JP2775312B2 - Receiving digital beamformer for phased array type ultrasonic equipment. - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial application field) The present invention relates to reception of a phased array type ultrasonic apparatus that converts a received signal into a digital signal as it is in an RF range and digitally performs phasing addition. Related to digital beamformers.

(Prior Art) When an ultrasonic pulse is emitted into space, a reflected wave returns from a point having a different acoustic impedance. By measuring the time from the transmission to the reception of the reflected wave, the distance to the reflection point can be known. Know the direction of the reflection point by arranging a number of elements in a plane and giving different time delays to the ultrasonic waves radiated from each element and giving them directional beams as directional beams. Can be. The phased array type ultrasonic apparatus is an apparatus that reconstructs and displays an image based on the above-described distance information and azimuth information based on a received signal.

In this phased array type ultrasonic apparatus, for example, in an ultra-short distance phased array type apparatus such as a medical ultrasonic apparatus, an echo signal received by each element or an element group is divided into blocks and each block is reserved. 2. Description of the Related Art There is an ultrasonic apparatus having a high-frequency digital receiving beamformer that AD-converts an echo signal that has been phase-synthesized and performs digitally all procedures for delay addition performed thereafter. If it is requested to minimize unnecessary echo components existing in the skirt portion of the directivity, which is so-called directional characteristics away from the target area, the directivity synthesized by this receiving beam forming method, the phase resolution of the RF echo signal is reduced.
About 22.5 to 36 mm is required. In order to obtain this phase resolution, the time delay of the delay line is required to have a resolution of 10 ns or less.

(Problems to be Solved by the Invention) When this resolution is required as the time axis resolution of a shift register used as a delay means or an SRAM system for a temporary memory of a waveform, the feasibility is almost lost. Therefore, conventionally, the vernier control as the ultrafine control means is weighted not for the time delay but for the in-phase component i and the quadrature component q obtained by separating the received signal by 90 °. By doing that. This process is a phase rotation process similar to the butterfly process in FFT (Fast Fourier Transform).
It is performed by one complex number multiplication. In this system, even if phase rotation within one wavelength can be performed, it is necessary to separately perform a time delay in a region exceeding one wavelength.

The present invention has been made in view of the above points, its purpose is to realize a phased array type ultrasonic apparatus having a vernier control of delay addition means as a more effective method, particularly as the means, An object of the present invention is to realize a receiving digital beamformer of a phased array type ultrasonic apparatus using a convolver.

(Means for Solving the Problems) The present invention for solving the above problems is a phased array type ultrasonic apparatus which converts a received signal into a digital signal as it is in an RF region and digitally performs phasing addition. In a receiving digital beamformer, an AD converter that converts a receiving signal of each receiving element or element groom into a digital signal, and a shift register that accommodates a plurality of parallel signals of a plurality of bits input in parallel as a serial signal. A coefficient memory for inputting and storing a kernel which is a coefficient sequence, and a number equal to the coefficient sequence for multiplying the data stored in the shift register and shifted for each clock input with the data stored in the coefficient memory. A convolver comprising: a multiplier that sequentially adds the output of the multiplier that changes according to the clock and outputs the result. A shift register for digital delay charge of main delay to the output data of Nboruba, is characterized in that it comprises an adder for adding the data from all elements that have undergone delay processing.

(Operation) A signal input for each receiver element or element group is converted into a digital signal by an AD converter, and stored in a shift register of a convolver in a bit serial manner, and each coefficient of a kernel stored in a coefficient memory. Are multiplied bit by bit by a multiplier for each clock, and the outputs of the multipliers are sequentially added by an adder according to the clock to perform vernier control of phasing addition.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram of one embodiment of the present invention. In the figure, reference numeral 1 denotes a receiver element for receiving a transmitted ultrasonic signal and converting it into an electric signal, and shows only one channel. Digital 2 is a first-stage amplifier for amplifying the electrical signals from the receivers element, its output signal by four times the frequency or clock frequency higher than that of the center frequency f ₀ of the ultrasonic signal in the AD converter 3 Converted to a signal. Reference numeral 4 denotes a serial register which reads 12-bit parallel signal inputs from the AD converter 3 and stores a shift register 4a in which a signal of each bit moves to the right for each clock input, and a coefficient of a kernel (coefficient sequence). Bit coefficient memory 4b
And eight multipliers 4c that take out data stored in the shift register 4a and the coefficient memory 4b in parallel and perform multiplication for each bit, and receive and add the outputs of the multipliers 4c, This is a convolver configured with an adder 4d that outputs a 16-bit signal. The kernel of the convolver 4 is externally set in the coefficient memory 4c, and is modified as needed to change the direction of the received beam. Reference numeral 5 denotes a digital delay shift register for providing a main delay of the signal whose phase rotation or time shift has been finely adjusted by the convolver 4, and 6 denotes an adder for adding the above-mentioned processed signals from the respective receiver elements 1. , The digital delay shift register 5 and the adder 6 constitute a main beamformer.

Next, the principle and operation of the embodiment configured as described above will be described. The ultrasonic echo signal received by the receiver element 1 is converted into an electric signal and amplified by the first-stage amplifier 2. This output signal is converted into 4f ₀ or more digital signal by the clock in the AD converter 3.
In the convolver 4, the following calculation is performed. The output signal S _{in of the} AD converter 3 stored in the shift register 4a
Is multiplied for each bit by the multiplier 4c with the data of the coefficient memory 4b in which the set of coefficient sequences is stored. The input signals stored in the shift register 4a are arranged in series, and are shifted to the right in the figure by a clock having the same frequency as the clock input to the AD converter 3. Therefore, the multiplication by the multiplier 4c is performed between the data of the input signal, which is shifted to the right by one bit in response to the clock input and arranged in series, and each bit of the coefficient sequence. The outputs of the multipliers 4c are added in the adder 4d and output as a _Cout signal. Eventually, the convolver 4 performs convolution integration of the input signal and the kernel, finds a correlation between the signals, and performs vernier control of phase rotation and time shift. In the signal processing by the convolver 4, the processing content from the signal input _Sin to the correlation output _Cout can be freely changed depending on what kind of kernel is stored in the coefficient memory 4b. Therefore, the kernel may be fixed or fixed during use, but can be rewritten or modified in real time using a μPU or the like, and is useful.

When a band-pass type kernel is used for the kernel stored in the coefficient memory 4b, the frequency of the input signal can be selected. This band-pass type kernel is characterized by having no DC component, and the relative distribution of coefficients appearing at each tap of the function form of the kernel, that is, the coefficient memory 4b, is as follows.
A desired frequency characteristic is obtained by performing an inverse Fourier transform.

On the other hand, such a relative distribution is determined from a desired frequency characteristic, but its absolute position, that is, a place where the function is packed in the frame of the kernel coefficient memory 4b can be freely determined. This is another degree of freedom with respect to the degree of freedom relating to the above-described processing contents. In other words, the delay time or phase relationship between input and output can be controlled by how to decide where to store the functions. This sampling manner on the time axis and amplitude axis towards the kernel, for example, put the idea to the amplitude resolution of 8 bits can be taken, at most very fine vernier control even at a sampling rate of 4f ₀ Can be.

The digital delay shift register 5 gives a time delay to the output signal C _out of the convolver 4. The adder 6 adds all the processed signals of the received signals from the respective receiver elements, and outputs a phasing-added signal. Subsequent signal processing is performed in the same manner as a normal ultrasonic device.

Here, considering the function of the convolver 4, the above-described vernier control function of the phase rotation or the time delay of the convolver 4 is performed in the phased array system by the following main beamformer (by the digital delay shift register 5 and the adder 6), that is, The time axis resolution for each channel in the main delay addition procedure is facilitated.
In other words, the time axis resolution can be quantized by the amount of time required to shift the kernel, which can be shared by the vernier control side, left and right. For example, it is typically rougher from the Nyquist rate to somewhat coarser, and can be roughened from about one wavelength to several wavelengths.

The control provided by the kernel also has a function of determining the gain or scale factor between input and output according to the manner in which the amplitude is sampled, that is, the scale factor.
This allows weighting or apodization within the aperture of the receiver. In other words, for the element whose weight is to be reduced, the amplitude of the kernel may be reduced.

As for the usage of the convolver 4, in the above case, the kernel has been regarded as a simple sample point sequence filter (bandpass filter). However, the present invention is not limited to this. If it is inserted, it becomes a matched filter, and it is also possible to use a filter (equalizer) having an inverse characteristic such that the transmitted pulse or the echo signal is deconvoluted to narrow the pulse width. In addition, variations in acoustic characteristics or variations in circuit characteristics for each element (each channel) of the receiver,
It is also possible to share the work including the work of equalizing by the function of the wave receiving equalizer.

Of particular note is the correlation or compression process (decoding process) and the correlation of the received signal with the "dispersion compression method" or the scrambled (by phase modulated wave) transmission and reception using the modulated transmission signal. This is a result obtained when a decoding process (descrambling process) for performing a process to obtain a correlation is performed by a convolver for each element. In the case of the FM chirp method weighted to the amplitude,
The transmitted signal itself may be used as the kernel, and the transmitted signal itself may be similarly used as the kernel when a long m-sequence is used. The contents of the modulation work or codes, modulation waveforms, etc. used in the process of correlating the transmitted or received signal of the dispersion-compressed or phase-modulated wave into a time signal are common to all elements in the aperture. Alternatively, it may be independent for each element group or for each element. In any case, the function of the vernier control relating to the amplitude and the time axis (relative delay, phase) can be realized in the same manner as described above.

The convolver 4 has another property that it can be made to perform a thinning-out or interpolation operation on the frequency axis or the time axis with respect to the data sample sequence. This property is another representation of the possibility of reordering, renaming or duplicating data or processing any possible linear combination between samples according to the requirements of the subsequent main beamformer. Proper lossless thinning is very effective in simplifying the hardware of the subsequent main beamformer and its work procedures.

The present invention is not limited to the above embodiment.
Second embodiment is a block diagram of another embodiment of the present invention. In the figure, the same parts as those in FIG. 1 are denoted by the same reference numerals. In the figure, reference numeral 11 denotes a group of elements of the phased array type receiver, and in the figure, eight elements constitute one group. Reference numeral 12 denotes an amplifier group that is connected to the group n of the group of elements of the receiver element group 11 and amplifies the received echo signal, and 13 distributes the output signal of the amplifier group 12 to each tap of the delay line 14. Is a cross point switch. Amplifier group
12, a cross-point switch 13 and a delay line 14 constitute a preliminary phasing stage, and phasing-add the received signals from the element group 11 divided into small groups.

In this embodiment, 4 to 8 elements of the receiver element group 11 are grouped together to form a group, and for each group, the pre-phasing is performed by the cross point switch 13 and the delay line 14 with the LC type tap as before. The above operation is performed using the resulting signal as one unit. 64
1 requires a huge convolver 4 and a digital delay shift register 5 to compose the circuit of FIG. If this is divided into 16 to 32 groups, the number becomes feasible. The effect shown in the embodiment in FIG. As an example of grouping, when the total number of elements is m (m = 2 ^M ), Choose It is appropriate to create groups. However, the method of this grouping is not limited to the above, and any method may be selected. In the above-described correlation processing by the convolver, the content of modulation, code, modulation waveform, or any other specific transmission signal used in the process of dispersion compression, scrambling, and descrambling is essentially the purpose of the present invention. It does not matter, and any type can be used.

(Effects of the Invention) As described in detail above, according to the present invention, by using a convolver, in particular, a convolver made into an LSI,
Vernier control for facilitating the digital beam forming procedure can be realized, and the practical effect is great.

[Brief description of the drawings]

 FIG. 1 is a block diagram of one embodiment of the present invention, and FIG. 2 is a block diagram of another embodiment of the present invention. 1 ... receiver element, 3 ... AD converter 4 ... convolver, 4a ... shift register 4b ... coefficient memory, 4c ... multiplier 4d, 6 ... adder 5 ... shift register for digital delay 11 Receiver element group 13 Cross point switch 14 Delay line

Continuation of the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G01S 7 / 52,15 / 89 A61B 8/00 G01N 29/26 503

Claims (2)

(57) [Claims] 1. A receiving digital beamformer of a phased array type ultrasonic apparatus for converting a received signal into a digital signal as it is in an RF region and digitally performing phasing and addition, wherein each receiver element or An AD converter (3) for converting the received signal of each group into a digital signal, a shift register (4a) for accommodating a parallel signal of multiple bits input in parallel as a serial signal, and a kernel as a coefficient sequence are input. The coefficient memory (4b) stored and stored in the shift register (4a) is the same as the coefficient column for multiplying the data stored in the coefficient memory (4b) and shifted at each clock input by the data stored in the coefficient memory (4b). Number multiplier (4
c), an adder (4d) that sequentially adds and outputs the output of the multiplier (4c) that changes according to the clock, and a main delay for the output data of the convolver (4). A digital delay shift register (5), and an adder (6) for adding data from all the elements subjected to the delay processing. Forma. 2. A convolver (4) for: (i) realizing signal processing on a frequency axis for an input signal, particularly realizing a band-pass filter function; (ii) controlling a delay time or a relative position of an output signal viewed from the input signal; (Iii) control of the scale factor or gain of the output signal as viewed from the input signal, (iv) compensation and equalization of variations in the characteristics of each receiver element, and (v) input signal as if it was scrambled or encoded. Correlation signal processing for descrambling or decoding, i.e. convolution integration,
(Vi) Any two or more operations such as spatial or temporal decimation or interpolation signal processing for one or more input signal sample sequences are performed by a function form of a kernel waveform or a relative phase or delay time for the frame. 2. A digital beamformer according to claim 1, wherein the digital beamformer is operated in a manner to perform a vernier control.