High Resolution Software Defined Radar System for Target Detection

Transcription

1 High Resolution Software Defined Radar System for Target Detection Sandra Costanzo, Francesco Spadafora, Antonio Borgia, Oswaldo Hugo Moreno, Antonio Costanzo, and Giuseppe Di Massa DIMES University of Calabria Rende (CS), Italy Abstract. A high resolution Software Defined Radar system is implemented in this work by adopting the new generation Universal Software Radio Peripheral USRP NI2920, a software defined transceiver. The enhanced available bandwidth due to the Gigabit Ethernet interface is exploited to achieve the high range resolution features. At this purpose, a specific Labview application implementing the radar operations is developed. The realized SDRadar system is successfully validated by preliminary outdoor tests accurately retrieving the distance of a reference target. Keywords: Software Defined Radio, Radar, Slant Range Resolution. 1 Introduction The flexibility of software based systems and their easy adaptability make them useful for many different applications. The Software Defined Radar (SDRadar) system is a special type of versatile radar in which operations and components, typically realized by specific hardware (i.e., mixers, filters, modulators and demodulators), are implemented in terms of software modules [1]. To implement a SDRadar, some recent researches and studies [2], [3] were conducted through the use of FPGA and/or DSP. The Universal Software Radio Peripheral (USRP) transceiver can be used to develop Software Defined Radio applications like SDRadar, thus leading to obtain a low cost radar sensor. A first attempt to adopt USRP for radar applications was performed by the authors in [4], where a SDRadar system was implemented through the adoption of first generation USRP. Due to the bandwidth limitations imposed by the available USB connection, the solution presented in [4] gives a limited slant-range resolution equal to 75 m, so alternative solutions have been investigated to enhance the radar performance. Other excellent results have been conducted in [5], [6], [7], [8], where the characterization of the USRP N200 e N210 in radar field has been considered. In particular, National Instruments (NI) has recently manufactured a new generation of USRP for wireless communications teaching and research. It successfully combines the NI LabVIEW software and the USRP hardware to deliver an affordable and easy-to-use software-reconfigurable RF platform that works well Á. Rocha et al. (Eds.): Advances in Information Systems and Technologies, AISC 206, pp DOI: / _94 Springer-Verlag Berlin Heidelberg 2013

2 998 S. Costanzo et al. for communications, education, experimentation, research, and rapid prototyping [9]. In this paper, the potentiality of the NI new generation USRP is exploited to enhance the radar resolution of the first SDRadar prototype proposed in [4]. A specific LabVIEW code is developed to control the SDRadar system, with the implementation of a signal processing compression-based technique to achieve a strongly enhanced slant-range resolution equal to 6 m. In the following sections, a complete description of the hardware and the relative control algorithm is provided. Furthermore, experimental results obtained by outdoor tests are discussed to prove the enhanced radar resolution. 2 USRP NI2920 Hardware Description The first USRP motherboard was designed by Matt Ettus at the National Science Foundation in Nowadays, four versions are available, namely USRP, USRP2, USRP N200 and USRP N210. In the last year, the National Instruments has realized three new boards, namely USRP 2920, 2921, 2922, interfacing with the PC through Labview software. The USRP 292X main features are as follows: 2 channels ADC, 400MS/s; 2 channels, 100MS/s; 1 GIGABIT ETHERNET for PC interface; Xilinx Spartan-6; 25MHz of operating bandwidth. The block scheme is shown in Figure 1. Fig. 1. USRP 2920 block diagram

3 High Resolution Software Defined Radar System for Target Detection 999 Incoming signals attached to the standard SMA connector are mixed down from RF using a direct-conversion receiver (DCR) to baseband I/Q components, which are sampled by a 2-channel, 100 MS/s, 14-bit analog-to-digital converter (ADC). The digitized I/Q data follows parallel paths through a digital down-conversion (DDC) process that mixes, filters, and decimates the input 100 MS/s signal to a user-specified rate. The down-converted samples are passed to the host computer up to 20 MS/s over a standard Gigabit Ethernet connection. For transmission, baseband I/Q signal samples are synthesized by the host computer and fed to a USRP-292x up to 20 MS/s over Gigabit Ethernet. The USRP hardware interpolates the incoming signal to 100 MS/s using a digital up-conversion (DUC) process and then converts the signal to analog with a dual-channel, 16-bit digital-to-analog converter (DAC). The resulting analog signal is then mixed up to the specified RF frequency [9]. The main limitation of the SDRadar technology is due to the interface with the PC, that reduces the radar performance in terms of slant range resolution [10]. The first generation USRP, by Matt Ettus, uses a USB 2.0 interface to connect to the PC, thus imposing the adoption of the low USB band for data transmission, which leads to very low slant range resolutions. The behavior and the analysis of the first generation USRP in radar field was conducted by the authors in a recent work [4], where a SDRadar system was implemented with a slant range resolution equal to 75 m. However, radar applications typically require more refined precisions for target detection, so alternative solutions are investigated to enhance the system bandwidth and thus the SDRadar resolution. In particular, the adoption of the new USRP NI 2920 is considered in this work to exploit the associated Gigabit Ethernet interface in order to improve the SDRadar capabilities. 2.1 Signal Processing Algorithm In order to demonstrate the range resolution improvement using the USRP 2920, a signal processing technique, called Stretch Processor [10], is implemented in Labview code. This processing is a particular pulse compression technique which consists of four distinct steps. First, the radar returns are mixed with a replica (reference signal) of the transmitted waveform. This is followed by Low Pass Filtering (LPF) and coherent detection in order to avoid the high frequency response achieved at the output of the Mixer (see figure 2). Next, Analog to Digital (A/D) conversion is performed, and finally a bank of Narrow Band Filters (NBFs) is used to extract the tones proportional to the target range, since stretch processing effectively converts time delay into frequency. A block diagram for a stretch processing receiver [10] is illustrated in Figure 2. The transmitted signal is an Linear Frequency Modulated (LFM) waveform expressed by the following equation: 2, 0 (1)

4 1000 S. Costanzo et al. where μ / is the LFM coefficient, B gives the bandwidth, is the chirp start frequency and is the chirp duration. The slant range resolution ΔR is given by: (2) On the basis of the above equation, the use of the USRP 2920 NI, giving a maximum available bandwidth B = 25 MHz, leads to have a slant range resolution equal to 6 m, which is significantly enhanced with respect to the value of 75 m achieved with the first generation USRP where B is equal to 2 MHz [9]. Fig. 2. Stretch processing block diagram 3 3 SDRadar System The idea is to implement a SDRadar system able to scan a complete area under analysis and to locate, through N different radar scannings in different horizontal positions, the surface topology. In Figure 3 is reported the system block diagram through which the USRP 2920 is used to transmit and receive data by two linear array antennas, that are rotated by a controlled motor. The system is interfaced by a Labview window running on a Single Board Computer (SBC) which processes all the transmitted and received data to determine the topology of the area under analysis. This interface is able to control the motion motor too. A Power Amplifier (AMP) and a Low Noise Amplifier (LNA) are connected to the transmitting (TX) and the receiving (RX) antenna to increase the power along both the transmission and the receiving paths.

5 High Resolution Software Defined Radar System for Target Detection 1001 Fig. 3. SDRadar block diagram 3.1 SDRadar Algorithm for Complete Horizontal Scanning The proposed SDRadar leads to scan, N times, different frames of the area under analysis (eg. Mountain, Landslide, Topography surfaces, Glaciers ) through a horizontal movement of the radar antenna controlled by a driver motor. Figure 4 shows the proposed algorithm, summarized in the following step: 1. Parameter Definition Footprint (antenna illuminating area) of each scan defined by the distance between the radar antenna and the analyzed area, the azimuth and the elevation antenna beam widths, the grazing angle and the operating frequency. The receiving window, that ensures the correct recognition of any type of topology of the surface under analysis, defined by R min and R max (minimum and maximum target range required). Total Area size: which gives the exact number N of radar scanning necessary to retrieve the total topology. 2. An N scan matrix, defined by the parameter of the previous step. This Matrix, made up by N rows, that correspond at N scan produced, and M columns, that depend of the receiving window. 3. A For loop is started for each N scan. The scans are obtained by the motor that rotate the antenna by an angle θ scan N times. For each n<n the Matrix is filled with the results from the Stretch Processor described in the previous section and retrieved from the USRP. 4. When the matrix is completed, a colors assignment is performed like in a radargram [11]. The colors are helpful for the remote view of the topology.

6 1002 S. Costanzo et al. Fig. 4. SDRadar algorithm 4 Outdoor SDRadar Tests Specific tests are performed on the USRP NI2920 in order to identify key features of the device in radar field. At this purpose, the USRP is connected to an host PC through Gigabit Ethernet and it is controlled by an own developed Labview application illustrated in Figure 5. Fig. 5. Labview SDRadar application window To demonstrate the enhanced Slant Range Resolution, an outdoor experimental setup is assessed (Fig. 6), with a broadband ridged horn antenna employed for the transmission and a broadband logarithmic antenna adopted for the reception. A metal plate, positioned at different distances in line of sight direction from the transmitting/receiving platform is assumed as target under test. This preliminary test is performed without motors so the algorithm test described in the previously section was considered for only one scan with θ scan equal to 0 degree. The real and software retrieved target positions are successfully compared in Table 1 for various target distances. The relative signal peaks, properly retrieved by the implemented Stretch Processor technique, are illustrated in Figure 7. As a further validation, the SDRadar map for a single scan is illustrated in Figure 8, where three different targets at m are displayed with different colors.

7 High Resolution Software Defined Radar System for Target Detection 1003 Table 1. Real and retrieved target positions with the USRP 2920 Real target position [m] Retrieved software position [m] Fig. 6. Software Defined Radar Test Fig. 7. Retrieved signal peaks for different target positions

8 1004 S. Costanzo et al N Radar Scanning Effective Distance [m] Fig. 8. SDRadar map for a single scanning 0 5 Conclusions A low cost, flexible, versatile and small dimensioned solution to create a high performance radar system has been proposed in this work. The USRP NI2920 has been adopted to realized a SDRadar system giving a 6 m Slant Range Resolution, significantly enhanced with respect to that achieved in the existing SDRadar solutions. A specific Labview application has been developed to implement the high resolution radar processing algorithm and outdoor experimental validations are performed to demonstrate the theoretical features. Acknowledgments. This work has been carried out under the framework of PON 01_01503 National Italian Project Landslides Early Warning, financed by the Italian Ministry of University and Research. References 1. Debatty, T.: Software Defined RADAR a state of the art. In: Second International Workshop on Cognitive Information Processing, Brussels, Belgium, pp (2010) 2. Zhang, H., Li, L., Wu, K.: 24ghz software-defined radar system for automotive applications. In: European Conference on Wireless Technologies, Munich, Germany, pp (2007) 3. Kauffman, K., Garmatyuk, D., Schuerger, J.: Multifunctional software-defined radar sensor and data communication system. IEEE Sensors Journal 11, (2011) 4. Aloi, G., Borgia, A., Costanzo, S., Di Massa, G., Loscrì, V., Natalizio, E., Pace, P., Spadafora, F.: Software Defined Radar: synchronization issues and practical implementation. In: COGART, International Conference on Cognitive Radio and Advanced Spectrum Management, Barcelona (2011) 5. Prathyusha, C., Sowmiya, S.N., Ramanathan, S., Soman, R., Amrita, K.P., Deepthi, V.V., Chinnam, M., Nandhini, J.: Implementation of a low cost synthetic aperture radar using software defined radio. In: International Conference on Computing Communication and Networking Technologies (ICCCNT), Karur, TamilNadu, India, pp. 1 7 (2010)

9 High Resolution Software Defined Radar System for Target Detection Manuel, F., Martin, B., Christian, S., Lars, R., Jondral Friedrich, K.: An SDR-based Experimental Setup for OFDM-based RadaR. In: 7th Karlsruhe Workshop on Software Radio Karlsruhe, Germany (March 2012) 7. Marcus, M., Martin, B., Manuel, F., Jondral Friedrich, K.: A USRP-based Testbed for OFDM-based Radar and Communication Systems. In: 22nd Virginia Tech Symposium on Wireless Communications, Blacksburg (June 2012) 8. Fernandes, V.: Implementation of a RADAR System using MATLAB and the USRP, CSUN ScholarWorks (2012) 9. NI USRP-2920, NI USRP-2921, National Instruments Data-sheet, Mahafza, B.R., Elsherbeni, A.Z.: Simulations for Radar Systems Design. Chapman & Hall /CRC (1999) 11. Skolnik, M.: Radar handbook, 3rd edn., pp Mc Graw Hill, San Francisco (2008)