IEEE Aerospace and Electronic Systems Magazine - June 2020 - 52

Fast Fully Adaptive Signalling for Target Matching

TARGET RESPONSE ESTIMATION
The waveform designed from the environmental estimation phase is transmitted to interrogate the extended target
response. Returns collected from a number of transmissions of the pulsed waveform are processed in a similar
manner to the PU evaluation data using the MDL procedure from the "PU Environment Estimation" section mentioned above. The procedure diverges from the PU
estimation in the formation of the FTM. Whereas the PU
processing creates qi 2 f0; 1g, the target response estimation creates qi 2 ½0; 1 corresponding to the normalized
PSD of the k largest PSD estimates. This new FTM will
include frequency components from the PU transmissions
in addition to the required target response. To ensure that
the FTM is restricted to the estimated target response, it is
scaled by the FTM from the PU estimation process, which
eliminates the contributions from the PU as
qTMI ¼ qTARGET qPU

(13)

where qTMI is the FTM required for creating the TMI
waveform, qTARGET is the FTM created by the target
response estimation, and qPU is the FTM created from PU
estimation.

TMI WAVEFORM DESIGN
The TMI waveform design follows precisely the procedure described in the "Waveform Design" section above.
The use of the composite FTM from the "Target Response
Estimation" section enables the ERA to create the required
waveform.

SIMULATION AND EXPERIMENTAL CONFIGURATION
A common framework for simulation and hardwarein-the-loop (HITL) experimentation has been developed
in MATLAB running under the Windows operating system. The framework start-up parameters dictate whether
transmitter, environment, target, and receiver interactions
are achieved by simulation or are carried out over-the-air
using hardware transmitters and receivers. The functionality is abstracted such that consciousness of the mode of
operation is restricted to the interface functions, which
either communicate with or simulate the hardware components. In the simulation mode no attempt is made to simulate the over-the-air path losses between the hardware
components or distortion effects introduced by the transmitter and receiver chains. These effects could easily be
added to the simulation components in the future. However, delays due to the cable length connecting the system
rack to the radar heads are included such that the range
delay measurements are common between simulation and
HITL operation.
52

Figure 3.
Physical layout representation for joint spectral coexistence and
TMI experiments.

Figure 3 illustrates the components of the simulation/
HITL framework. The timing control system is responsible for coordinating all activities. The remaining three
control blocks implement the control and processing activities required by the radar, target, and PU simulation
components.
It can be seen that the target simulator is represented
as a transmitter-receiver pair. In both full simulation and
HITL modes no physical target is employed. Instead, the
target is either fully simulated in software, or in HITL, by
receiving the over-the-air radar transmission in a separate
receiver, convolving this signal with the current target
response in software, and retransmitting the results overthe-air for reception by the radar. This allows for flexible,
repeatable control of the target behavior across both the
full simulation and HITL experiments.
Details of the dynamic PU emulation, and the dynamic
target response employed in the target simulator is described
in the "Problem Formulation" section.
The mapping of the framework from pure simulation to
HITL requires a hardware solution providing highly flexible, highly capable functionality. To provide this capability,
the framework is designed around the use of the Cognitive
Radar Experimental Workspace (CREW) [64] at The Ohio
State University ElectroScience Laboratory. However, by
replacing the components specifically related to interfacing
with the CREW hardware, it is equally applicable to alternative hardware platforms.
The CREW system consists of four transmitting channels plus four receiving channels, each operating at
94 GHz, with an instantaneous bandwidth of 1 GHz. Real
value sampling is carried out at 3 GHz, and the Hilbert
Transform applied to create complex samples over
1.5 GHz with a usable bandwidth of 1 GHz. Each of the
channels has its own transmitting or receiving head, containing the final up/down RF conversion stages, connected

IEEE A&E SYSTEMS MAGAZINE

JUNE 2020

IEEE Aerospace and Electronic Systems Magazine - June 2020

Table of Contents for the Digital Edition of IEEE Aerospace and Electronic Systems Magazine - June 2020