Pulse compression method
The pulse compression method is a method of signal processing which u. a. is used in location processes such as radar technology , lidar and sonar . The otherwise used, very short and powerful transmission pulses are replaced by longer, modulated transmission pulses with reduced power. With the help of modulation, for example as chirp or, in the case of digital methods, in the form of pseudo-noise , the same resolution can be achieved with the time-stretched and reduced power transmission pulse by means of a correlation in the receiver as with the very short and powerful transmission pulses .
How classic radar works
With pulse radar without pulse compression , extremely short pulses with powers in the megawatt range are emitted. With increasing pulse power, greater ranges can be achieved because this is the only way to ensure that the reflected, very weak signal is not covered by the inherent noise of the receiver. The shorter the duration of the transmission pulses, the better the distance resolution and different objects of almost the same distance can be displayed separately. In practice, this means high pulse powers with a transmission duration of a few microseconds, which places high demands on the respective transmission tubes .
To illustrate the resolution, the transmission signal is shown in red on the left-hand side of the picture in the following pictures, the echoes from two targets are shown in blue. In the two images on the left, the background noise (inherent noise of the receiver) has been omitted, which would be almost as large as the weak received signals. Since the amplitude characteristic of the transmission signal is known, which can signal-to-noise ratio by an adapted matched filter ( matched filter will be greatly improved). As a result, the originally rectangular envelope becomes triangular in shape, making it difficult to separate targets of similar distance.
| Received signal before the optimal filter if there were no receiver noise |
filtered signal at the output of the receiver. The transmission pulse is hidden |
|---|---|
How a radar works with pulse compression
With the use of amplifier modules in semiconductor technology , high pulse powers cannot be achieved or only with great difficulty. With pulse compression, significantly longer transmission pulses are generated, the frequency of which is specifically changed within the transmission pulse duration. This frequency modulation enables a time reference within the transmission pulse, similar to how it is carried out with frequency-modulated continuous wave radar devices (FMCW radar). The time course of the frequency change can be controlled linearly or non-linearly in analog devices with voltage-controlled oscillators ; in digital devices, coded pulse-phase modulation is also used.
The considerable increase in the duration of the transmission pulses would have considerable effects on the resolving power of the radar device if it were not possible to significantly shorten (compress) the relatively long pulse duration of a few microseconds so that the transit time of the radar signal can be measured precisely. Special pulse compression filters were developed for this purpose: Acoustic surface wave filters with a special arrangement of the “fingers” as a hardware solution and implemented as an FIR filter for digital devices .
The following images show the modulation with a chirp and the signal improvement through the compression filter (matched filter) adapted to the transmission pulse:
Advantages and disadvantages of pulse compression
- The impulse maximum due to the correlation with the transfer function of the compression filter is sharply limited in time and thus allows a high spatial resolution.
- Reflections on different objects can be separated and differentiated in the receiver, the distance between which is smaller than the duration of the transmission pulses would allow.
- The reduction of the pulse power has the significant advantage for air surveillance radar that an enemy reconnaissance of the radar is more difficult and is often only possible if the reconnaissance system knows the exact image of the modulation. That is why the term “silent radar” is often used.
- Since the noise is always broadband and the frequency-synchronous component of the noise is rather low compared to the echo signal due to the statistical distribution, this type of filtering reduces the component of the noise in such a way that a clear output signal is achieved even if the input signal is so small that with a classic radar it has long been drowned out in the noise and is therefore lost for simple demodulation.
- The radar receiver becomes considerably more sensitive with pulse compression and a suitable choice of pulse duration and frequency deviation , the power of the received signal appears to be amplified by the factor (usually in the range 20 to 30).
The following figures show how the received signal before the pulse compression filter is disturbed by strong additional noise and how insignificant this noise is after the pulse compression filter.
The main disadvantage of the pulse compression method is the worsening of the minimum possible distance, because nothing can be received as long as there is transmission (“dead zone”). The receiver is switched off for the duration of the temporally stretched transmission pulse, so the radar is "blind". Since this is a disadvantage especially for air traffic control radar devices, they usually work alternately with both methods. Between the frequency-modulated pulses for a long range, small and very short pulses are transmitted which only have to cover the close range and therefore do not require a large pulse power.
Example of a radar application
The following data from the American Cobra Dane radar should serve as an example of a technical implementation : The transmission frequency is set within the duration of a transmission pulse (1000 µs) with a rate of change of 0.2 MHz / µs (up-chirp) from the starting frequency 1175 MHz to 1375 MHz increased. As a reference, the oscillator frequency of the heterodyne receiver is changed from 1665 MHz to 1865 MHz and is correlated with the received signal, which is why incoming received signals are mixed down to the intermediate frequency 490 MHz. After a further reduction to 60 MHz and amplification, sampling is carried out by two analog-digital converters in order to provide the baseband signals required for the I&Q process . Further processing is carried out by digital signal processing .
Technical details on frequency modulation
Linear frequency modulation
With this pulse compression method, the transmission pulse is linearly frequency modulated. This has the advantage that the circuit can still be kept relatively simple. The transmission pulse is divided into a number of time intervals with an assumed constant frequency. Special filters for precisely the frequency in the respective time interval produce an output signal that is added to an output pulse in a cascade of delay lines and summing stages.
Linear frequency modulation has the disadvantage that so-called "sweepers" can generate interference relatively easily. The RRP 117 can be named as an example of an application of linear frequency modulation . The disadvantage of susceptibility to interference is compensated for by the transmission of two different carrier frequencies, each with linear frequency modulation.
In the adjacent circuit example, the principle is shown using five frequencies in the transmission pulse. The high level of circuit complexity can be managed with today's integration options. There are practically two basic ways to technically implement this process:
- a processor-controlled data processing (after an A / D conversion)
- with an AOW filter
Surface acoustic wave filters
SAW filter , and (SAW filter English Surface Acoustic Wave ) called, are often used in radar systems with pulse compression and compress the frequency-modulated echo signal in an analogous manner. They work on the piezoelectric principle.
A broadband transducer is vapor-deposited on a piezo crystal, which converts the electrical vibrations into mechanical vibrations in the crystal. However, these mechanical vibrations propagate at a much slower rate than the electrical signals on a line. Therefore, relatively long delay times are achieved. Frequency-dependent converters that convert the mechanical energy back into electrical signals are also vapor-deposited on the same crystal.
Due to the necessarily different distance between these different converters and the excitation system, the different frequency components of the input signal are given a different time delay, so that all frequency components of the input signal are shifted to the same point in time and are thus interpreted by the radar device at the same distance. The sum of the delay must be taken into account when calculating the distance, i.e. subtracted from the total running time.
Time sidelobes
Since the frequency-dependent converters, like any filter, can also be excited by harmonics , in addition to the sharp output pulse, disturbing side lobes, so-called “time-side lobes”, arise, which often have to be compensated for by complex processes.
Since both the temporal and the amplitude distance are constant, these side lobes can be reduced to an acceptable value by weighting the signal amplitudes. If this amplitude weighting is only carried out on the receive path, however, it also causes a deterioration in the filter and reduces the signal-to-noise ratio. The size of these side lobes is an important quality criterion in the pulse compression process and can be reduced to a value in the range of −30 dB by this amplitude weighting.
Non-linear frequency modulation
Pulse compression with non-linear frequency modulation has some distinct advantages. So she needs z. For example, for the suppression of the time-side lobes that arise during compression, there is no longer any amplitude weighting, since the form of the modulation already fulfills the function of the otherwise necessary amplitude weighting. This enables a filter adjustment with steeper edges. In this way, the losses in the signal-to-noise ratio that otherwise occur due to the amplitude weighting are avoided. There are two ways of non-linear frequency modulation:
- a symmetrical shape, and
- an asymmetrical shape.
The symmetrical form of the modulation has an increasing (or decreasing) frequency change during the first half of the transmission pulse duration and a decreasing (or increasing) frequency change in the second half. An asymmetrical form of modulation is obtained if only half of the symmetrical form is used.
The disadvantages of pulse compression with non-linear frequency modulation are:
- a very complicated circuit structure and
- a complicated modulation so that each transmission pulse has exactly the same properties in the amplitude weighting.
Usually this type of frequency modulation is generated by special waveform generators that generate a processor-controlled pulse shape.
Pulse compression with phase modulation
The phase-coded pulse shape differs from the frequency-modulated pulse shape in that the long overall pulse is divided into smaller sub-pulses of the same frequency. These sub-impulses always represent the smallest resolvable distance. These sub-pulses all have the same length and within this pulse duration the phase is constant. A phase jump can be programmed between the sub-pulses. This phase jump is usually linked to a binary code. With a number of 13 sub-pulses in the transmission pulse ( Barker code as shown in the adjacent picture), the time side lobes have a size of −23 dB. (This pulse pattern was also used by the AN / TPS-43 radar .)
The binary code consists of a sequence of logical states. Depending on this binary code, the phase position of the transmission signal is switched between 0 and 180 °. In contrast to the greatly simplified picture shown, the transmission frequency is not necessarily a multiple of the frequency of the switching pulses. The coded transmission frequency is generally switched disharmoniously at the phase reversal points .
Web links
Individual evidence
- ^ JR Klauder, A. C, Price, S. Darlington, WJ Albersheim: The Theory and Design of Chirp Radars , In: Bell System Technical Journal , 39, 1960, p. 745.
- ^ Achim Hein: Processing of SAR Data: Fundamentals, Signal Processing, Interferometry . Springer, 2004, ISBN 3-540-05043-4 , pp. 38-44.
- ↑ Merill I. Skolnik (Ed.): Radar Handbook . 3. Edition. Mcgraw-Hill Professional, New York NY 2008, ISBN 978-0-07-148547-0 , Chapter 8 “Pulse-Compression-Radar” , pp. 8.36 .