Simulation, test of stepped frequency radar systems

Technology News | October 18, 2012

By eeNews Europe

In any radar receiver, the received echo signals contain the target return and background clutter. Detection of the target in an environment with background clutter requires high range and cross-range resolution in the radar system. The traditional way to accomplish this goal involves use of short duration pulse waveforms and wideband-FM pulses. However, this approach requires a complex system architecture and results in higher implementation cost due to its wideband receiver usage. Another way to achieve high range resolution, without increasing system complexity, is to employ Stepped Frequency Radar (SFR), a scheme well known for its use in non-destructive testing and ground searching applications.

With SFR, the echoes of stepped frequency pulses are synthesized in the frequency domain to obtain wider signal bandwidth. Using frequency hopping, both high resolution and a high signal-to-clutter ratio can be received. Because of its high resolution and low cost it is today widely used in both the commercial and aerospace/defense (A/D) industry. However, it is very difficult to get an analytical solution for SFR receiver performance in the presence of background clutter caused by reflections from ground, structures, vegetation, and so on. As a result, simulation becomes more important. Using it to accurately design, verify and test SFR systems under real-world environments has become absolutely essential.

Understanding SFR
To better understand why SFR is so advantageous, first consider the pulse radar waveform shown in Figure 1 (left-most image).

Figure 1: On the left is a pulse radar waveform. The right-most image depicts a SFR waveform.

Assuming the pulse width is τ and the bandwidth of the signal is ƒ₀= 1/ τ, the range resolution, Rs, can be calculated as

(1)

where c is the velocity of the light.

As an example, assuming the pulse width τ = 0.25 µs and the pulse repetition interval T = 10 µs, the range resolution would be 37.5 m. For a resolution of less than 1 meter, from (1) the pulse duration would have to be shortened to say, T = 3.9 ns. The resulting range resolution would then be 0.58 m and instead of handling a 4-MHz bandwidth the new system bandwidth would be 250 ns/3.9 ns = 64 wider than the original system bandwidth.

To achieve a high resolution at the 0.58 m without reducing the pulse duration, SFR could be employed. As shown in Figure 1, SFR transmits sequences of N pulses at a fixed pulse-repetition frequency, but not at a fixed radar frequency. Unlike the pulse signal, each pulse in the sequence of a stepped frequency waveform has the same pulse width and time duration, but different carrier frequency. That frequency is given by fi = fo+N*dF, where dF is the amount of frequency increased, indicating that frequency hopping and time division are used.

Assuming the N-step stepped frequency is used, the pulse width and pulse repetition interval are still τ = 0.25 µs and T = 10 µs where N = 64, as from the previous example, and dF = 4 MHz, the resulting range resolution bandwidth would be Rs = c/{2*(ƒo+(N-1)*dF)} = 0.58 m. As is clearly evident from this result, SFR has a high range resolution (less than one meter). Moreover, it was achieved without having to shorten the resolution, making it preferable to pulse radar in this scenario.

Platform for designing, testing SFR
In SFR radar, clutter interferes with target detection, making it difficult to find the actual number of targets or even causing it to fail in detecting small targets. Finding a closed-form analytical solution that enables target detection to be analyzed in the presence of this clutter is also difficult. Because of the significance in analyzing these types of scenarios, simulation becomes critically important, as does the use of a platform solution for simulation of SFR systems under real-world environments. The platform can also be used for verification and testing of SFR systems. The simulation platform with test environment must include return signal radar cross section (RCS) and background clutter.

To better understand how such a platform might be used to design, verify and test a SFR system, a template SFR design is provided below. By customizing the template SFR design for their own systems, engineers can run simulations in the platform to evaluate the design’s performance. When design simulation is combined with test equipment, the simulation platform can also be used as a test platform for SFR component hardware testing. As an example, an SFR system with two target returns and ground clutter is presented in which the platform is used for both simulation and hardware test.

Simulating an SFR system
Consider the basic SFR design shown in Figure 2. In the signal generator, a SFR source is followed by an RF modulator, then two target models and a clutter model are used. At the SFR receiver input, received signals include target return and clutter.

Figure 2: This example of an SFR simulation is performed using Agilent Technologies’ SystemVue electronic system level design platform.

The received signal is measured at the input of the SFR and displayed in Figure 3. Note that the plot of frequency versus time in Figure 3C, is in keeping with what one would expect for the SFR signals based on the carrier frequency calculation previously described. The unwrapped phase is also expected. Additionally, the SFR receiver works fine in simulation.

Figure 3. Shown here is the spectrum (A), magnitude of the waveform that reflects the random characters of the target return as well as the clutter property (B), frequency hopping in the received signal (C), and the unwrapped phase (D) of a received SF radar signal measured at the receiver input.
Click on image to enlarge

Figure 4 shows that using this high-resolution SFR design, two targets close to one another can be easily detected. To detect the same two targets using a pulse radar, the pulse width would have to be increased at least 8 times, significantly increasing system cost.

Figure 4. A high-resolution SFR design was used to detect these two targets near one another.

Testing an SFR system
Once a simulation model is built that replicates the actual SFR system, it can be used for testing under real-world environments including RCS and background clutter. Moreover, the same simulation platform used to design the high-resolution SFR system can now be used for hardware receiver testing. Performing this test requires a SFR signal generator. The received signal includes target returns with environments such as ground clutter and noise.

First a test signal with two targets near each other and clutter is generated using the design in Figure 2. The signal is then downloaded into a vector signal generator for up-conversion to RF frequencies. Next, a signal analyzer measures the input of the SFR receiver and, along with signal analysis software, uses the measured data to verify the test signal.

For SFR transmitter test, a SFR receiver is needed. The SFR receiver can be created using the simulation platform. The signal is measured using a vector signal analyzer running vector signal analysis software. The received signal is then downloaded from the signal analyzer into the EDA software for demodulation, as well as detection and recovery of the original target signals. The created SFR software receiver can be used to test real received SFR signals using the test setup in Figure 5.

Figure 5. This test setup can be used for hardware test of the SFR transmitter and receiver.

Click on image to enlarge

Conclusion
A platform solution for simulating systems under real-world environments is the ideal tool for engineers designing, verifying and testing today’s radar systems. It is particularly useful with high-resolution Stepped Frequency Radar systems, where target detection in the presence of clutter can be difficult to analyze with closed-form solutions. The simulation platform is well suited for designing and testing SFR systems. In the latter case, a simulated software receiver can be used for transmitter component testing, while a simulated software transmitter can be used for receiver component testing.

About the author:
Dingqing Lu, Scientist – Eesof EDA, Software and Modular Solutions Division, Agilent Technologies

Dingqing Lu has been with Agilent Technologies/Hewlett Packard Company since 1989 and is a scientist with Agilent EEsof EDA, working on modeling, simulation, testing and implementation of Military and Satellite Communications and Radar EW systems. From 1981 to 1986 he was with University of Sichuan as Lecturer and Assistant Professor. He was a Research Associate in the Department of Electrical Engineering at University of California (UCLA) from 1986 to 1989. He is IEEE senior member and has published 20 papers on IEEE Transactions, Journals and Conference Proceedings. He also holds a US patent on fast DSP search algorithm. His research interests include system modeling, simulation and measurement techniques.