Research

Research Interests

Radiation pattern and co-site interference prediction for antennas on electrically large platforms
Radar cross section prediction for electrically large platforms
Modeling of indoor propagation and wireless propagation in urban environments
RF/microwave modeling and characterization of electrical connector fault precursors
Design, modeling and simulation of wireless and embedded sensors
Breathing, heartbeat and presence detection using Doppler and UWB radars
Waveguide propagation modeling and characterization
High performance computing for computational electromagnetics
Antenna design and modeling for various low and high frequency applications
Design and optimization of radar absorbing materials and radomes
Modeling of radiated electromagnetic interference from power converters
Low frequency electromagnetic shielding of enclosures

Fast and accurate techniques are desired to predict radiation pattern and co-site interference for antennas placed on electrically large platforms.
Unlike full-wave methods, uniform geometrical theory of diffraction (UTD) scales with frequency, enabling huge savings in computation time and memory in analysis of antenna placement on electrically large platforms.
UTD is the only asymptotic technique that accounts for creeping rays.
Ray tracing is applied to faceted structures to identify ray mechanisms.
UTD and Geometrical Optics are applied along traced rays to find fields.
This research was partially supported by the Proposal Writing Research Enhancement Grant from the Office of Research and Sponsored Programs.

Iterative physical optics (IPO) allows for faster computation compared to full-wave methods by enabling use of coarser mesh to represent an object.
IPO convergence has been improved using forward-backward approach.
The computational speed of IPO has been significantly improved using fast far field approximation, model based parameter estimation as well as MPI parallelization on CPUs and CUDA implementation on GPUs.
IPO has been used in various applications including prediction of radar cross section of electrically large platforms as well as modeling of indoor propagation and wireless propagation in urban environments.

A typical aircraft contains miles of wiring and hundreds of connectors.
Manual inspection of precursors to electrical faults is labor intensive.
Electrical faults may cause fire and loss of flight critical system functionality.
Detection of fault precursors enables timely maintenance and mitigation.
MIL-DTL-38999 series III is a connector family widely used in avionics.
Transmission line modeling, and FEKO coaxial line modeling and simulations were done to model a partially inserted contact within a connector.
Nonlinear least squares and FEKO optimization tools were employed to optimize connector parameters to match measured data.
An UWB radar was used to measure connector transfer function, which was employed to model it as an alternative to network analyzer measurements.
This research was funded by the NASA Ames Research Center while Dr. Tokgoz was with United Technologies Research Center, East Hartford, CT.

Modeling and simulations were performed for different types of wireless and embedded sensors for a variety of applications including the following:
- Application of Doppler and UWB radars as well as signal processing techniques for breathing, heartbeat and presence detection,
- Modeling and simulation of communications and wireless power transfer through rotor blade carbon spar for wireless sensors,
- Modeling of a direct write fabricated wear sensor embedded in an abradable coating for industrial and aero gas turbine applications,
- Modeling and simulation of communications and wireless power transfer between reader and tag coils for an embedded RFID sensor,