Russian electronics offer more scope

by Rob van den Berg

Over the past few decades, an unorthodox school of thought on radio and radar testing techniques evolved in the former Soviet Union. Whereas Western thinking focuses on using single-frequency signals, our Eastern counterparts preferred the use of pulse techniques, in which each separate pulse could cover a wide frequency range. Russian scientists have managed to extract valuable information from signals that appear indecipherable to Western eyes. For example, a single air-borne radar measurement provided information about both the vegetation on a site and the objects hiding under it.

Ex-Soviet radar sees more efficiently than Western radar

Ten years ago, Dr. Ir. Leo Ligthart, professor at Delft University of Technology, started research to find out how antenna and radar measurement tests were being carried out at Russian research centres. At the time, he was the only person from the west showing an interest in this aspect of Russian radar technology. His pioneering work resulted in a flourishing co-operation with Russian research institutes and universities, and a unique test facility at the Faculty for Information Technology and Systems. By combining his knowledge of electronics from behind the former Iron Curtain with state-of-the-art Western signal processing techniques, he created a unique facility that has generated enormous interest from industrial clients.
In the science of measuring, the time-honoured adage of Kamerlingh Onnes, ‘To measure is to know’, often changes to ‘To continue to measure is to know’. Anyone attempting to determine antenna characteristics across both a wide frequency range and a large angular range must have ample measuring time
at his disposal. That is, if the classic measuring method is used. There is an alternative, much quicker, method that uses pulses in quick repetative succession. However, this method places such high demands on the radio and signal processing equipment, that the traditional, labour-intensive method has become generally accepted practice all over the world.

At first glance, the two instrumentation systems on the table look very similar. The one on the left is clearly an American product, manufactured by the well-known Hewlett-Packard company.
‘Straight off the shelf’, says Ligthart, managing director of the International Research Centre for Telecommunication Transmission and Radar (irctr) at Delft University of Technology. The same does not apply at all to the equipment on the opposite side of the table. A closer investigation reveals it to be of Russian origin. It takes a little detective work on the front panel texts to work out that one of the boxes can be used to generate impulse-like waves. Ligthart points out the serial number, k-263-3.
‘This is number three of the series. And the sampling scope next to it is the only one of its kind in the world, manufactured specially for the irctr for antenna measurements’, Ligthart says.
He walks to a heavy metal door that slowly slides to one side. The door provides access to the anechoic antenna measuring chamber, an insulated, foam-rubber clad room that completely absorbs radio waves. Inside the room, the properties of different types of antenna can be measured.
There are two different measuring methods, based on entirely different principles. In the West, the usual procedure is to determine the behaviour of an antenna across its full bandwidth in frequency steps. The Russian approach is quite different. Based on the principle that impulse signals comprise a large number of frequencies, the antenna's behaviour for the entire frequency range can be determined in one measurement by feeding it repeating pulses. Both methods can be used together in this antenna measuring chamber, which in itself is reason enough to consider the facility on the 22nd floor of the Electrical Engineering sub faculty unique.

Characteristics
Earlier, Ligthart explained about the procedures involved in measuring and characterizing very high-frequency analogue microwave circuits, from single high-frequency components to complete systems, and also antennas. Some form of electronic signal has to be fed to the device, and then the researchers have to analyze the behaviour of the component or system by measuring the signal responses. Generally speaking, a part of the input signal will pass through the component, while another part will be reflected. The usual measuring method uses single-frequency harmonic signals. For antenna measurements, these signals range from a few hundred megahertz up to about one hundred gigahertz (so-called millimetre waves). The output signal of the system under scrutiny can generally be described by means of its amplitude and time relationships relative to the original input signal. Where the original signal started by increasing from a zero value, the reflected signal might well show a decrease in value, i.e. a phase shift of p radiants, or 180 degrees.

Observations like this can reveal many things to those in the know. The only thing left to do is to measure the accepted and reflected signals to determine the characteristics of the circuit or system. However, to determine the characteristics of an antenna across a sufficiently wide frequency range requires continued measurements due to the multitude of frequencies, in particular if the angle dependency of the antenna signals is to be measured. Ligthart shows that in this way, a standard 180 degree elevation and 360 degree azimuth measurement for 100 frequencies will take well over 100 hours to complete. Cumbersome indeed, especially when you know that there is a more efficient way.

Fourier
Why not present the antenna with all the frequencies of interest in one go? After all, the mathematics involved have been known for a long time. The French mathematician, Jean Baptiste Fourier, laid the foundations for this in the early nineteenth century.
Ligthart: ‘According to Fourier’s theorem, one can consider a periodically repeated pulse lasting a few picoseconds to be the sum of a large number of different waves that are harmonically time-dependent. So in fact, a pulse-like signal that will cover the entire range of required frequencies simultaneously. Added to that, it is relatively simple to produce large numbers of pulses per second. Each pulse fed into the circuit will pass or be reflected in the same way. The results can then be added together in order to be able to distinguish even the weakest signals from all the unwanted noise. Using Fourier analysis, the distortion can be used as a basis for retrieving information on every frequency. It sounds like the perfect solution, but there are some drawbacks. The technique places high demands on the detection equipment, which must be extremely reliable, sensitive and stable. In addition, the signal to noise ratio tends to be rather low and although this can be remedied to some extent by increasing the power of each pulse, this is not sufficient. So, we've opted for a high pulse repetition frequency. By generating 100,000 pulses per second, we can measure and determine the average of 100 consecutive pulses. Basically, this method still enables us to carry out the measurement up to 1000 times faster than we could by using the standard technique of measuring in stepped frequencies.’

Even so, the time domain method is rarely used, and the time-consuming, classic frequency domain techniques still prevail. The main reason for this is the lack of stable pulse generators that are capable of producing the required pulse sequences. There is also a lack of equipment capable of sampling the short duration measuring signals at sufficiently high speeds. Probably as a result of the continued lack of interest in the West, the specifications of commercially available generators and sampling equipment as produced by the major manufacturers are now lagging behind the capabilities offered by equipment from Eastern Europe and Russia.

Inaccessible
The pulse method also offers advantages in other fields than measuring techniques. Ligthart, who for years has been researching civil radar applications such as ground penetrating radar, explains that the former East bloc countries developed radar systems for many different applications based on the principle of repeating very short duration pulses, as opposed to the radar technology being developed in the West, which focused on the classic constant-frequency methods. The political tension between East and West meant that the technology from behind the Iron Curtain remained inaccessible.
Ligthart: ‘Through conferences and contacts with Russian research centres I still managed to gather quite a lot of information about radar research in the former Soviet Union. I had a fair idea about what might be interesting. Shortly after the fall of the Berlin Wall, I went to Moscow to see their pulse radar techniques for myself. I couldn’t believe what I saw! ...They used simple components, many with enormous specification in tolerances... Even though each component was different but within the tolerance after installation the various parts without exception resulted in very reliable systems.

Respect
Ligthart learned to respect the great achievements of the Russians. ‘Without using any forms of automatic adjustment or computer control, and without any advanced real-time signal processing techniques, they still managed to extract all sorts of information from the (pulse) signals, enabling them to realize superlative radar capabilities.’
For example, pulsed radar used for earth observation purposes can be directed at a wooded area, and the high frequencies in the pulse will be reflected by the leaves in the top of the trees, whereas the lower frequencies will penetrate the crown. This makes it possible to see both the trees and what’s under them in a single observation.
Ligthart: ‘You would never have found that type of system here in the West. In the first place, in our environment the frequencies in radar pulses would have overlapped all kinds of communication channels. In addition, the high-energy transmission signals required across a wide frequency spectrum would have resulted in unacceptable interference levels.’
Ligthart was one of the few people to see great opportunities in co-operating with scientists from the former Soviet Union to develop advanced high-frequency measuring systems.
‘Linking their measuring equipment to our processing power simply had to result in something great,’ Ligthart explains. This was in the early nineteen nineties, when they still had a long and arduous road ahead. The researchers in Russia and Lithuania used special oscilloscopes that had been developed for Vimpel, an institute comparable to tno in Holland. During the nineteen eighties, Vimpel, supported solely by defence contracts, had grown into an organization with 100,000 employees. After the fall of communism, part of the concern was privatized, but engaging the remainder of Vimpel to set up a joint development operation in Russia proved highly complex.
‘They just haven’t quite mastered the art of working in projects’, Ligthart recalls. ‘Also, working in Russia or Lithuania requires a very specific approach, and this was a very high-tech project. Nothing worked when we tried to integrate the existing Russian equipment into the existing irctr facilities. They used different types of connectors, different thread dimensions, and different cable specs. And interfacing is very important. After all, we had to be able to adjust and control the Russian hardware and software using our own equipment.’

A student at the time, René de Jongh, played a major role in this. He stayed in Moscow for two months, and later went to Vilnius, Lithuania, to gain experience with the measuring equipment, and to make the integration with the equipment at Delft a bit easier. The stays taxed his powers of improvisation to the limit. Finish and attention to detail turned out to be non-existent qualities in the ex-Soviet Union. And to top it all, only a handful of the people involved had an adequate command of the English language. Even so, the final result, a unique measuring facility, is magnificent.
Prof. Ligthart: ‘This is the only place in the world where antenna characteristics can be measured with these specifications and under identical conditions, using both methods simultaneously, i.e. in both the frequency domain and the time domain. This not only opens the way to fascinating comparative studies, it also enables us to make good use of the advantages offered by each of the techniques.’

Lack of interest is the last thing Ligthart’s group at Delft can complain about. The manufacturer of (military) radar electronic equipment Hollandse Signaalapparaten and the esa/estec space technology centre have announced their desire to use the facility for measurements.
stw, the Netherlands Technology Foundation, has also contributed to the project. Thanks to STW, Russian scientists were able to come to Delft. The Erasmus University in Rotterdam was also involved, helping to market the Russian electronics products, and offering services such as supervision and advice for patent applications to ensure that the intellectual properties of the Russian partners and the IRCTR remain respected.

The Delft facility is currently staffed by seven scientists. The first phase of a recently proposed stw project has now been approved, and will provide 55 man-years of work over the next four years. The project involves co-operation with departments of the Delft ITS faculty and the sub faculty of Applied Earth Sciences.
The project will investigate new systems and measuring methods to make Ground Penetrating Radar systems suitable for the detection and classification of land mines in former theatres of (civil) war. Again, the Russian research centres and the Russian measuring methods provide a major contribution.

For more information please contact Prof. Dr. Ir. L.P. Ligthart (l.p.ligthart@its.tudelft.nl), phone +31 15 278 6230