Fast communication
DOA estimation for wideband signals based on sparse signal
reconstruction using prolate spheroidal wave functions
$
Nan Hu
a,b
,XuXu
a,b
, Zhongfu Ye
a,b,
n
a
Department of Electronic Engineering and Information Science, University of Science and Technology of China, Hefei, Anhui 230027,
People's Republic of China
b
National Engineering Laboratory for Speech and Language Information Processing, People's Republic of China
article info
Article history:
Received 27 February 2013
Received in revised form
10 September 2013
Accepted 11 September 2013
Available online 18 October 2013
Keywords:
Direction-of-arrival
Prolate spheroidal wave function
Sparse signal reconstruction
Orthogonal matching pursuit
abstract
The idea of block-sparse signal reconstruction, as an alternative perspective compared
with the conventional approach, is exploited to formulate the problem of direction-of-
arrival (DOA) estimation for wideband signals. Prolate spheroidal wave functions (PSWFs)
are used to form the block-wise bases for this problem, due to its excellent performance in
extrapolating bandlimited signals, and the block orthogonal matching pursuit (BOMP)
algorithm is directly employed to verify its efficiency. Simulation results show that the
proposed method yields better performance when the number of samples is highly
limited.
& 2013 Elsevier B.V. All rights reserved.
1. Introduction
Direction-of-arrival (DOA) estimation for wideband
signals has been a crucial task in recent years. The
conventional subspace-based methods such as incoherent
signal subspace methods (ISSM) [1] and coherent signal
subspace methods (CSSM) [2] process the data which
have passed a bank of narrowband filters (NF), whereas
the NF output is highly correlated when few samples is
obtained. Moreover, to resolve multiple incident signals,
these methods require a large number of samples to obtain
multiple frequency snapshots and hence form the sample
covariance matrices. A subspace-based method without
NF was proposed in [3] for the localization of wideband
signals, while this method still needs several data blocks
for subspace construction and DOA estimation.
Recently, the perspective of sparse signal reconstruc-
tion has been introduced into DOA estimation [4], which
exploits the spatial sparsity of the incident signals, while
in the wideband case the NF approach has to be employed
for pre-processing. Due to the diversity of estimation in
each frequency band, direct adjustment of narrowband
sparse signal reconstruction algorithms for the wideband
case is not feasible, and simple averaging will cause
serious confusion in spatial spectrum. The feature of joint
sparsity among multiple frequency bands was taken
advantage of in [5], where an orthogonal matching pursuit
(OMP) [6] method was employed to solve the multiple-
dictionary joint optimization problem. A sparse signal
reconstruction method without NF pre-processing was
proposed in [7] under the assumption that all the incident
signals are identical, which is, however, not common
in reality. A recent work of DOA estimation for wideband
signals via sparse signal reconstruction was proposed
in [8], where the homotopy approach was employed to
solve the corresponding basis pursuit de-noising (BPDN)
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/sigpro
Signal Processing
0165-1684/$ - see front matter & 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.sigpro.2013.09.018
☆
This work is supported by the National Natural Science Foundation of
China (No. 61101236).
n
Corresponding author at: University of Science and Technology of
China, Department of Electronic Engineering and Information Science,
Huangshan Road, Hefei, Anhui 230027, People's Republic of China.
Tel.: þ86 551 63601314.
E-mail address: yezf@ustc.edu.cn (Z. Ye).
Signal Processing 96 (2014) 395–400