Electro-Optical Correlator

The invention relates to a system for electro-optically correlating two signals.

Such a system may be used for comparing a transmitted measuring signal to a returning echo signal by correlation. In the medical field such a procedure may find application, for example, in echography, in which ultrasonic sound pulses or wave packets are passed through a patient and the reflection caused by one or more organs of the patient is received. A similar procedure takes place in sonar systems.

Supposing that one signal, for example a transmitted signal, is g(t) and the other signal, for example a received signal, is f(t), the correlation function reads: ##EQU1## IN WHICH .tau. IS THE TIMING DISPLACEMENT BETWEEN F(T) AND G(T).

In general, the correlation function provides a signal that is defined better than signals achieved by the conventional techniques, such as envelope detection.

Hitherto, however, it has been difficult to form a correlation function as this requires both a continuous multiplication of two signals which are displaced relative to each other through all possible values of .tau., and an integration. These operations require complex computing arrangements and even then the computation of the correlation function takes a rather long time, so that this function is not available right away.

It is an object of the present invention to provide a system for electro-optically correlating signals which does not entail the above drawback. To achieve this object, in accordance with the invention a system of the above type is characterized by means for forming a running representation of the amplitude variation of one signal, by at least one masking means having a geometric configuration representing the amplitude variation of the other signal, and by means for observing the running representation through the masking means as well as for integrating the observed signal.

The invention will be described in detail hereinafter with reference to the accompanying drawings, in which:

FIG. 1 shows possible signal shapes f(t) and g(t) and the associated correlation function;

FIG. 2 shows an embodiment of a system according to the invention;

FIG. 3 shows a correlation mask for use in the system shown in FIG. 2;

FIG. 4 shows a variant of the embodiment shown in FIG. 2;

FIG. 5 shows another variant of the embodiment shown in FIG. 2; and

FIG. 6 shows an embodiment of a masking means suited for a specific purpose .

By way of example, FIG. 1 shows the correlation function Y(.tau. ) of two signals f(t) and g(t). For example, g(t) may be the electrical representation of an ultrasonic wave packet used in medical echography or in sonar, while f(t) may be the received signal.

FIG. 2 shows a system according to the invention. The signal f(t) is applied to a cathode ray tube 1 for modulating the intensity of an electron beam 2. The beam can be circularly deflected by means of cosinusoidal or sinusoidal voltages, indicated by cos .omega.t and sin .omega.t, respectively, which voltages are applied to horizontal and vertical plates 3 and 4, respectively. Thus a circular light track is formed on the screen 5, which track retains the intensity distribution of the signal f(t) for a short while due to the phosphorescence of the screen. In order to reproduce also the negative portions of f(t) on the screen 5, a DC component may be added to f(t). Preferably, the persistence has a value so that the wave packet f(t) or a relevant portion thereof is stored temporarily.

The screen 5 of the cathode ray tube may be reproduced through an objective 6 on the photocathode 7 of an image intensifier tube 8. The tube 8 comprises vertical deflection coils 9 and horizontal deflection coils 10, to which sinusoidal and cosinusoidal currents are applied, respectively, which currents correspond to the voltages applied to the vertical and horizontal deflection plates of the cathode ray tube 1, respectively. In this manner it is realized that the dot travelling on the screen 5 of the cathode ray tube 1 is reproduced on the anode 11 of the image intensifier tube 8 as a stationary dot located in the center of this anode, the phosphorescing track of varying intensity being reproduced on the screen 5 as an arcuate light track 12 originating from the central dot and rotating about this dot at an angular frequency .omega..

FIG. 3 shows a rotation-symmetric correlation mask 13 mounted on the outside of the anode 11, the light transmission of the mask in radial direction corresponding with g(t). As g(t) can have positive and negative values and negative light transmission is impossible, the mast 13 must have a mean transmission greater than zero. The correlation mask may be manufactured by known photographic techniques, starting from the shape of the transmitted wave packet.

The light track 12, which is a representation of f(t), is transmitted in this manner in accordance with a representation of g(t). The transmitted light signal, if necessary after deletion of a DC component, will be proportional to f(t - .tau.).g(t), which, as known from the theory of the correlation function, provides the same result as the multiplication f(t).g(t - .tau.).

FIG. 2 further shows a photomultiplier tube 14 mounted behind the correlation mask 13, which tube performs the integration over the area contributing to the correlation function. Instead of a photomultiplier tube a different light detector may be used. At the output 15 of the photomultiplier tube 14 a signal is produced in this manner which consists of the correlation function Y(.tau. ) and a DC voltage component, the latter being of no further interest.

In the arrangement described above the mask 13 corresponds to the transmitted signal g(t). However, it is also possible to use a mask including a function differing from the transmitted signal so as to be able to detect certain characteristic reflections. It is known from the art of medical echography that the different tissues have characteristic reflections, which implies that also tissue deviations have characteristic reflections. When a plurality of correlation masks is available which each correspond to a characteristic reflection, it is possible to detect known deviations, for example by successively placing the correlation masks behind the image intensifier. The amplitudes of the resultant correlation functions provide an indication of the deviation present.

FIG. 4 shows a further embodiment of the system shown in FIG. 2. In this embodiment the input signal is divided by means of an amplifier 40 into a positive and an inverted, negative portion, so that no DC voltage component need be added so as to be able to reproduce also the negative portion of the signal as a variation in the light intensity. The correlation is performed in a push-pull correlation system with appropriate correlation masks. The push-pull correlation system comprises two systems according to FIG. 2 as well as an additional input amplifier 40 and a recombination amplifier 41 at the output. The result is a correlation function Y(.tau. ) without added DC component. This has the advantage of optimally low photon noise.

FIG. 5 shows a correlation system suited for Doppler detection. Its operation is as follows: due to reflections of the ultrasonic wave packet to organs or objects with a velocity component in the direction of the sound wave, the received wave packet is slightly lengthened or shortened in time. When the received signal is applied to the input, the Doppler displacement can be determined by means of two correlation masks 52, 53 included in the system, one mask producing a slight magnification and the other mask producing a slight reduction relative to the mask associated with the transmitted signal. To this end, a second objective 50 is mounted behind the image intensifier tube 8, which objective 50 is followed by a beam splitter 51. Correlation masks 52, 53 of the type described above are mounted on either side of the beam splitter, each mask followed by a photomultiplier tube 54, 55, respectively. In this manner two output signals are produced whose ratio is indicative of the Doppler displacement and whose sum represents the intensity of the signal received.

It is observed that in actual practice the above systems may include a radially extending grey filter to compensate for the decreasing intensity of the light track on the anode screen of the image intensifier. This filter may be mounted between the image intensifier and the correlation mask.

The rotation frequency .omega. is a compromise determined by the persistence of the screen of the cathode ray tube and the track length to be used for each wave packet as, if after one rotation the phosphor of the cathode ray tube is not yet fully extinguished, a residual signal is formed which has an undesirable effect on the correlation function.

It is further observed that in the system described the input signal is reproduced on the screen of the cathode ray tube along a circular track. However, a different shape of the track, for example a straight line, is also feasible. In the event of a linear track the correlation mask should be adapted accordingly.

Finally, FIG. 6 shows the manner in which Doppler displacements can be detected by means of only one masking means. In this embodiment, the running representation of one signal consists of a succession of parallel lines 60 each formed by extending points (61) on a linear running representation having, for example, a light intensity exceeding a given value. The masking means 64 includes radially extending regions 62, 63 having different light transmission factors.