R. V.

College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 1 M.Tech in Communication Systems

Chapter 2
Literature Review
The term RADAR was originally an acronym for RAdio Detection And Ranging
coined in 1941, and has become a standard English word today. The basic principle of a radar
system is to transmit an electro-magnetic (EM) signal into space and receive echo signals
reflected by targets, which are carefully processed to provide information about them. Early
radar was developed at a rapid pace driven by military demands, but after World War II,
radar has been used for diverse civilian purposes as well. The wide range of radar
applications and developments in the signal processing domain stimulate radar researchers to
design and implement more and more sophisticated radar systems in order to meet the
increasing accuracy requirements.
In recent years, a concept termed multiple-input multiple-output (MIMO) radar has
been attracting increasing attention, which was motivated by intensive research on MIMO
wireless communications since 1990s. This is not surprising due to the fundamental
similarity between communication systems and radar in that they both utilize antennas to
transmit and receive EM signals. There is no explicit definition of MIMO radar as many
different research groups have tackled MIMO radar problems from various perspectives.
MIMO radar can be broadly defined as a radar system deploying multiple antennas to
simultaneously transmit arbitrary waveforms and utilizing multiple antennas to receive
signals which are then processed jointly.
Researchers in this field have demonstrated considerable potential gains from MIMO
radar in scintillation mitigation, resolution enhancement, and interference suppression, etc.
By fully exploiting these benefits, MIMO radar is capable of significantly improving target
detection, parameter estimation, tracking and recognition performance compared with its
conventional phased-array counterpart [3].
In conventional monostatic radar, range resolution can be significantly improved by
using very short pulses. However, this results in a decrease in the transmitted power, leading
to a decrease in receiver signal to noise ratio (SNR). In order to increase the pulse-width
R. V. College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 2 M.Tech in Communication Systems

while maintaining the same range resolution, many pulse compression techniques have been
employed extensively in the past [2]. Although there are analog pulse compression
techniques, digital pulse compression is more popularly used. These include binary phase
codes, polyphase codes and frequency codes [1]. The performance criteria of a pulse
compression sequence are judged by their autocorrelation properties.
With the rapid development of multiple input and multiple output (MIMO) technique
in communication, MIMO radars and multistatic radars are becoming popular due to their
ability to image a target from a variety of angles; thus improving the information received
from the target and capability of modern radars in terms of detection, identification and
classification of targets under surveillance [4] [5]. The orthogonal waveforms used by the
MIMO radar systems must be carefully designed to avoid the self-interference and detection
confusion. For high range resolution and multiple target resolution, the aperiodic
autocorrelation function of sequences should have a low of peak sidelobe level. MIMO radar
also requires orthogonal pulse transmitted waveforms with low cross-correlation between
them. Hence, the successful design of orthogonal code sets with low autocorrelation and
cross-correlation is crucial for implementing MIMO radar systems.
Motivated by breakthroughs in communications theory, one important notion of
MIMO radar is proposed by Fishler in [7] that the system performance can be dramatically
enhanced by taking advantage of the spatial diversity gain provided by the array-target
configuration. More specifically, such a gain is determined by the inter element spacing of
the antenna array, the size of the target, and the distance between the antenna array and the
target. Spatial diversity gain is one of the two major benefits realized by MIMO
communication systems, and it is often achieved by transmitting the same signal through
different sub-channels and combining the information at the receiver.
Diversity gain is used to combat channel fading and thus enhance the link reliability
of the system [8, 9]. The same idea can be exploited in radar scenarios. It has long been
understood that common radar targets are complex bodies, and large scintillations in the
amount of energy reflected by a complex target can occur with only very small changes in
the illuminating direction. The antennas of MIMO radar are widely separated such that
different antennas observe different aspects of the target, and the target returns resulting from
independent illuminations are combined together leading to a spatial diversity gain. This is
R. V. College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 3 M.Tech in Communication Systems

similar to that obtained in the communication systems when the data is transmitted through
independent channels. The underlying idea of diversity gain in MIMO radar is that any
individual view of the target might have a small return with a significant probability, but by
increasing the number of look directions, the probability that all directions have small returns
can be very low. The target deep fading or scintillation problem, which severely degrades the
performance of conventional radars whose antennas are closely spaced, is overcome by
taking advantage of spatial diversity gain, leading to a significantly enhanced target detection
and parameter estimation performance of MIMO radar.
Moreover, for a radar system with a specified configuration, it is obvious that
waveform design is a crucial problem as the choices of transmitted signals affect system
performance considerably.
The waveform design problem based on optimization of the ambiguity function in the
traditional SIMO radar has been well studied. Several waveform design methods have been
proposed to meet different resolution requirements. These methods can be found in [6] and
the references therein. In the traditional SIMO radar system, the radar receiver uses a
matched filter to extract the target signal from thermal noise. Consequently, the resolution of
the radar system is determined by the response to a point target in the matched filter output.
Such a response can be characterized by a function called the ambiguity function [6].
2.1MIMO radar model
Consider a monostatic MIMO radar that contains M transmitters and N receivers with
their antennas configured as uniform linear arrays, as shown in Fig. 2.1. We assume a point
target and also that the target, transmitters and receivers lie in the same 2-D plane. Let d
T
and
d
R
represent the spacing between consecutive transmitters and receivers respectively, and let
= d
T
/d
R
. Then spatial frequency of the target is defined as
f

P
(2.1)
where is the target angle with respect to the broadside direction and is the wavelength of
the RF carrier of the transmitted waveforms. Let and Y be the target delay (which is a
measure of target range) and Doppler frequency (a measure of target velocity), respectively.
R. V. College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 4 M.Tech in Communication Systems






Figure 2.1: Transmitters and receivers in a MIMO radar
The spatial frequency f corresponds to the angular location of the target with respect
to the arrays of the radar. Let {u
m
(t)}, m {0, 1 , M 1} represent the M
transmitter waveforms. Then, the waveform received at the n
th
receiver antenna can be
expressed as [10].

(2.2)


R. V. College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 5 M.Tech in Communication Systems

2.2 MIMO ambiguity function
The resolution of a radar system is determined by the response to a point target in the
matched filter output. This response can be characterized by a function called the ambiguity
function. The ambiguity function (AF) represents the time response of a filter matched to a
given finite energy signal when the signal is received with a delay and a Doppler shift Y
relative to the nominal values (zeros) expected by the filter. The traditional Woodward
ambiguity function for a SIMO radar is given as
utu

t e

ut

(2.3)

In the above expression, and Y represent the delay and Doppler mismatch at the receiver
respectively. The ideal ambiguity function should be sharp around the region of zero-
mismatch, i.e. (, Y) = (0, 0). This idea has been extended to the MIMO case in [10].
Consider the expression for the received signal in a MIMO radar given in (2.2). Let (
1
, Y
1
,
f
1
) represent the true parameters of a target, and let (
2
, Y
2
, f
2
) be the assumed parameters at
the receiver. The summed match filter output is given as

(2.4)

In the above expression, the second term corresponds to the ambiguity function when
only one receiver is present, while the first term brings out the effect of having multiple
receivers. To simplify the ambiguity function and the waveform design problem, the first
term can be decoupled from the above expression. The resulting expression is termed the
MIMO ambiguity function. We can now consider (
1

2
) to be the delay mismatch and
(Y
1
Y
2
) to be the Doppler mismatch, and rewrite the expression.
R. V. College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 6 M.Tech in Communication Systems

The MIMO radar ambiguity function is thus given as [10]
f f

(2.5)

where and Y represent the Doppler and delay mismatch at the receiver, f represents the
targets true spatial frequency, and f represents the assumed spatial frequency at the receiver,
and
m,m
(, Y) represents the cross-ambiguity function between the waveforms u
m
(t) and
u
m
(t).

tu

t e

ut

... (2.6)
The MIMO radar ambiguity function cannot be expressed as a function of the difference of
the spatial frequencies, namely f-f. Therefore, we need both the target spatial frequency f
and the assumed spatial frequency f to represent the spatial mismatch. The value | (0, 0, f,
f)| represents the matched filter output without mismatch. Therefore, the sharper the function
| (, Y, f, f)| around the line {(0, 0, f, f)}, the better the radar system resolution.
2.3 Fitness Function
A fitness function is a particular type of objective function that quantifies the
optimality of a solution (that is, a chromosome) in a genetic algorithm so that that particular
chromosome may be ranked against all the other chromosomes. Optimal chromosomes, or at
least chromosomes which are more optimal, are allowed to breed and mix their datasets by
any of several techniques, producing a new generation that will (hopefully) be even better
than the previous one. An ideal fitness function correlates closely with the algorithm's goal,
and yet may be computed quickly. Speed of execution is very important, as a typical genetic
algorithm must be iterated many, many times in order to produce a usable result for a
problem and this is usually one of the main drawbacks of GAs in real world applications.
This limits their applicability in certain industries. Definition of the fitness function is not
straightforward in many cases and often is performed iteratively if the fittest solutions
produced by the GA are not as good as desired. In some cases, it is very hard or impossible to
come up even with a guess of what the fitness function definition ought to be [14] [26].
R. V. College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 7 M.Tech in Communication Systems

The total cross correlation function and off-peak autocorrelation function must be
minimized to achieve desired correlation properties. In our design, the cost function (which is
to be minimized) is chosen as the sum of the total autocorrelation side lobe energy for each
waveform and totals all possible crosscorrelation.
Thus we boil down to a cost function (E) that is the sum of the Autocorrelation side
lobe peak and energy (Autocorrelation side lobe peak squared) and cross correlation peak
and corresponding energy (cross correlation peak squared).
E
max A k

+ max A

2
max C k

max C

2

(2.10)

Where E is the cost function, the 1
st
term is the sum of Autocorrelation side lobe
peaks of the signals, the 2
nd
term is the sum of Autocorrelation side lobe energies, the 3
rd
is
the sum of cross correlation peaks and the 4
th
term is the sum of cross correlation energies.
The solution set consists of a group of 4 waveforms thus the 1
st
two sum term contain 4 terms
each and the last two contain
4
C
2
i.e. 6 terms each.
Above equation is that of the cost function, and we want its value for our solution to
be as small as possible, whereas the fitness function is defined as the inverse of cost function
scaled by some value. Hence it is desirable to have larger fitness function values. The larger
the value the better solution it is.
Thus, the fitness function needs to be optimized in order to get orthogonal polyphase
code sets that may be used in MIMO systems.
The MIMO radar waveform design problems have been studied in [11, 12, 13, 14,
15]. These methods can be broken into three categories: (1) covariance matrix based design
[16, 17, 18], (2) radar ambiguity function based design [11, 13, 14, 15], and (3) extended
target based design [12, 15].
R. V. College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 8 M.Tech in Communication Systems

In the covariance matrix based design methods, the covariance matrix of the
waveforms are considered instead of the entire waveform. Consequently, this kind of design
methods affects only the spatial domain. In [16], the covariance matrix of the transmitted
waveforms is designed such that the power can be transmitted to a desired range of angles. In
[17], the authors have also designed the covariance matrix of the transmitted waveforms to
control the spatial power. However, in [17], the cross-correlation between the transmitted
signals at a number of given target locations is minimized. This can further increase the
spatial resolution in the receiver. The covariances between waveforms have been optimized
for several design criteria based on the Cramer-Rao bound matrix. In [19], given the
optimized covariance matrix, the corresponding signal waveforms are designed to further
achieve low PAR (peak-to-average-power ratio) and higher range resolution.
The radar ambiguity function based methods optimize the entire waveforms instead
of just their covariances. Thus these design methods involve not only the spatial domain but
also the range domain. The angle-Doppler-range resolution of the radar system can be
characterized by the MIMO radar ambiguity function [20, 21, 22]. In [13, 14], the sidelobe of
the autocorrelation and the cross correlation between waveforms are minimized. This
sharpens the radar ambiguity function. In [11], the waveforms are directly optimized so that a
sharper radar ambiguity function can be obtained. Thus the spatial and range resolution of
point targets can be improved.
In the extended target based methods also, the entire waveform is considered as in the
radar ambiguity function based approaches. However, unlike the ambiguity function based
methods which consider the resolutions of point targets, these methods consider the detection
or estimation of extended targets. These methods require some prior information about the
target and/or clutter impulse response. The extended target based methods have been also
studied in the SIMO case. In [23], the waveform is optimized to maximize the SINR subject
to the constraint that the waveform is similar to a desired waveform. This constraint
improves the PAR, and the range resolution of the waveform. In [24], the optimal radar code
which considers detection probability, Doppler frequency estimation accuracy, PAR and the
range resolution is proposed.
In traditional (SIMO) radars, the ambiguity function of the transmitted pulse
characterizes the compromise between range and Doppler resolutions. It is a major tool for
R. V. College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 9 M.Tech in Communication Systems

studying and analyzing radar signals. Recently, the idea of ambiguity function has been
extended to the case of MIMO radar.
In [16], the covariance matrix of the transmitted waveforms has been designed to
form a focused beam such that the power can be transmitted to a desired range of angles. In
[17], the authors have also focused on the design of the covariance matrix to control the
spatial power. However in [17], the cross-correlation between the transmitted signals at a
number of given target locations is minimized. In [12, 15], unlike [17], the entire waveforms
have been considered instead of just the covariance matrix. Consequently these design
methods involve not only the spatial domain but also the range domain. These methods
assume some prior knowledge of the impulse response of the target and use this knowledge
to choose the waveforms which optimize the mutual information between the received
signals and the impulse response of the target. The waveform design which uses prior
knowledge about the target has been done in the traditional SIMO radar system.
Deng has proposed a set of orthogonal polyphase sequences for use with netted radars
which use orthogonal signals and multiple matched filters at each radar station to form
orthogonal netted radar systems (ONRS) and can operate both monostatic and multistatic
modes [25]. The sequences have desired properties of aperiodic autocorrelation and cross-
correlation which are generated based on a hybrid algorithm integrating statistical simulated
annealing algorithm and traditional iterative code selection method. In that work Doppler
properties have not been taken into consideration resulting in sequences which suffer from
severe degradation, even at small Doppler shifts. Liu have proposed an algorithm to design
and optimize orthogonal signals which is based on a statistical genetic algorithm [14].
Singh have also investigated the similar topic and proposed a modified simulated
annealing algorithm which is a combination of simulated annealing and Hamming scan
algorithm to design polyphase coded signals for ONRS [26]. However, their investigations
have not placed emphasis on Doppler resilience and Doppler properties of those sequences
are similar to Deng sequences. Doppler problems have been mentioned by Khan [27] who
have used Hadamard matrix construction and cross entropy (CE) technique to optimize
polyphase sequences which maintain Doppler tolerance. But the Doppler tolerance levels are
not good enough and can be further improved.
R. V. College of Engineering Bangalore 560059

Dept of Electronics and Communications 20112012 10 M.Tech in Communication Systems

Hongping Hu and Bo Liu has proposed an effective approach based on genetic
algorithm (GA) and a good evaluation criterion was presented for the design of orthogonal
ployphase code sets that can be used in MIMO radars and ONRS radars [28]. Because the
main parts of polyphase code optimization methods are evaluation criterion of the optimal
polyphase code set and automatic search algorithm. So a valid evaluation criterion was
proposed to select the optimal polyphase code set from the code sets which shows feasible
and effective simulation results.
San Antonio [22] have extended the radar ambiguity function to the MIMO radar
case. It turns out that the radar waveforms affect not only the range and Doppler resolution
but also the angular resolution. It is well-known that the radar ambiguity function satisfies
some properties such as constant energy and symmetry with respect to the origin [6]. These
properties are very handy tools for designing and analyzing the radar waveforms.
The MIMO radar ambiguity function characterizes the resolutions of the radar
system. By choosing different waveforms, we obtain a different MIMO ambiguity function.
Therefore the MIMO radar waveform design problem is to choose a set of waveforms which
provides a desirable MIMO ambiguity function. Directly optimizing the waveforms requires
techniques such as calculus of variation. In general this can be very hard to solve.