RADAR SIGNAL PROCESSING SYSTEM
This invention relates to signal processing circuits, particularly adapted for the processing of radar video 5 signals, which compress the dynamic amplitude or intensity range of the video signals in a fashion to make most effective use of the display. The display unit can be a CRT or some other type of display unit with limited intensity range. That is, such display units have a lim- 10 ited range of gray shades which can be distinguished by an operator, and to make maximum use of the received information it should be presented so that information of significance falls into regions of intensity most readily visible by the operator. 15
In its most general form the present invention relates to a signal processing system intended to adapt a received signal for use with a display having a limited dynamic range. The phase "dynamic range" is used to denote the ratio of maximum to minimum brightness 20 discernable on the display. The system comprises apparatus for estimating the mean value of the signal over a selected portion of the area of the display, a variable gain amplifier in the path of the signal coupled to an analog-to-digital converter to provide a sequence to 25 signal samples, a statistical analyzer responsive to the signal samples to measure the spread of the amplitude distribution of the samples, means controlling the gain of the amplifier to maintain the spread of the distribution matched to that of a model distribution and signal 30 compression means responsive to the output of the amplifier to produce a signal of limited dynamic range for use with the display. The mean value of the signal is normally the arithmetic mean but other mean values may be used in different applications. The phrase 35 "spread of the amplitude distribution" is used as a general term corresponding to the variance of a binomial distribution or a similar measure of deviation from the mean value.
Also specifically disclosed is a method and apparatus 40 for estimating the mean of a sequence of digital signals comprising: generating pseudo-random sequence of pulses synchronized to the digital signal sequence to obtain a sequence of sampling signals from a characteristic of the pseudo-random sequence, selecting and stor- 45 ing such digital signals as occur in coincidence with the sampling signals, accumulating the values of the selected and stored digital signals up to a predetermined number thereof, thereafter increasing the accumulated amount by each next selected digital signal and decreas- 50 ing the accumulated amount by the value of the digital signal occurring earlier by the predetermined number in the sequence selected, and dividing the number in the accumulator by the predetermined value to obtain a measure of the mean. 55
The present invention also contemplates a method and apparatus for maintaining the spread of the amplitude distribution of a signal matched to the spread of a model distribution comprising the steps of: passing the signal through a variable gain amplifier, sampling the 60 output of the amplifier to provide a sequence of discrete sample values, counting those sample values which exceed a preset upper tail break-point value and using the measure thus determined to control the gain of the amplifier. 65
Yet another aspect of the invention contemplates a method and apparatus for transforming input signal samples of a certain amplitude distribution to a com
pressed amplitude distribution, including a system for establishing the break-points of the compressed distribution, comprising: storage means having addresses corresponding to each possible discrete input sample amplitude, means incrementing by one the number stored at an address on detection of an input sample of corresponding amplitude, a first counter counting the total number of signal samples and dividing the total by a factor representing the number of possible amplitudes in the compressed distribution to obtain the number of samples occurring at each amplitude level of the compressed distribution, a second counter responsive to the storage means to accumulate the stored numbers when read out in sequence including means to indicate when the accumulated count exceeds the last-mentioned number thereby establishing the address last read as the first break-point and further indicating when the accumulated count exceeds twice the last-mentioned number thereby establishing the second break-point and so on.
An alternative manner for performing this last described function is described, being a method and apparatus for compressing the amplitude distribution of a series of signal samples falling within a finite amplitude range, comprising: supplying the series of samples in parallel to a plurality of comparators having monotonically increasing level settings, detecting the comparator having a level setting immediately above the amplitude of the applied signal sample and generating an output signal indicative of that comparator, counting the output signals from each comparator, and adjusting the discriminating level settings so that the counts attributable to each comparator remain substantially equal.
Particular embodiments of the invention will now be described in conjunction with the accompanying drawings in which:
FIG. 1 is a block diagram showing the interrelationship of the inventive components of a specific embodiment of the invention,
FIG. 2 is a diagrammatic representation of the amplitude distribution of a typical radar display signal,
FIG. 3 is a block diagram similar to FIG. 1 but in greater detail,
FIG. 4 is a block diagram of the mean estimating circuit 11 of FIG. 1,
FIG. 5 is a block diagram of a circuit for maintaining a constant variance,
FIG. 6 is a block diagram of an adaptive dynamic range compression circuit,
FIG. 7 is a typical histogram of an exemplary amplitude distribution curve showing break-points, and
FIG. 8 shows a further embodiment of an adaptive dynamic range compression circuit.
DETAILED DESCRIPTION OF THE
FIG. 1 shows a block diagram of typical components of a signal compression system which functions according to this invention to compress a wide dynamic range of radar video signal returns to a narrow dynamic range of intensity sufficient for the usual CRT display. Such signal returns, for example, may have of the order of 12 bits of information while only 4 bits are needed at the CRT display. The dynamic range is modified by the invention to provide the operator with the maximum amount of information within the capability of the CRT display. The radar video signal is supplied to terminal 10 as an analog signal. The amplitude distribution of the video signal is preferably modified by subtracting the