IEEE Aerospace and Electronic Systems Magazine - September 2020 - 6

DOI. No. 10.1109/MAES.2020.3006367

NOISE RADAR SPECIAL ISSUE-PART 1
Konstantin Lukin, Institute for Radiophysics and Electronics of the National
Academy of Sciences of Ukraine; and University of Pardubice
Andy Stove, Stove Specialties
Christoph Wasserzier, Fraunhofer Institute for High Frequency Physics and
Radar Techniques

oise radar" is a convenient name for radars
that use essentially random signals either as
the transmitted signal or as the modulation.
Employing analog random signals can be traced back to the late
19th and early 20th centuries. At that time, noise pulses of a
spark generator were used by Christian H€ulsmeyer, in 1904, in
his Telemobiloscope, now considered as "radar precursor" and
by Alexandr Popov, in 1897, in his experiments on ship detection in the bistatic configuration. The use of random signals
with coherent reception was first considered in the late 1950s.
The first paper on radar based on noise signals was published by
R. Bourret in 1957, followed by that by B. M. Horton in 1959.
Early experiments used physical delay lines to match the
timing of the correlation reference to the target range and
could only process a few range cells. The technique was
nonetheless of significant interest because intercept receivers
would see nothing but noise from the transmitter. The fact
that the waveform is nonrepetitive also means that noise
radars have no ambiguities in either range or Doppler.
Guest Editors at Pendik, near Istanbul, in December 2018. Left to
However, the noise radar concept did not attract much
right: Christoph Wasserzier, Konstantin Lukin, and Andy Stove.
attention from radar engineers at that time, mainly because of
the lack of appropriate sources of noise signals and the difficulties in their coherent reception. With advances in microwave and digital electronics, it is now possible to perform both the random
signal generation and the matched filtering digitally using Fast Fourier Transform techniques so that targets can be detected efficiently over a useful range swathe. Interest in noise radars has naturally expanded in the last three decades and has created a significant community, as may be seen from the set of conferences, special issues, and dedicated sessions on noise radar technology.
The random signals may be samples of a physical random process or they may now be generated digitally. Once the process has
been sampled, the radar itself "knows" what it has transmitted and so can receive its signals by applying the matched filter. The
matched filter theorem shows us that noise radar can have the same sensitivity and resolution as any other radar with the same characteristics-the mean power determines the sensitivity, the bandwidth determines the range resolution, and the coherent integration
time determines the Doppler resolution. Additionally, interest in covert radars has also grown and stealth goals can be met by noise
radars since one can prove that there is no way to intercept them more effectively than one can detect random noise.
Another development that has provided a "market pull" for noise radar research has been the increasing demands on the
radio spectrum from commercial users. Noise modulation provides multiple quasi-orthogonal waveforms and so offers the
potential for many radars to operate the same spectrum at the same time.
In the limit of infinite integration time two noise waveforms will have zero cross-correlation, but for finite integration times the
cross correlation will be below the autocorrelation by a ratio equal to the time bandwidth product of the signal. The random nature
of the reference waveform means that many of the artifacts which limit the performance of other radar systems, such as range sidelobes, the effects of receiver non-linearities, and deterministic interference, are converted into random signals, with, in practice, the
same statistical properties as noise. This means that the power in all these signals is spread over all the possible range cells and are
reduced by a factor equal to the time bandwidth product, which is of course the product of the signal bandwidth and the integration
time used in the correlator. In order to reduce the levels of the unwanted signals designers of noise radars have even more incentive
than most radar designers to maximize this product.

IEEE A&E SYSTEMS MAGAZINE

SEPTEMBER 2020

IEEE Aerospace and Electronic Systems Magazine - September 2020

Table of Contents for the Digital Edition of IEEE Aerospace and Electronic Systems Magazine - September 2020