Aerospace and Electronic Systems - October 2018 - 51

Book Review

DOI. No. 10.1109/MAES.2018.180062

Book Review of "MIMO Radar: Theory and Application"
Jamie Bergin and Joseph R. Guerci

Review by Fred Daum, IEEE Fellow

IMO radar is a very controversial hot topic. MIMO radar researchers have claimed 10 dB or 20 dB advantages relative to
boring old phased array radars, but crusty old radar engineers
have doubted such assertions and poked holes in their analysis. Ten
years ago the Office of Naval Research (ONR) asked me to become
embroiled in this debate, resulting in [5]. The authors of this new
book tackle this problem head on. In particular, they say clearly and
emphatically that: "The hype surrounding MIMO radar has come
with a fair amount of skepticism. MIMO radar is not a cure-all for
every radar problem.... Unfortunately, the benefits of MIMO radar
have sometimes been overstated, which has created a cloud of controversy that has generally slowed its adoption for radar modes and
applications where it could have great benefit. In this book we attempt to present a fair analysis of MIMO radar technology in a way
that clearly highlights its benefits and pitfalls" (page 11). In this
review, we shall see how well the authors have succeeded at this
task. For example, on pages 72 to 75 there is a nice simple analysis
of the bad impedance match of a typical MIMO antenna with free
space, which results in a very poor radiation efficiency, and hence
potential damage to the radar transmitter. That is, if the microwave
energy is not radiated into free space, then it stays in the transmitter,
which heats up and could suffer severe damage. That is, the typical
MIMO antenna does not radiate efficiently like a boring old phased
array radar or an extremely boring old dish radar. In fact, the MIMO
radar transmitter might actually melt as a result. This is a simple
fact of physics, first explained by Oliver Heaviside [9] well over
one hundred years ago, and it is taught to every student of antenna
design or electromagnetic physics. Unfortunately, the authors do
not mention that there is a very simple solution to this problem: design the MIMO transmit pattern in beam space rather than element
space, in which each beam is designed with high efficiency of radiation as explained in [3]. About 15 years ago a famous MIMO radar
researcher (who will remain nameless) once visited us to extol the
virtues of MIMO radar, and one of our smart aleck antenna designers in the back of the room pointed out that the MIMO transmitter
would melt, to the eternal embarrassment of the famous professor.
Chapter 3 of this book gives a nice simple analysis of the extremely large computational complexity required for MIMO radars that have N2 receiver channels rather than only a few receiver
channels as in SISO radar or only N receiver channels as in SIMO
radar. Likewise, this chapter explains the difficulty of actually calibrating all N2 receiver channels, including all N2 transmit-receive
paths through the antenna, waveform generators, transmitters and
receivers as well as the numerous propagation paths (through the
troposphere or ionosphere) and possibly the radome, compared

OCTOBER 2018

with the relatively simple
task of calibrating a boring
old phased array radar which
has only a single propagation path. Also, this chapter
explains the extra difficulty
of collecting enough data
samples for adaptive MIMO
radar algorithms, considering
that the number of required
samples grows quadratically
with the number of MIMO
degrees of freedom. Obtaining enough data samples is
especially challenging in real
world clutter or jamming,
which is typically highly heterogeneous and non-stationary, in contrast with textbook clutter or jamming found only in
academic papers.
Chapter 5 gives a nice clear tutorial introduction to optimal
MIMO radar, using the standard least squares approach for the
general case of correlated MIMO waveforms, not just orthogonal
MIMO waveforms. Unfortunately, almost all of the other chapters
in this book assume N orthogonal MIMO waveforms, which suffer
the maximum loss in signal-to-noise ratio relative to a boring old
phased array radar (SISO or SIMO), owing to zero transmit antenna gain and the resulting omnidirectional transmit antenna pattern. Most serious MIMO radar experts today agree that orthogonal
waveforms are a bad idea in practice, and that correlated MIMO
waveforms should be used in essentially any application, as explained in [2]. Nevertheless, the academic literature as well as this
book is still dominated by the assumption of orthogonal MIMO
waveforms. Curiously, Chapter 5 fails to mention the most fundamental fact about optimal MIMO radars; namely, that the boring
old phased array radar (SISO or SIMO) maximizes signal-to-noise
ratio, and any MIMO radar is always suboptimal [12], assuming
thermal front end noise and no clutter or jamming.
Figures 3.10 to 3.13 compare MIMO radar with so-called "conventional radar" performance, based on the Monte Carlo simulations reported in a 2009 IEEE Conference Proceedings paper that
does not exist in the list of references at the end of this chapter, and
hence we cannot learn exactly what kind of "conventional radar"
the authors are talking about. It is not clear what the authors mean
by "conventional radar"; is it a SISO radar or a SIMO radar? Was

IEEE A&E SYSTEMS MAGAZINE

Aerospace and Electronic Systems - October 2018

Table of Contents for the Digital Edition of Aerospace and Electronic Systems - October 2018