Pulse Doppler Radar With Variable Pulse Repetition Rate

1.Field of the Invention

The present invention relates to a pulse doppler radar system with variable pulse repetition rate, and especially to such a system having an arrangement for cancelling permanent echos and for integrating the reflected pulses.

2. The Prior Art

Various systems have been developed for ranging, moving and fixed targets using doppler radar, and such systems which have fixed pulse repetition rates are adapted to processing unwanted multiple-time-around echos.

When a pulse doppler radar system is employed with periodic switching of the pulse repetition rate, using a doppler filter to furnish a moving target indication, the cancellation of permanent echos from multiple-time-around ranges deteriorates. Whereas fixed echos from the unambiguous range are returned unaltered at various pulse repetition rates, the permanent echos from multiple-time-around ranges fall into different ranges, so that the permanent echo amplitude usually changes in every range. Accordingly, the basis for cancelling permanent echos in the doppler filter is lost, and the varying amplitude permanent signals can no longer be discriminated from moving target echos signals, and can only partially be cancelled by the doppler filter.

Due to the change of the pulse repetition rate from pulse to pulse (sometimes referred to as staggering), the echos of individual pulses, or noise signals that simulate reflected target pulses from multiple-time-around ranges fall into different range gates, or individual ranges. Therefore, the single pulses falling into the several range gates cannot be integrated. The probability of detection of such pulses is considerably reduced in the presence of superimposed noise. When an extensive clutter region extends over several ranges, the clutter corresponds to signals representing a plurality of distance ranges for the radar system, and then the probability that a plurality of target echos from a plurality of multiple-time-around ranges will accumulate within the same range gate rises considerably. In this case, permanent echos that derive from multiple-time-around targets can lead to erroneous operation.

As described in a German patent DE OS 3,110,279, it is known to stagger the pulse repetition rate such that at least two different pulse repetition rates are transmitted during the target sweep time. Targets from multiple-time-around ranges are distinguished from those in the unambiguous range because the echo signals from each individual pulse repetition fall into a different range gate. As a result, the permanent echos supply fewer signals to be integrated in accord with the number of pulse repetition rates employed and are attenuated in comparison to signals from the unambiguous range of the radar system. Since every switchover of the pulse repetition rate leads to transients in the moving target filter, which transients result in changes in the amplitude of permanent echos and thus disrupt the moving target indication, the first reception periods after each changeover of the pulse repetition radar are blanked out. Without a noticable loss of sensitivity, this type of staggering of the pulse repetition rate is particularly suitable for radar installations that have a high number of reflected signals during the target sweeptime. Because of cost consideration, the integration of the echo signals does not usually take place in a coherent fashion. Circuits which integrate the echo signals after they are received are relatively inexpensive but do not achieve an optimum effect.

Accordingly, it is desirable to provide a system with pulse doppler radar of the type having periodic switching of the pulse repetition rate in which deterioration of the moving target indication is prevented.

It is a principal object of the present invention to provide an apparatus and method for allowing pulse doppler radar of the type periodically switching the pulse repetition rate to prevent deterioration of the indication of a moving target because of pulses received as multiple-time-around echoes.

This object is achieved in the present invention by subjecting the echo signals to a coherent integration in an integration circuit comprising of plurality of filters, such integration being modified by means of simplified vectorial addition of the doppler signal pulses. The signal amplitudes obtained at the output of each filter are individually compared to the mean value of the video signals of all other filters, and manifested as a target indication in accordance with the result of the comparison.

The present invention allows multiple-time-around echo signals to be discriminated on a spectral basis from moving targets of the unambiguous range, with a small number of filters. The present invention also guarantees an optimum detection of moving targets from the unambiguous range in the presence of strong multiple-time-around clutter due to the modified coherent integration, since the echo signals of a range gate are vectorially added over a plurality of reception periods. The mean power of the multiple-time-around echos is distributed over a plurality of range gates as a result of the spectral staggering of the pulse repetition rate. The modified coherent integration enables the entire doppler frequency range of interest to be covered with a small number of filters and to statistically combine the different pulse periods.

In accordance with a further development of the invention, the advantages of coherent integration are also obtained in the case of the appearance of a single pulse, when the integration circuit is preceeded by a circuit for cancelling single pulses.

Reference will now be made to the accompanying drawings in which:

FIG. 1 is a functional block diagram of the signal processing branch of a pulse doppler radar incorporating vectorial integration;

FIG. 2 is a functional block diagram of a circuit for modified vectorial integration of the echo signals;

FIG. 3 is a functional block diagram of a circuit for implementing filter comparison;

FIGS. 4 and 5 are diagrammatic illustrations showing the staggering of the pulse repetition rate; and

FIGS. 6 and 7 are graphs illustrating transfer of functions of the integration circuit for two different staggerings of the pulse repetition rate.

Referring to FIG. 1, a functional block diagram of a digital processing system incorporating the present invention is illustrated. An analog to digital converter 1 is supplied at its input side with video signals from the I channel and from the Q channel by a phase detector (not shown). The analog to digital converter A/D unit 1 produces digital signals at its output which are supplied to a number of stages connected in cascade. The stages consists of EZPU circuit 2 for single pulse cancellation, an integration KOH-INT circuit 3, a filter comparison FV circuit 4, and a ZA target indicator 5. Since single pulses which are produced either as noise or from unrelated radar installations can represent multiple-time-around echos and seriously complicate a target acquisition, the single pulse cancellation unit 2 is employed for removing such single pulses. By this means detection of a moving target can be significantly improved in a environment with much noise. Since circuits for a single pulse cancellation are known, it is not necessary to describe the circuit 2 in detail herein.

The integration unit 3 performs a modified coherent integration of the echo signals. It considerably reduces the complexity of the filters which would be required for a coherent integration in a pulse doppler radar with variable pulse repetition rate. The only requirements for execution of the modified coherent integration consists in the use of a specific staggering of the pulse repetition rate, and of a phase backing of the signal voltages that has a constant value. When these conditions are met, nearly optimum target detection is provided. The echo signals of a range gate are vectorially added over a plurality of reception periods, and multiple-time-around clutter is largely cancelled by recognizing that the mean power of the multiple-time-around echos is distributed over a plurality of range gates. The desired effect can be achieved by means of a specific staggering of the pulse repetition rate, and when the pulse repetition periods are statistically combined, a large number of coherent filters is required when applying a coherent integration in order to cover the entire doppler frequency range of interest. Using the modified coherent integration according to the present invention, a substantially simpler and more economical construction results, by which multiple-time-around echos can be discriminated from the moving targets of the unambiguous range using a small number of filters. The discrimination itself takes place by means of a filter comparison performed in the circuit stage 4. Due to the noise character of multiple-time-around echos, they are greatly attenuated in power as a result of the narrow bandwidth of the coherent filters. In contrast to the moving targets within the unambiguous range, the multiple-time-around echos are transmitted in all of the filters.

In order to discriminate reflected pulses from the unambiguous range from reflected pulses resulting from the multiple-time-around ranges, the amplitude of the output voltage of each filter is compared to the mean value of amplitudes of all neighboring filters. As a result of this comparison, a target signal is furnished to the unit 5 only when the output signal of an individual filter has an amplitude that is a prescribed factor greater than the measured mean value of all other filters.

FIG. 2 shows a block diagram of the filter unit 3. Digital signals are supplied from its input to a shift register SR.sub.1. A series of such shift registers are connected in cascade, with the last shift register being designated SR.sub.n. There is one shift register provided for each of the signal values to be integrated, and they each supply a delay which corresponds to the pulse repetition period of the radar. The output of each shift register is connected to the input of an individual multiplier M.sub.1 through M.sub.n+1. The output of each multiplier is connected to inputs of a common adder AD. The adder AD adds the products of all of the multipliers M.sub.1 -M.sub.n+1 to arrive at a sum which is proportional to the mean value of all of the products.

In order to perform vectorial addition of the samples with coherent integration, a phase backing of the individual samples is necessary. This must normally be derived from the momentary pulse repetition rate and the filter frequency. In the case of the modified coherent integration of FIG. 2, all samples at the outputs of the shift registers are subjected to a weighting in their respective multipliers of a factor e.sup.-j .multidot..omega..sub.o .multidot.n.multidot.PRT. This operation brings about a constant phase baking so that the required phase condition for the vectorial addition is restored. In the weighting factor, .omega..sub.o is 2.pi. times the center filter frequency, n is the number of samples, and PRT is a mean period duration of the pulse repetition rate.

The discrimination of multiple-time-around echos from moving targets is explained in greater detail referring to FIG. 3 which is a block diagram of the circuit 4 of FIG. 1. The signal values at the output F.sub.1 through F.sub.n of all of the filters is supplied to a circuit MS which forms the mean value, and whose output signal GMW is supplied to a subtractor SS for further evaluation. All filter outputs F.sub.1 through F.sub.n are successfully sampled by means of a selection switch WS, and as a result multiple-time-around signals are furnished to the selection switch WS. At any position of the selection switch, one filter output is supplied as an upper end to the subtractor unit SS. In this way the mean value GMW of the output signals of all filters is changed into a mean value of the output signals of all filters other than the sampled one, which in the illustration of FIG. 3 is the filter F.sub.n. Thus, the output of subtractor SS is the mean value of all filters F.sub.1 through F.sub.1-n, forming a new mean value B. The new mean value B is compared in amplitude to the filter output signal A from the filter F.sub.n, which output has been weighted by means of the multiplier MW and supplied as the A input to the comparator KS. When a comparator KS determines that A is larger than B, such that the weighted value of the signal from the filter F.sub.n exceeds the mean value of signals from all of the filters, an output is produced from the comparator KS which is employed in the unit 5 to indicate a target. If A is less than B, no output signal is produced so that the signal is suppressed.

In order to obtain a comb filter structure in the integration circuit for the case of a quasi-statistical pulse repetiton frequency, the staggering of the pulse repetition times must be selected such that the sum of two successive pulse repetition times produces a value which is constant for all pairs of successive pulses. The comb structure becomes more regular if the staggering takes place in a more ordered fashion. A pronounced comb structure is produced when the n.sup.th transmission pulse follows the first transmitted pulse within an integration interval after (PRT+(-1).sup.n .multidot..DELTA.t), where PRT is a constant period duration and .DELTA.T is a constant fraction of this period.

Two examples of a possible staggering arrangement are illustrated in FIGS. 4 and 5. Both examples are based on a mean pulse repetition rate PRF=4 kHz having a corresponding pulse repetition time PRT of 250 .mu.s. As illustrated in FIG. 4, the period of successive pulses alternates between the values 2/3.multidot. PRT and 4/3.multidot.PRT. The mean value of any two successive pulse repetition periods is a constant value, namely 250 .mu.s. Such a pulse repetition rate staggering is defined according to the equation

In the example of FIG. 4, .DELTA.T=1/3PRT.

FIG. 6 shows the transfer function of a modified coherent integration in accordance with the present invention of 16 pulses in accordance with the staggering of FIG. 4. The phase backing of the signal values as the constant value .DELTA..phi.=2.pi..multidot.f.sub.o .multidot.PRT, in which f.sub.o is the center of filter frequency and is equal to 2 kHz. The comb structure resulting from the modified coherrent integration is clearly shown in the transfer function of FIG. 6. An important feature of this transfer function is that the frequency range around the doppler frequency 0 lies in the stop band, so that only moving targets result and signals pass through to the target indicator 5.

In FIG. 5, a second example of the staggering of the pulse repetition rate is shown. It is based on the same pulse repetition period PRT=250 .mu.s. This staggering of the pulse repetition rate of FIG. 5 is formed according to the equation

The number of different pulse repetition frequencies which may be employed in accordance with this equation is not restricted to two different frequencies, but can assume any large value. The essential feature of the staggering is that any two successive repetition periods have a constant mean value.

FIG. 7 shows the transfer function obtained with the coherent integration of 16 pulses staggered in accordance with the example of FIG. 5. In this arrangement, .DELTA.T is one 1/3.multidot.1/16.multidot.PRT. The constant phase backing necessary to the modified coherrent integration is again .DELTA..phi.=2.multidot.f.sub.o .multidot.PRT, whereby f.sub.o =2 kHz. The transfer function of FIG. 7 also shows the comb structure with a null in the region of the doppler frequency 0.

It is apparent that the arrangement of the present invention furnishes a simple and economical means of discriminating against over-the-horizon pulses which would interfere with the clear indication of a target within the unambiguous zone. Various modifications and additions may be made to the apparatus of the present invention without departing from the essential features of novelty thereof, which are intended to be defined and secured by the appended claims.