form of sinx/x, which becomes 0 at Δ=2π/nω. The direction corresponding to this Δ becomes the first zero point. On the other hand, in the conventional method, the 3rd line of the 4th page
b Using the formula of _P , _o-1 〓 ^p=0 b _P = _o-1 〓 ^p=0 a{t-τ ₁ +pΔ-nτ ₄ }= _o-1 〓 ^p=0 A{t-τ ₁ +pΔ −nτ ₄ }sin {ω(t−τ ₁ +pΔ−nτ ₄ )} _o−1 〓 ^p=0 Bsin{ω(t−τ ₁ −nτ ₄ )+ωpΖΔ}, and the change to Δ is due to the present invention. Become the same. Therefore, the azimuth resolution is also exactly the same.

Next, we will discuss the influence of the delay time accuracy of the delay circuit. The reference signal can be processed digitally, the required delay time can be easily obtained, and τ ₄ ~ _{τ 2} = 0
It is possible to set On the other hand, the portions that delay the received signal components (DS _P , D _P shown in FIG. 2) have amplitude information, making the configuration complicated and making it difficult to improve the time accuracy. Therefore, this _DSP ,
The delay time error of DC _P becomes a problem. Assuming that this delay time setting error is Δp, the actual delay time _DEP of D _P and D _P with respect to the set delay time pτ ₅ is _DEP = (np) τ ₅ −Δp. Therefore, the outputs j _SP and j _CP from the delay means corresponding to equations (8) and (9) are as follows: j _SP = A(τ _P ′)/2 cosφ _P j _CP = A(τ _P ′)/2 sinφ _P Become. Here, we are considering the received signal strength for sound waves incident from the target direction, so
The set delay amount τ ₅ is τ ₅ −τ ₂ (=Δ′)=0. Therefore, τ _p ′=t−τ ₁ +Δp−nτ ₅ , and the summed outputs k _s and k _c obtained by adding these together are obtained from equations (10) and (11), k _s = 1/2 _o-1 〓 ^{p =0} A(t-τ ₁ +Δp-nτ ₅ ) cosφ _p K _c =1/2 _o-1 〓 ^p=0 A(t-τ ₁ +Δp-nτ ₅ ) sinφ _p . Here, the phase difference φ _p is φ _p =ω{p (τ ₄ − τ ₂ ) − _{τ 1} }, and since it is an object in the target direction, similarly τ ₄ − _{τ 2} (=Δ)=0 , and φ _p =ωτ ₁ . From the above formula, k _s = 1/2 cos (-ωτ ₁ ) _o-1 〓 ^p=0 A{t-τ ₁ +Δp-nτ ₅ } k _s = 1/2 sin (-ωτ ₁ ) _o-1 〓 ^{p= 0} A{t−τ ₁ +Δp−nτ ₅ }. Reception output C(t) for the target direction when such a delay time error Δp exists
is C(t)=√ _s2 + _c2 = 1/2 _o-1 〓 ^p=0 A{t-τ ₁ +Δp-nτ ₅ }. Here, A{t-τ ₁ +Δp-nτ ₅ } has a reception start time of τ ₁ −Δp+nτ ₅ , which is the time length τ ₀
This is a rectangular pulse. For this reason, the error Δp changes depending on the receiving element p, so the reception time changes, and the sum of these, C(t), becomes longer than τ ₀ , as shown in FIG. 4. That is, assuming that n is large and Δp is uniformly distributed, _o-1 〓 ^p=0 A{t-τ ₁ +Δ _p −nτ ₅ } corresponds to the maximum value Δ _pn of Δ _p as shown in Figure 4. Change. As can be understood from this figure, by configuring the delay circuit DC _p so that Δ _pn ≦τ ₀ /2, it is possible to extract the target signal without reducing the maximum value.

On the other hand, in the conventional case, it is a direct delay addition of A(t)sin(ωt), and due to the error of Δp, the addition output becomes _o-1 〓 ^p=0 A(t)sin(ωΔp). Therefore, in order for the addition output not to decrease, ωΔp<<π is required, and Δp<<π/3=1/2τ ₃ . Here, since this τ ₀ is usually much longer than the above-mentioned τ ₃ , the delay time accuracy of DS _p and DC _p is much easier than that of the delay circuit DL _p in the conventional system.

In this way, it is possible to obtain a signal from a specific location, and by using this signal to modulate the brightness of the cathode ray tube, it is possible to learn the shape of an object, or to learn the nature of an object at a specific location by analyzing the reflected waveform. becomes.

The above is the operation when configuring the device faithfully, but various simplifications can be made by utilizing the feature of this method that focuses on envelope information.

Attention is paid to the output k _s of the adder S _s on one side in FIG. As shown in equation (12), the received output from the object in the target direction is k _s = nA (t-τ ₁ - nτ ₄ )/2cos (-ωτ ₁ ), which is expressed as cos in equation (14). (−ωτ ₁ ). This corresponds to the output amplitude changing (sensitivity changing) as τ ₁ changes (distance changing). This situation is shown in FIG.
In this way, fluctuations in sensitivity occur with the sound wave propagation time τ ₆ periods having the relationship ωτ ₆ =π. But this τ ₆
The distance interval Δr corresponds to the speed of sound in the propagation medium C _s ,
When the sound wave wavelength is input, Δr=C _s τ ₆ /2=C _s π/2ω=C _s /4f=λ/4, and when using a 2MHz sound wave underwater, Δr=0.19 (mm) (C _s : 1500 m /s), and in the case of an object with a finite size consisting of a large number of reflection points, such minute sensitivity changes pose no problem at all.

In other words, there is only one reflection point, and the fifth
If the position is fixed at the zero point of |k _s | shown in the figure (τ ₁ is fixed), the reflected signal will be lost.
However, usually the object moves, such as a living body or underwater side, or the observation point moves, and the relative position changes. Therefore, the sound wave propagation time τ ₁ changes,
Even if the reflected signal momentarily disappears, it immediately reappears. In particular, in the case of an object with a finite size, there are many reflection points, so even if the object is fixed, some reflection point will always be observed. Even if you only do this, you will never lose sight of the reflector.

As a result of the above, it is also possible to use a receive beam configuration method that uses only either e _sp or e _cp as a reference signal as a modification of the method shown in Figure 2. In this case, the configuration shown in Figure 2 will be approximately halved. The device becomes significantly simpler.

Also, according to this method, the delay time accuracy of DS _p is 4th
As shown in the figure, this is the accuracy of the envelope length. Therefore, when τ ₄ is small compared to τ ₀ , it is possible to group i _sp (or i _cp into multiple groups and then delay them) as shown in Figure 6. In this way, S _a , S _b ,
Signals la, l _b , l _c added by adder S _c
are as shown in FIG. 6, and no decrease in amplitude appears at all. These l _a , l _b , l _c are respectively DS _a ,
By delaying 12τ ₂ , 8τ ₂ , and 4τ ₂ by DS _b and DS _c , q _a , q _b , and q _c in Fig. 7 are obtained, and by adding these three signals using an adder SS, the target signal output u is maximized. It becomes possible to obtain the amplitude without decreasing the amplitude. The above is based on 4 signals (i _s0 ~ i _s3
Although the system was explained by configuring one unit (L _a, etc.) and three groups, the system is not limited to this configuration, and arbitrary division is possible under the restriction that the maximum amplitude of U does not decrease. is obvious. With this configuration, as is clear from FIG. 6, the number of delay circuits is greatly reduced (to 1/4 in the configuration of FIG. 6), making the device configuration easier.

[Problem to be solved by the invention]

In the above-described received acoustic beam combining method, the requirement regarding the accuracy of the delay circuit is relaxed by delaying the low frequency component obtained by multiplying the received signal and the reference signal. However, Fig. 2,
In the example shown in Figure 6, an analog delay line is used as the delay circuit, so in order to satisfy the required accuracy and obtain a large delay, it is inevitable that the delay circuit will be large and expensive. I don't get it.

Furthermore, although it was omitted in the explanation of Figures 2 and 6, as time elapses from the transmitting time, initially the reflected sound waves are received from the reflection point from a short distance, and then the reflected sound waves are gradually received from the distant points. By gradually reducing the curvature of the delay time distribution for phasing the received signal as it reaches the wave element, the reflected signal regardless of the distance can be received with the same phase at all times. is possible. In order to use analog delay numbers as shown in Figure 1 or Figure 6 when adopting such a reception method, each delay line must have a large number of intermediate taps, and furthermore, the intermediate taps must be It becomes necessary to switch the taps at high speed during the reception period after one wave transmission.

SUMMARY OF THE INVENTION Therefore, it is an object of the present invention to provide a wave receiving apparatus that can accurately and easily sequentially switch the delay time distribution even when receiving waves whose phases are always matched after the time of wave transmission.

[Summary of the invention]

Rather than delaying the waveform of the received signal itself, the present invention discretely samples and holds the received signals of multiple channels, and adds the held received signals of the multiple channels to create a desired delay time distribution. It is characterized by the configuration in which control is achieved so that the curvature corresponding to the delay time becomes smaller as time passes from the time of wave transmission.

[Effect]

[Embodiment] FIG. 9 shows the configuration of an embodiment of the present invention. In this embodiment, the analog delay numbers DS _a , DS _b ,
Adder instead of the part consisting of DS _c and adder SS
These are connected to the outputs of S _a , S _b , and S _c . However, instead of aligning the outputs la, _{l b} , and l _c of S _a , S _b , and S _c , a similar configuration can be used when aligning the outputs i _sp and i _sc in Figure 2. It is.

FIG. 9 will be explained with reference to FIG.

As shown in Fig. 8, the signals _la , l _b , l _c (i _sp , i _cp
(Although a configuration in which alignment is performed with respect to Identical parts of these signals (at different times) l _a1 , l _b1 , l _c1
are selected by multiplexers MX _a , _b , _c and applied to the same adder MA. Similarly, l _a2 , l _bz , l _c2
MA ₂ and sequentially apply specific portions of the signal through each adder up to MA ₇ . Next l _a8 , l _b8 ,
l Apply _c8 to MA ₁ again and repeat this operation. When such processing is performed, the outputs of the adders MA ₁ to MA ₇ become v ₁ to v ₇ , respectively, and the same portions of the signals are each sampled and extracted separately. These adder outputs v ₁ to _{v 7} are integrated by integrators IT 1 to IT ₇ for a finite time τ ₈ (time during which _la , l _b , and _lc are obtained).
This integrator can be reset immediately after the l _c component is output to l.

The outputs of such integrators are ω ₁ to ω ₇ , respectively _{, and the final values of the integrals are approximately the same as the amplitudes added after adjusting the times of la, l b and} l _c , respectively. In other words,
l _a , l _b , l _c sequentially sampled by multiplexers MX a , MX b , MX _c and appearing _{at the} output of adder MA ₁
Although the appearance times of the signals of each channel are different from each other, the values are sequentially held in the integrator and added together, so a delay distribution is realized. If such integration results are sequentially read out by the output multiplexer MPX just before resetting, x in Fig. 8 is obtained.
As shown in FIG. 7, a waveform substantially the same as u in FIG. 7 is obtained, which means that an envelope delay circuit has been equivalently realized. In this way, the multiplexer switch,
When configured with an adder and an integrator, all time adjustment can be performed using a digital clock, greatly improving the stability of the device. The signal selection mentioned here,
The control signal that controls the integral reset is shown in Figure 9.
Created by CX and CY. If the control signal that controls the timing of sequential selection of multiplexers MX _a , _b , and _c created by CX has a delay distribution with a certain curvature, the phase of the spherical waves from the reflection point at the desired position can be adjusted. If this curvature is made smaller over time, it becomes possible to receive reflected sound waves that are always in phase from nearby objects to distant objects. On the other hand, as in this embodiment, when phase processing is performed on each received signal by multiplying it with a reference signal, it is necessary to give the above-mentioned delay time of curvature that changes with time to this reference signal as well. Therefore, the circuit shown in FIG. 10 is used instead of the circuit shown in FIG. delay circuit
DV ₁ , DV ₂ , ... DV _p ... are delay circuits that can control the delay time, and the configuration of DV _p uses a normal monostable multivibrator whose output pulse width can be changed depending on the voltage. This can be easily achieved by Furthermore, considering that the reference signals e _sp and e _cp are sine waves, it is clear that a reference signal having such a quadratic curvature can be substituted by a simple phase shifter configured for only a single frequency. .

〔Effect of the invention〕

As described above, according to the present invention, since the delay is realized by holding the discrete values of the received signals of multiple channels, it is possible to easily realize an arbitrary pattern of delay distribution by time alignment using the digital clock, and as time goes on. Due to the delay distribution of changing curvature, phase-aligned reception can be easily performed from short distances to long distances.

[Brief explanation of the drawing]

1 is an explanatory diagram showing the time relationship between the transmitted waveform and the receiver output, , , is an explanatory diagram of the operation of the conventional method, FIG. 2 is a configuration diagram of the method that is the premise of the present invention, and FIG. Figure 2 shows an example of the configuration of the reference signal generator used in the method shown in Figure 4. Figure 4 shows the influence of errors in the delay circuit on the output waveform. Figure 5 shows the distance when configured with one type of reference signal. FIG. 6 is an explanatory diagram of a configuration that simplifies the delay circuit section, FIG. 7 is a diagram showing the output with the configuration of FIG. 6, and FIG. 8 is a diagram showing the delay circuit section according to FIG. 9. FIG. 9 is a circuit diagram of an embodiment of the present invention, and FIG. 10 is a diagram showing an additional circuit for maintaining focus at short distances.

Claims (1)

[Claims] 1. A received sound wave beam that is delayed and added to received signals of multiple channels obtained from a plurality of arrayed receiving elements to match reflected waves from a predetermined direction or position. A wave receiving device for synthesizing, which generates a plurality of reference signals of a predetermined frequency, the time relationships of which are controlled in accordance with the incident time of the wavefront of the reflected wave received at each position of the wave receiving element. a reference signal generating means having two input terminals and inputting corresponding ones of the reception signal of the reception element and the reference signal, respectively, and changing the phase of the reception signal of the plurality of channels and outputting the phase. a processing circuit, a means for sampling and extracting the values of mutually identical phase portions of the output of the phase processing circuit, a means for storing the extracted values and adding together a plurality of received signals, and the above-mentioned extraction and addition. a discrete delay means for providing a delay corresponding to the difference in the incident time of the wavefront, and means for adding the respective outputs of the discrete delay means. A wave receiving device characterized in that the time relationship is sequentially changed corresponding to the time from the sound wave transmission time to perform phase correction corresponding to the change in the wave front. 2. The wave receiving device according to claim 1, wherein the change in the wavefront and the curvature are inversely proportional to time.