Rafal Rytel-Andrianik

Optimal detection and estimation in FMCW radar

The paper is mainly concerned with detection of an accelerating target echo by FMCW radar. The processed signal is modeled as a polynomial phase signal, contaminated either by white or colored additive Gaussian noise. Obtained tests, optimal in the Neyman-Pearson sense, are linked with the maximum likelihood estimation procedures dealt with in earlier papers. An example shows the importance of properly suiting the test to the noise statistics.

Micro-range, micro-Doppler joint analysis of pedestrian radar echo

We present in the paper results of a generalised micro-Doppler analysis that is suited to high range resolution radar. The target echo energy is here represented in the three dimensional space of time, Doppler, and additionally range, what can provide more information on target micro-motion than the standard micro-Doppler analysis which is performed in the time-Doppler plane. The analyzed signals were recorded with a Ka-band FMCW radar of a high 1GHz bandwidth, which could provide good resolution in range, time and velocity. Here, we present a new method to visualize jointly micro-range and micro-doppler effects of a running man observed on the background of ground interferences.

Efficient matched filtering and beamforming for coherent MIMO radar

In the paper there are shown four realization structures of matched filtering and beamforming in a coherent MIMO radar with collocated antennas. Particularly, in one of the structures only one matched filtering (per beam and per Doppler frequency) and one receive beamforming (per beam) is necessary, while the MIMO transmit beamforming can be pre computed. This allows to realize detection algorithm for MIMO radar with computational requirements close to that for multi-beam phased array radar.

Simple X-band polarimetric micro-Doppler analyses of ground moving targets

This paper presents the results of polarimetric micro-Doppler analyses using X-band Frequency Modulated Continuous Wave (FMCW) Radar. In July 2014 the measurement campaign was jointly carried out by Metasensing BV and Warsaw University of Technology. It took place nearby Metasensing premises in Noordwijk, Holland. The radar sensor used in the experiments has a two-channel receiver. Radar data was gathered in two modes: co-polarized (HH) and cross-polarized (HV). During the measurements two cooperating targets were used - a cyclist and a running pedestrian. This paper presents results of the analyses of the observed targets. The obtained results indicate the possibility of distinguishing between different types of objects.

Design of MTD filters with maximal average improvement factor

Moving Target Detection (MTD) radar, to detect a target of unknown Doppler frequency, employs a filter bank where each filter corresponds to an assigned part of the unambiguous Doppler frequency range. In this correspondence the optimal MTD filters are designed. The optimality criterion is the maximization of the SNR improvement factor averaged over range of Doppler frequencies assigned to the filter. The noise component of the signal can consist of white receiver noise, clutter echoes and interferences. The obtained result is discussed and a link with the Kaiser window is established.

On circularity of complex-valued radar signals

In radars, complex signals arise naturally as a result of a quadrature demodulation or Fourier transform. Also, the Doppler spectrum is not an even function of frequency and the Doppler signal is thus naturally complex. Random complex radar signals used to be automatically assumed to be circular. As a result, most of signal processing algorithms (such as detection and estimation) were derived for circular noises and interferences only. In the recent literature an interest in complex signals that are not circular can be seen. It was noticed that many common signals, particularly in communications, are not circular and processing methods exploiting this noncircularity were presented. In this paper we review complex random signals and their properties of circularity and properness. Then we analyze a few complex-valued signals that arise in radars and show that some of them are indeed noncircular to some extend.

On the Ambiguity Function for Accelerating Target in FMCW Radar

In the paper we are concerned with the FMCW radar detection of an accelerating target, echo of which is buried in an additive white Gaussian noise. For this problem, we derive and analyze three-dimensional generalized ambiguity function for target range, velocity and acceleration. We interpret known properties of this function and obtain new ones. This allows us, for example, to specify resolution and regions of unambiguous velocity and acceleration.

A versatile signal processor and control system for AESA radar

The paper presents a concept of a versatile signal processor designed for a modern, active electronically scanned array (AESA) radar. The flexibility of the design is high enough to support most advanced modes of operation, including multiple-input multiple-output (MIMO) modes and cognitive radar control, reacting in real time to changing air situations. The required signal processing power calculations are provided together with hardware solutions meeting high computational power and low latency requirements.

Detection and tracking experiments with a fixed FMCW radar in an urban scenario

The paper presents the very first results of an FMCW radar measurements experiment carried out within the framework of the SOS project, the aim of which is to develop a modern sensors system for the detection and tracking of dangerous materials in order to increase airport security in the indoor landside area. The trials were held in 2013 in Noordwijk, the Netherlands. The main goal of the experiment was to test the detection and tracking capabilities of the FMCW radar in an urban environment.