Aerospace and Electronic Systems Magazine January 2018 - 34

Feature Article:

DOI. No. 10.1109/MAES.2017.160143

Focus-Before-Detection Radar Signal Processing:
Part II-Recent Developments
Jia Xu, Beijing Institute of Technology, Beijing, China
Ying-Ning Peng, Tsinghua University, Beijing, China
Xiang-Gen Xia, Beijing Institute of Technology, Beijing, China and University of
Delaware, Newark, DE, USA
Teng Long, Er-Ke Mao, Beijing Institute of Technology, Beijing, China
Alfonso Farina, SELEX-Sistemi Integrati (Retired), Rome, Italy

INTRODUCTION
Due to high-speed, highly maneuvering, and weak targets as well
as strong clutter and jamming, it is known that the environment of
modern radar becomes increasingly challenging [1-7]. To deal with
the challenges, focus-before-detection (FBD) [8-15] has been introduced in the first article [1] of this two-article series. As some of the
typical FBD methods, Radon-Fourier transform (RFT) and generalized RFT (GRFT) originally proposed in [8, 9] have briefly introduced in [1] for coherent integration of a long time on target (TOT)
to focus the energy of a target in parameter space with arbitrary parametric motion. That is, the motion is modeled with a finite number of
translational and rotational motion parameters, such as velocity, acceleration, and also jerk. The FBD methods can overcome the effects
of across range cells (ARC), across Doppler cells (ADC), and even
across beamwidths (ABW) in a long TOT. As a result, by combining accurate environment sensing from echoes and effective resource
management for optimization, the proposed FBD-based methods can
effectively improve target detection and parameter estimation without changing system parameters in a complicated environment.
Nevertheless, there are many related problems that should be
dealt with before effectively applying the proposed FBD methods.
First, the proposed RFT and GRFT may be regarded as the optimization in a multidimensional space, which is normally compuAuthors' current addresses: J. Xu (deceased), T. Long, E.-K.
Mao, School of Information and Electronics, Beijing Institute
of Technology, Beijing 100081, China; Y.-N. Peng, Department of Electronic Engineering, Tsinghua University, Beijing
100084, China; X.-G. Xia, School of Information and Electronics, Beijing Institute of Technology, Beijing 100081, China, and
Department of Electrical and Computer Engineering, University of Delaware, Newark, DE 19716, USA; A. Farina (retired),
SELEX-Sistemi Integrati, 00131 Rome, Italy.
Corresponding author is X.-G. Xia, E-mail: (xianggen@udel.edu).
Manuscript received June 24, 2016, revised October 27, 2016,
and ready for publication January 5, 2017.
Review handled by D. O'Hagan.
0885/8985/18/$26.00 © 2018 IEEE
34

tationally expensive, so the fast implementation of an FBD-based
method is urgently desired [14-24]. Second, in the transformed
parameter space some unwanted responses like blind speed sidelobes (BSSL) will be generated by RFT/GRFT, which should
be effectively suppressed. Third, the echoes of targets contain the
energy in multidomains, such as space, frequency, polarization,
and waveforms, and effective methods are needed to realize the
coherent integration of echoes in multiple domains [25]. Fourth,
real targets are moving in a three-dimensional space while the conventional FBD is only defined in the range direction as RFT or
GRFT [26]. Therefore, it is much more attractive to extend FBD
methods to high dimensions, such as range-azimuth or range-azimuth-elevation, which can retrieve the target's multidimensional
location and motion information. Fifth, high resolution is always
desired for dense target scenarios, such as the well-known multiple
aircraft formation (MAF) [27]. The resolution on the spatial and
motion parameters should be discussed for the proposed FBD in
depth. Sixth, the performance of the proposed FBD-based methods
in the complicated environment with strong clutter and jamming
should be discussed [28-31]. Seventh, much more effective methods should be proposed for maneuvering targets with complicated
parametric motion like micromotions [37, 38]. Eighth, the scale effect of an ultrahigh-speed target [39] will cause the signal-to-noise
ratio (SNR) gain loss and effective compensation methods should
be proposed. Ninth, efficient nonparametric processing should be
introduced to speed up the implementation of the FBD methods.
For solving the above real problems, nine recent research developments are discussed in this article for the FBD methods, namely
fast implementations, BSSL suppression, multiple domains, multiple dimensions, multiple targets, background cancellation, micromotion application, ultrahigh-speed target detection, and nonparametric processing. Furthermore, some application limitations
and research aspects are discussed for the proposed FBD-based
methods, such as, the limitations on the radar line-of-sight (LOS)
and carrier frequency difference, the optimal hybrid integration
structure, and the exploration of multidimensional high-resolution
information. The contributions of this article are not only to demonstrate the effectiveness of the existing FBD methods reported in

IEEE A&E SYSTEMS MAGAZINE

JANUARY 2018

Table of Contents for the Digital Edition of Aerospace and Electronic Systems Magazine January 2018