THE USE OF STEPPED-FREQUENCY CONTINUOUS WAVE (SFCW) RADARS to detect moving people and vehicles

WHAT IS A SFCW AND WHY DO WE USE IT?

A stepped frequency continuous wave (SFCW) consists of N equidistant discrete frequencies within a given bandwidth B.  So the frequency graph vs. time will consist of a step diagram, where each frequency is a continuous wave. Figure 1-(a) and (b) show the stepped frequency diagram and the corresponding continuous waves, respectively. A color-coded notation is used to indicate the discrete frequencies from the lower frequency fStart to the higher one fStop. Frequency is changed from pulse to pulse in uniform frequency steps over a burst of frequencies around the central frequency f0. Figure 1-(c) shows how the sum of the signal reflected by the target at the different frequencies allows to determine the range position of the target.

The SFCW is a radar technology that uses the frequency diversity to detect a target based on its answer to the microwave signals for each discrete frequency. The range resolution of a target is inversely proportional to the bandwidth while the maximum unambiguous range distance within which targets can be located depends on the number of frequency steps. For instance, a SFCW radar using 2000 frequency steps within a bandwidth of 200 MHz provides a maximum detectable range area of 1.5 km with a range resolution of 75 cm.

The magnitude of microwave signals scattered by the target is proportional to its radar cross-section (RCS), a measure of its “microwave reflectivity” depending on its size and dielectric properties.  In many cases the RCS is normalizes to a reference target in the scene so resulting so providing the normalized radar cross-section (NRCS).

Figure 1: Stepped Frequency Continuous Wave (SFCW) radar: (a) step diagram of discrete frequencies; (b) corresponding continuous waves for each discrete frequency; (c) range resolution of a target when imaged with the set of discrete frequencies.

The use of stepped-frequency continuous wave is useful and practical because its technology is very simple and inexpensive. In fact, we are able to get a good range resolution at a really effective cost.

HOW WILL WE USE IN THE BorderUAS Project?

The SFCW technique will be used in the BorderUAS Project to implement a MIMO (Multiple-input multiple-output) radar. MIMO radar is a linear array of transmitting (TX) and receiving (RX) antennas. The observation geometry diversity provided by the array ensures to distinguish targets with a good cross-range resolution that depends on the length of the array and is usually poorer than the range resolution.

The geometry diversity of the array coupled to the frequency diversity of the SFCW provides a means to acquire microwave images of the scene in any weather and sun-illumination conditions. The main advantage of the SFCW MIMO radar is its capability to collect SAR images whatever the movement of platform on which it is installed, also in case of persistent acquisitions when the platform is not moving.

WHICH INFORMATION WILL IT PROVIDE?

The SFCW MIMO radar will be used to detect moving targets, such as group of people and vehicles, using microwave images. To this end the change detection technique will be used. This technique consists in comparing two or more georeferenced radar datasets of the same scene, acquired at different times, in order to identify changes of the NRCS. The basic idea is the following. If the observed scene is stable in time, the targets of the scene should be characterized by a stable NRCS within the expected statistical variation. However, if a target of the scene is moving, i.e. vehicles or group of people, the NRCS values in radar datasets will change in time based on the actual position of the moving target. So the moving target will be detected in radar data as it will increase the NRCS of the stable background.

As an example, Figures shows three temporal profiles of the NRCS referred to three “background” targets located at the range distances R1 = 101,25 m, R2 = 108,75 m and R3 = 116,26 m. A moving vehicle appeared in the scene observed by the radar at the time t1 = 46 s and moved from R1 to R3. The position of the vehicle within the temporal window is indicated by a downward-pointing triangle. When the vehicle passes through a specific range position this affect the NRCS of that background target as the dielectric properties at that range position are changed due to the presence of the car. Usually this causes a peak as the car is a metallic object having a larger NRCS with respect to a natural scene. The analysis of a stack of radar dataset will help to track in time the movement of the target.

Figure 2: Example of NRCS change in radar dataset due to the movement of a target. From the top to the bottom the time profiles of the NRCS are plotted for the background scene at three different range distances. The peak of NRCS of the moving vehicle is indicated with the red triangle.
Tags: