Maximum likelihood estimation of DOD and DOA for bistatic
MIMO radar
Bo Tang
a,
n
, Jun Tang
b
, Yu Zhang
a
, Zhidong Zheng
a
a
Electronic Engineering Institute, Hefei 230037, Anhui Province, China
b
Department of Electronic Engineering, Tsinghua University, Beijing 100084, China
article info
Article history:
Received 24 February 2012
Received in revised form
7 November 2012
Accepted 12 November 2012
Available online 20 November 2012
Keywords:
Bistatic multiple input multiple output
(MIMO) radar
Direction finding
Maximum likelihood estimation
Alternating projection
abstract
In this paper, the maximum likelihood estimation (MLE) of the direction of departure
(DOD) and direction of arrival (DOA) of multiple targets for bistatic multiple input
multiple output (MIMO) radar is addressed. We derive the maximum likelihood
estimator of the DOD and DOA with the assumption that the targets are unknown but
deterministic. Moreover, we provide a compact expression of the Cramer Rao bound
(CRB) under this nonrandom framework. Since the MLE of the target DOD and DOA is
related to a high-dimensional nonlinear optimization problem, we propose alternating
projection (AP) to solve it efficiently. Numerical simulations demonstrate that the AP
based MLE can provide accurate estimations of the target DOD and DOA and achieves the
CRB in the asymptotic region. Furthermore, results also show that the proposed
algorithm outperforms the existing ESPRIT and MUSIC algorithm for the uniform linear
array (ULA) configuration of the transmitted and re ceived array.
& 2012 Elsevier B.V. All rights reserved.
1. Introduction
Multiple input multiple output (MIMO) radar has
become a hot topic which attracts great interests in recent
years (see, e.g., [1–3] and the references therein).
Compared with traditional phased array radar, which trans-
mits multiple coherent waveforms, MIMO radar offers
enhanced performance through waveform diversity and
careful radar configurations [4–7]. Currently, MIMO radar
can be classified into several types, the first of which is
called statistical MIMO radar, i.e., MIMO radar with widely
separated antennas. Since different aspects of the target can
be viewed simultaneously by statistical MIMO radar, diver-
sity gain and high resolution capabilities can be obtained
[8,9]. The second type is the monostatic colocated MIMO
radar in which the transmitters and receivers are close
enough so that the direction of departure (DOD) of the
target and the corresponding direction of arrival (DOA) are
identical. In [5,6,10], several advantages of monostatic
colocated MIMO radar, including better parameter identifia-
bility, lower-velocity target detection and more flexible
transmit beampattern design, have been demonstrated.
Slightly different from monostatic colocated MIMO radar,
the transmitters and receivers of bistatic MIMO radar are
separated away while the transmit elements as well as
receive elements are all colocated, respectively. Therefore,
the DOD and DOA of a given target are distinct.
In bistatic MIMO radar, the estimation of target DOD
and DOA is of much importance and has been extensively
discussed recently (see, e.g., [11–17] and the references
therein). In [11], a two-dimensional (2D) Capon method
was used to estimate the target DOA and DOD. However,
it is required to perform exhaustive 2D search to obtain
high resolution angle estimations. In [12], estimation of
signal parameters via rotational invariance technique
(ESPRIT) was applied to the joint estimation of target
DOD and DOA, in which both transmit and receive array
are assumed to be uniform linear arrays (ULA). Lately, the
ESPRIT algorithm proposed in [12] was further improved
in [13] where the DOD and DOA are automatically paired.
In [14], close form solution for DOD and DOA estimation
with ESPRIT was derived, in which spatially colored noise
Contents lists available at SciVerse ScienceDirect
journal home page: www.elsevier.com/locate/sigpro
Signal Proce ssing
0165-1684/$ - see front matter & 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.sigpro.2012.11.011
n
Corresponding author. Tel.: þ 86 055165927608.
E-mail address: tangbo06@gmail.com (B. Tang).
Signal Processing 93 (2013) 1349–1357