Fast and accurate computation of light path deviation due to atmospheric refraction is an important requirement for real-time simulation of optical imaging sensor systems. A large body of existing literature covers various methods for application of Snell’s Law to the light path ray tracing problem. This paper provides a discussion of the adaptation to real time simulation of atmospheric refraction ray tracing techniques used in mid-1980's LOWTRAN releases. The refraction ray trace algorithm published in a LOWTRAN-6 technical report by Kneizys (et. al.) has been coded in MATLAB for development, and in C-language for simulation use. To this published algorithm we have added tuning parameters for variable path segment lengths, and extensions for Earth grazing and exoatmospheric "near Earth" ray paths. Model atmosphere properties used to exercise the refraction algorithm were obtained from tables published in another LOWTRAN-6 related report. The LOWTRAN-6 based refraction model is applicable to atmospheric propagation at wavelengths in the IR and visible bands of the electromagnetic spectrum. It has been used during the past two years by engineers at the U.S. Army Aviation and Missile Research, Development and Engineering Center (AMRDEC) in support of several advanced imaging sensor simulations. Recently, a faster (but sufficiently accurate) method using Gauss-Chebyshev Quadrature integration for evaluating the refraction integral was adopted.
Rendering of point scatterer based radar scenes for millimeter wave (mmW) seeker tests in real-time hardware-in-the-loop (HWIL) scene generation requires efficient algorithms and vector-friendly computer architectures for complex signal synthesis. New processor technology from Intel implements an extended 256-bit vector SIMD instruction set (AVX, AVX2) in a multi-core CPU design providing peak execution rates of hundreds of GigaFLOPS (GFLOPS) on one chip. Real world mmW scene generation code can approach peak SIMD execution rates only after careful algorithm and source code design. An effective software design will maintain high computing intensity emphasizing register-to-register SIMD arithmetic operations over data movement between CPU caches or off-chip memories. Engineers at the U.S. Army Aviation and Missile Research, Development and Engineering Center (AMRDEC) applied two basic parallel coding methods to assess new 256-bit SIMD multi-core architectures for mmW scene generation in HWIL. These include use of POSIX threads built on vector library functions and more portable, highlevel parallel code based on compiler technology (e.g. OpenMP pragmas and SIMD autovectorization). Since CPU technology is rapidly advancing toward high processor core counts and TeraFLOPS peak SIMD execution rates, it is imperative that coding methods be identified which produce efficient and maintainable parallel code. This paper describes the algorithms used in point scatterer target model rendering, the parallelization of those algorithms, and the execution performance achieved on an AVX multi-core machine using the two basic parallel coding methods. The paper concludes with estimates for scale-up performance on upcoming multi-core technology.
Frequency stepping is an established technique for increasing the range resolution of pulsed
Linear Frequency Modulation (LFM, or chirp) radar waveforms . When a monostatic radar
system employs this waveform for increased range resolution measurements on an object with
motion relative to the radar platform, simple changes in the received waveform arise, requiring
fine motion compensation on a per-pulse basis. These motion effects include phase, frequency
and frequency slope offsets which vary according to the transmitted pulse frequency and
frequency rate, and the object range and range rate. All three offsets are easily compensated by
complementary offsets in Direct Digital Synthesizer outputs used to form frequency conversion
LO signals in the radar receiver. Radars employing stepped frequency LFM waveforms may be
tested in a Hardware-in-the-Loop (HWIL) facility in simulations involving scenes or objects with
radar-relative motion. Under these conditions, the motion effects on the radar receiver input
signals must be accurately computed, synthesized and must modify the transmit signal prior to its
return to the receiver. Engineers at the U.S. Army AMRDEC Advanced Simulation Center have
developed signal processing techniques for accurate simulation of fine range motion effects to
support HWIL testing of pulsed LFM radar systems. This paper provides an analysis of the
signal processing involved for a simple model of an HWIL RF signal generation chain. Some
results are presented from successful application of the motion simulation methods in an HWIL
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.
Proc. SPIE. 7301, Technologies for Synthetic Environments: Hardware-in-the-Loop Testing XIV
KEYWORDS: Radar, Modulation, Digital filtering, Computing systems, Field programmable gate arrays, Signal processing, Data conversion, Analog electronics, Electronic filtering, Filtering (signal processing)
Matched filter processing for pulse compression of phase coded waveforms is a classic method
for increasing radar range measurement resolution. A generic approach for simulating high
resolution range extended radar scenes in a Hardware in the Loop (HWIL) test environment is to
pass the phase coded radar transmit pulse through an RF tapped delay line comprised of
individually amplitude- and phase-weighted output taps. In the generic approach, the taps are
closely spaced relative to time intervals equivalent to the range resolution of the compressed radar
pulse. For a range-extended high resolution clutter scene, the increased number of these taps can
make an analog implementation of an RF tapped delay system impractical. Engineers at the U.S.
Army Aviation and Missile Research, Development and Engineering Center (AMRDEC) have
addressed this problem by transferring RF tapped delay line signal operations to the digital
domain. New digital tapped delay line (DTDL) systems have been designed and demonstrated
which are physically compact compared to analog RF TDLs, leverage low cost FPGA and data
converter technology, and may be readily expanded using open slots in a VME card cage. In
initial HWIL applications, the new DTDLs have been shown to produce better dynamic range in
pulse compressed range profiles than their analog TDL predecessors.
This paper describes the signal requirements and system architecture for digital tapped delay
lines. Implementation, performance, and HWIL simulation integration issues for AMRDEC's
first generation DTDLs are addressed. The paper concludes with future requirements and plans
for ongoing DTDL technology development at AMRDEC.
High resolution millimeter wave RADAR has become a reality in today's sensor world. System development and simulation-based acquisition are increasing the demands for high fidelity environmental models. Therefore, high resolution clutter models are imperative. The RADAR ground clutter signal is most often treated as some random variable with a given probability distribution function with some mean value depending upon the particular RADAR, desired clutter properties, and relative geometry. This paper will show the results of implementing a technique to correlate adjacent clutter scatterers in the clutter model while maintaining the overall clutter statistics. This correlated clutter will be matched to a clutter class map which was derived from visual data of a specific terrain location. Processed images from a high resolution RADAR simulation will be shown and compared to the visual images and clutter maps.
Hardware-in-the-Loop (HWIL) simulation is an effective, low cost alternative to flight testing for performance evaluation and development of guided missile RF seekers. Accurate generation of a simulated RF environment including target and clutter return is of primary importance in creating a realistic HWIL simulation. This paper discusses the signal generation requirements for typical millimeter- wave seeker simulation and the hardware developed for their implementation. Special emphasis is placed on a device for computer-controlled I/Q modulation called a Digital Quadrature Modulator.