The concept of noise radar was pioneered more than 50 years ago by Billy Horton, who recognized that range and Doppler ambiguities or uncertainties could be eliminated using random noise as modulation. The ambiguity function—a 2D representation of radar resolution performance in range and Doppler coordinates—is characterized by an ideal thumbtack, i.e., a sharp needlepoint, that enables clear separation of small changes in either parameter. Moreover, range could be determined by cross-correlating the return signal with a time-delayed replica of the transmit waveform.1
Since those early days, narrowband noise radar has been refined. And in recent years, our group has made significant progress in developing ultra-wide-band (UWB) random noise radar.2, 3 UWB noise radar transmits a noise waveform with fractional bandwidth greater than 25%. Because the resolution is inversely proportional to the bandwidth, UWB waveforms achieve good range resolution. Noise radar satisfies important requirements for military systems, such as low probability of intercept and low probability of detection, owing to the featureless characteristics of the transmit waveform. Noise radar also efficiently shares the frequency spectrum. Each noise waveform is uncorrelated with the others, which means that several such radars can operate over the same frequency band.
Figure 1. Block diagram of noise radar using (a) homodyne and (b) heterodyne correlation. LSB: Lower sideband.
Figure 1(a) shows a typical block diagram of noise radar using the standard homodyne correlation approach. This correlation technique downconverts the target-reflected signal directly to DC, and consequently loses important phase information. Phase coherence can be injected into noise radar by cross-correlating the reflected signal with a time-delayed and frequency-offset replica of the transmit waveform: see Figure 1(b). This technique, called heterodyne correlation, preserves the phase of the reflected signal, permitting Doppler extraction, interferometry, polarimetry, angle tracking, and coherent imaging. Because the output of the correlator is always at the offset frequency, we observe that the UWB transmit waveform, despite its wide bandwidth, collapses to a single frequency at the output of the downconverter. Thus, the detection bandwidth can be shrunk at the correlator output, just wide enough to pass any Doppler modulation, to enhance the signal-to-noise ratio.
Figure 2. Single polarization and RGB (red, green, blue) color composite polarimetric inverse synthetic aperture radar images of a mock airplane. H: Horizontal polarization. V: Vertical polarization.
We obtained 2D polarimetric inverse synthetic aperture radar images of a mock airplane using 1–2GHz coherent UWB random noise radar by placing the target on a turntable and collecting data as the turntable rotated. Figure 2 shows (a) the HH polarization, (b) the VV polarization, and (c) the average of the VH and HV polarizations (where V is vertical and H is horizontal). The polarimetric characteristics of the target can be captured by forming a color RGB (red, green, blue) image of the target, which provides useful clues about its strong scattering centers for automatic target recognition applications. Figure 2(d) is an RGB color composite polarimetric image in which red corresponds to the VV return, green to the average of the VH and HV returns, and blue to the HH returns. The wings and fuselage of the airplane, seen as white (since reflections at all polarizations are strong), can be readily recognized. We see the green returns along the edges of the fuselage where the wings meet it, just where one might expect a dihedral type reflection. The outline of the airplane can be reasonably well identified in the color image.
Figure 3. Synthetic aperture radar image of humans behind a wall. (a) Image of one human at 2m range. (b) Image of two humans behind a wall at 1 and 2m range.
Recent work on UWB noise radar by our group includes through-wall detection,4, 5 radar networking,6 multiple-input–multiple-output systems,7 radio frequency tagging,8 millimeter-wave noise radar,9 compressive sensing,10, 11 and noise tomography.12 Much of the system development has been spurred by the advances in radio frequency miniaturization and high-speed sampling, resulting in the development of portable digital noise radar. Figure 3 shows a synthetic aperture radar image of one and two humans inside a room behind walls made of cinder blocks. The room was 4m wide and 6m long. Three side walls are covered with wave absorbers. The antenna set (composed of one transmit antenna and one receive antenna with 0.7m separation) was fixed on the scanner at a height of 1.25m, which is at chest level. The antenna set moved parallel to both the floor and the wall. The total translational distance of the antennas was 1.8m, with a step size of 5cm. We can clearly see the humans despite some ghost images, which can be eliminated by appropriate image thresholding.
In conclusion, tailored noise waveforms can be used to overcome dispersive effects caused by the propagation medium by suitably shaping the waveform spectral characteristics and improving target detectability. Noiselike waveforms, such as orthogonal frequency-division multiplexing, chaotic, and spread-spectrum types, offer new features for enhanced processing yet retain desirable characteristics of randomness. Current noise radar technology is fully mature for short-range applications (less than 1km), and thus ideally suited for covert close-range applications such as through-wall imaging, mine detection, and detection of concealed weapons through clothing. Pulsed noise radar can work for longer-range uses. As next steps, we plan to explore the use of noise waveforms for multifunctional radio frequency systems and multiple target tracking.
The Pennsylvania State University
University Park, PA
Ram Narayanan received his BTech from the Indian Institute of Technology Madras (1976) and his PhD from the University of Massachusetts at Amherst (1988). His current research interests include high-resolution radar systems, tomographic imaging, and advanced waveform design and analysis. He is a Fellow of SPIE, the IEEE, and the Institution of Electronics and Telecommunication Engineers.
1. B. M. Horton, Noise-modulated distance measuring systems, Proc. IRE 47(5), p. 821-828, 1959.
2. R. M. Narayanan, Y. Xu, P. D. Hoffmeyer, J. O. Curtis, Design, performance, and applications of a coherent ultrawideband random noise radar, Opt. Eng. 37(6), p. 1855-1869, 1998.
3. R. M. Narayanan, X. Xu, Principles and applications of coherent random noise radar technology, Proc. SPIE
5113, p. 503-514, 2003. doi:10.1117/12.484912
4. C. P. Lai, R. M. Narayanan, Ultrawideband random noise radar design for through-wall surveillance, IEEE Trans. Aero. Electron. Syst. 46(4), p. 1716-1730, 2010.
5. P. H. Chen, M. C. Shastry, C. P. Lai, R. M. Narayanan, A portable real-time digital noise radar system for through-the-wall imaging, IEEE Trans. Geosci. Remote Sens. 50(10), in press.
6. S. C. Surender, R. M. Narayanan, UWB noise-OFDM netted radar: physical layer design and analysis, IEEE Trans. Aero. Electron. Syst. 47(2), p. 1380-1400, 2011.
7. W. J. Chen, R. M. Narayanan, Comparison of the estimation performance of coherent and non-coherent ambiguity functions for an ultrawideband multi-input–multi-output noise radar, IET Radar Sonar Nav. 6(1), p. 49-59, 2012.
8. Q. Pan, R. M. Narayanan, Delay-modulated RF tag system using ultrawideband noise radar waveforms, Int'l J. Distrib. Sens. Netw.
2011(156582), p. 16, 2011. doi:10.1155/2011/156582
9. K. A. Gallagher, R. M. Narayanan, Simultaneous human detection and ranging using a millimeter-wave radar system transmitting wideband noise with an embedded tone, Proc. SPIE
8361, p. 836119, 2012. doi:10.1117/12.922471
10. M. C. Shastry, R. M. Narayanan, M. Rangaswamy, Compressive radar imaging using white stochastic waveforms, Proc. 5th Int'l Waveform Diversity Design Conf., p. 90-94, 2010.
11. Y. Kwon, R. M. Narayanan, M. Rangaswamy, Sample selection and adaptive weight allocation for compressive MIMO UWB noise radar, Proc. SPIE
8365, p. 83650T, 2012. doi:10.1117/12.920937
12. R. Vela, R. M. Narayanan, K. A. Gallagher, M. Rangaswamy, Noise radar tomography, Proc. IEEE Radar Conf., p. 720-724, 2012.