Joint specific emitter identification and tracking using device nonlinearity estimation

Open Access
Liu, Ming-Wei
Graduate Program:
Electrical Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
May 24, 2011
Committee Members:
  • John F Doherty, Dissertation Advisor
  • John F Doherty, Committee Chair
  • William Kenneth Jenkins, Committee Member
  • John Metzner, Committee Member
  • Allan Gunnar Sonsteby, Committee Member
  • specific emitter identification
  • tracking
  • communication system security
  • reconnaissance
  • OFDM
  • nonlinear distortion
  • frequency-selective fading channels
In this thesis, we present radio frequency (RF) front-end nonlinearity estimators to per- form joint specific emitter identification (SEI) and tracking. Our SEI systems discern radio emitters of interest through the estimation of transmitter nonlinearities caused by design and fabrication variations. These nonlinearity features provide unique signal sig- natures for each emitter, and we extract those characteristics through the estimation of transmitter nonlinearity coefficients. We first present a nonlinearity estimator which es- timates the power series coefficients of nonlinear devices in the radio frequency (RF) front end by observing the spectral regrowth in additive white Gaussian noise (AWGN) channel. Then another robust algorithm is also provided by using alternative degrees of nonlinearities associated with symbol amplitudes for initial estimation, and then iter- atively estimating the channel coefficients and distorted transmit symbols to overcome the inter-symbol interference (ISI) effect. The convergence and unbiasedness of the it- erative estimator are demonstrated semi-analytically. Based on this analysis, we also trade error performance for complexity reduction using the regularity of the estimation process. The algorithm is applicable to a wide range of multi-amplitude modulation schemes, and we present an SEI system designed for an orthogonal frequency division multiplexing (OFDM) system over an empirical indoor channel model with associated numerical results. This technology is then adapted to provide location tracking in multi- path environments, which locates the mobile stations (MS) based on the transmit power variation estimates. The method is simulated over a grid-based city map. In the last part of the thesis, complexity reduction methods are introduced to balance the convergence rate and identification performance.