Modern millimeter wave (mmW) radar sensor systems employ wideband transmit waveforms and
efficient receiver signal processing methods for resolving accurate measurements of targets
embedded in complex backgrounds. Fast Fourier Transform processing of pulse return signal
samples is used to resolve range and Doppler locations, and amplitudes of scattered RF energy.
Angle glint from RF scattering centers can be measured by performing monopulse arithmetic on
signals resolved in both delta and sum antenna channels. Environment simulations for these sensors
- including all-digital and hardware-in-the-loop (HWIL) scene generators - require fast, efficient
methods for computing radar receiver input signals to support accurate simulations with acceptable
execution time and computer cost. Although all-digital and HWIL simulations differ in their
representations of the radar sensor (which is itself a simulation in the all-digital case), the signal
computations for mmW scene modeling are closely related for both types. Engineers at the U.S.
Army Aviation and Missile Research, Development and Engineering Center (AMRDEC) have
developed various fast methods for computing mmW scene raw signals to support both HWIL scene
projection and all-digital receiver model input signal synthesis. These methods range from high
level methods of decomposing radar scenes for accurate application of spatially-dependent nonlinear
scatterer phase history, to low-level methods of efficiently computing individual scatterer
complex signals and single precision transcendental functions. The efficiencies of these
computations are intimately tied to math and memory resources provided by computer architectures.
The paper concludes with a summary of radar scene computing performance on available computer
architectures, and an estimate of future growth potential for this computational performance.
This paper describes the Sensor Emulator System developed at AMCOM using custom off the shelf image processing hardware combined with in-house designed interfaces to SGI Digital Video Port (DVP) input and output hardware. The system was designed to allow the emulation processing elements to be inserted in the image output path of the SGI computers currently being used in the Hardware-in-the-Loop (HWIL) facilities. This is accomplished by converting the SGI's DVP output to the emulator's input bus format, and after being processed, the output is converted back to DVP format. The images can then be input to in-house designed injector or projector interfaces. Sixteen bit images of 256 X 256 pixels, at frame rates of 800 Hz have been input, processed in parallel on 5 nodes, and output with this system. The system's processing elements are Matrox Genesis image processing boards. Each processing node consists of a Texas Instruments C80, a Matrox Neighborhood Operation Accelerator ASIC (NOA2) and a Matrox Video Interface ASIC (VIA). The NOA2 is a multiplier/accumulator (MAC) array capable of 32 simultaneous sum of products at 50 MHz. The VIA provides high- speed links between acquisition, processing and display devices by controlling multiple independent 32 bit wide busses. It controls the image acquisition and fan-out to the processing Nodes and output without adding overhead. The C80 provides the processing for sensor electronic functions such as gains, offsets, dead pixels, saturation, etc. This combination has the capability of processing large, high frame rate images in real time.
Hardware-in-the-loop (HWIL) testing of infrared missile seekers has been a proven method for seeker evaluation for many years. Infrared HWIL testing has two primary modes, projection or injection. With infrared projection HWIL testing, the seeker's optics and detectors are retained as part of the simulation since an infrared scene in the correct waveband is presented to the seeker's optics, and is then detected and processed. When using the injection mode of HWIL testing the infrared scene is injected directly into the seeker's electronics and bypasses the imaging and detection process. When this type of simulation is used it is critical to model the optical and electrical processes that would have degraded the image in a real-world scenario. Real-time modeling of sensor system modulation transfer functions and other forms of image degradation is a computationally intensive task. The types of calculations necessary for real- time sensor modeling often push the processing requirements past the capabilities of standard processors and custom processing hardware is required. This paper discusses solutions to this problem that have been implemented for infrared seekers at the U.S. Army's Aviation and Missile Command's, Missile Research, Development, and Engineering Center.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.