Motion Modeling in Phased Array Systems
Support for Motion Modeling
A critical component in phased array system applications is the ability to model motion in space. Such modeling includes the motion of arrays, targets, and sources of interference. For convenience, you can ignore the distinction between these objects and collectively model the motion of a platform.
Extended bodies can undergo both translational and rotational motion in space. Phased Array System Toolbox™ software supports modeling of translational motion.
Modeling translational platform motion requires the specification of a position and velocity vector. Specification of a position vector implies a coordinate system. In the Phased Array System Toolbox, platform position and velocity are specified in a Global Coordinate System. You can think of the platform position as the displacement vector from the global origin or as the coordinates of a point with respect to the global origin.
Let r0 denote the position vector at time 0 and v denote the velocity vector. The position vector of a platform as a function of time, r(t), is:
The following figure depicts the vector interpretation of translational motion.
When the platform represents a sensor element or array, it is important to know the orientation of the element or array local coordinate axes. For example, the orientation of the local coordinate axes is necessary to extract angle information from incident waveforms. See Global and Local Coordinate Systems for a description of global and local coordinate systems in the software. Finally, for platforms with varying velocity, you must be able to update the velocity vector over time.
You can model platform position, velocity, and local axes orientation
Platform Motion with Constant Velocity
Beginning with a simple example, model the motion of a platform over ten time steps. To determine the time step, assume that you have a pulse transmitter with a pulse repetition frequency (PRF) of 1 kilohertz. Accordingly, the time interval between each pulse is 1 millisecond. Set the time step equal to pulse repetition interval.
PRF = 1e3; Tstep = 1/PRF; Nsteps = 10;
Next, construct a platform object specifying the platform initial position and velocity. Assume that the initial position of the platform is 100 meters from the origin at (60,80,0). Assume the speed is approximately 30 meters per second (m/s) with the constant velocity vector given by (15,25.98,0) .
platform = phased.Platform('InitialPosition',[60;80;0],'Velocity',[15;25.98;0]);
The orientation of the local coordinate axes of the platform is the value of the
InitialOrientationAxes property. You can view the value of this property by entering
hplat.InitialOrientationAxes at the MATLAB™ command prompt. Because the
InitialOrientationAxes property is not specified in the construction of the
phased.Platform System object™, the property is assigned its default value of
[1 0 0;0 1 0;0 0 1]. Use the
step method to simulate the translational motion of the platform.
initialPos = platform.InitialPosition; for k = 1:Nsteps pos = platform(Tstep); end finalPos = pos + platform.Velocity*Tstep; distTravel = norm(finalPos - initialPos)
distTravel = 0.3000
step method returns the current position of the platform and then updates the platform position based on the time step and velocity. Equivalently, the first time you invoke the
step method, the output is the position of the platform at t = 0.
Recall that the platform is moving with a constant velocity of approximately 30 m/s. The total time elapsed is 0.01 seconds. Invoking the
step method returns the current position of the platform and then updates that position. Accordingly, you expect the final position to differ from the initial position by 0.30 meters. Confirm this difference by examining the value of
Platform Motion with Changing Velocity
Most platforms in phased array applications do not move with constant velocity. If the time interval described by the number of time steps is small with respect to the platform speed, you can often approximate the velocity as constant. However, there are situations where you must update the platform velocity over time. You can do so with the
phased.Platform System Object™ because the
Velocity property is tunable.
This example models a target initially at rest. The initial velocity vector is (0,0,0) . Assume the time step is 1 millisecond. After 500 milliseconds, the platform begins to move with a speed of approximately 10 m/s. The velocity vector is (7.07,7.07,0). The platform continues at this velocity for an additional 500 milliseconds.
Tstep = 1e-3; Nsteps = 1/Tstep; platform = phased.Platform('InitialPosition',[100;100;0]); for k = 1:Nsteps/2 [tgtpos,tgtvel] = platform(Tstep); end platform.Velocity = [7.07; 7.07; 0]; for k = Nsteps/2+1:Nsteps [tgtpos,tgtvel] = platform(Tstep); end
Track Range and Angle Changes Between Platforms
This example uses the
phased.Platform System object™ to model the change in range between a stationary radar and a moving target. The radar is located at (1000,1000,0) and has a velocity of (0,0,0) . The target has an initial position of (5000,8000,0 ) and moves with a constant velocity of (−30,−45,0). The pulse repetition frequency (PRF) is 1 kHz. Assume that the radar emits ten pulses.
PRF = 1e3; Tstep = 1/PRF; radar = phased.Platform('InitialPosition',[1000;1000;0]); target = phased.Platform('InitialPosition',[5000;8000;0],... 'Velocity',[-30;-45;0]);
Calculate initial target range and angle.
[initRng, initAng] = rangeangle(target.InitialPosition,radar.InitialPosition);
Calculate relative radial speed.
v = radialspeed(target.InitialPosition,target.Velocity,... radar.InitialPosition);
Simulate target motion.
Npulses = 10; for num = 1:Npulses tgtpos = target(Tstep); end
Compute last target position.
tgtpos = tgtpos + target.Velocity*Tstep;
Calculate final target range and angle.
[finalRng,finalAng] = rangeangle(tgtpos,radar.InitialPosition); deltaRng = finalRng - initRng
deltaRng = -0.5396
The constant velocity of the target is approximately 54 m/s. The total time elapsed is 0.01 seconds. The range between the target and the radar should decrease by approximately 54 centimeters. Compare the initial range of the target,
initRng, to the final range,
finalRng, to confirm that this decrease occurs.