Born in Tehran, Iran, on June 24, 1979, received the B.Sc. in electrical Engineering from Sharif University of Technology in 1999, M.Sc. and Ph.D. in Electrical Engineering from the University of Minnesota in 2003 and 2005 respectively. Present position: Senior R&D Engineer, America Online (AOL) Inc.
Ph. D. in Electrical Engineering, University of Minnesota, Minneapolis, MN 10/2005
Thesis: Tradeoffs in Resolution of Nonlinear Spectral Estimation, GPA:4.0/4.0
M. Sc. in Electrical Engineering, University of Minnesota, Minneapolis, MN 01/2003
Thesis: Subspace Spectral Analysis Based on State-Covariance and its Applications, GPA:4.0/4.0
B. Sc. in Electrical Engineering, Sharif University of Technology, Tehran, Iran 01/2000
· Awarded Doctoral Dissertation Fellowship for Ph.D. thesis proposal, University of Minnesota, 2004.
· Ranked 1st in the Nationwide Entrance Exam toward Graduate Studies (5500 examinees), Iran, 1999.
· Ranked 9th in the Third National Electrical Engineering Olympiad (6500 participants), Iran, 1999.
· Ranked 49th in the Nationwide University Entrance Exam (350,000 examinees), Iran, 1995.
Senior Research and Development Engineer 11/2005-Present
America Online (AOL) Inc., Mountain View, CA
Conducting research on Internet modeling, optimization, and control.
Graduate Research Assistant 01/2001 – 10/2005
University of Minnesota, Minneapolis, MN
Advisor: Professor Tryphon T. Georgiou
· Spectral Estimation based on State-Covariance Analysis
Subspace methods for spectral analysis can be adapted to the case where state-covariance matrix replaces the traditional Toeplitz matrix formed out of a partial autocorrelation sequence of a time-series. This forms the basis of a generalized framework for spectral analysis that classical techniques such as MUSIC, ESPRIT, and Maximum entropy are special cases. In order to quantify advantages of working with state-covariance data instead of the autocorrelation sequence I analyzed tradeoffs between resolution and robustness in spectral estimates due to the dynamics of state space filter. This gives rise to a novel class of high resolution spectral estimators that can be tailored to specific applications such as radar imaging, biomedical diagnostics, and sensor arrays and networks processing.
· Noninvasive Measurement and Control of Tissue Temperature
I worked with Prof. Emad Ebbini on noninvasive tissue temperature estimation from pulse-echo radio frequency signals obtained by standard diagnostic ultrasound imaging equipment. The temperature change in the tissue causes a physical phenomenon: the spectral contents of the backscattered signal exhibit frequency shifts related to the temperature change. This is a result of local change in the speed of sound and thermal expansion, which affects scatterers spacing. I devised a new high-resolution algorithm, based on state-covariance analysis, to track frequency shifts in the backscattered signal. I demonstrated the superiority of the new approach to the classical one. Experiments on tissue samples have shown a great promise to make a reliable noninvasive temperature sensing system for clinical treatments.
· Synthetic Aperture Radar (SAR) Imaging
Synthetic aperture radar imaging can be viewed as a parameter estimation problem, in which one seeks to estimate the scene reflectivity versus slant-plane location. The physics of SAR justifies modeling a backscattered signal as a superposition of complex sinusoids. For this reason, model-based spectral estimation techniques offer an attractive alternative to Fourier-based methods. The tuning property of our recently developed spectral analysis tools enabled us to adaptively control the resolution in different parts of the reconstructed SAR image. Furthermore, the framework allows for parsimonious models because of the ability to focus on smaller patches. This contrasts with the inherent difficulties of earlier techniques where, for faithful reconstruction of typical SAR scenery with multiple scatterers, a very high order model is needed. New SAR imaging algorithm has been developed and applied on the field data. Reconstructed images demonstrate great promises in reducing side-lobe artifacts and speckle noise.
· Multirate Multidimensional Spectral Estimation
A technique for spectral analysis in the context of multi-rate sampling by a collection of sensors, e.g. sensor networks, is proposed. Correlation of the time-domain samples gives rise to moment constraints for the unknown power spectrum. A homotopy-based technique is then used to identify consistent power spectra. The spectra we obtain are at a minimum distance in the Kullback-Leibler sense to a given “prior” and the “maximum entropy” power spectrum corresponds to the special case where the prior is white.
· Controller Design with Complexity Constraints
We proposed an approach for shaping closed-loop operators while keeping their Mcmillan degree bounded by the sum of unstable plant-poles and non-minimum phase plant-zeros. We make use of recent developments in analytic interpolation with degree constraint and we focus on the paradigm of sensitivity minimization. The sensitivity function can be obtained as the minimizer of a convex weighted-entropy functional. It is the choice of this weight that we formulated as a convex optimization problem.
· Classification of Motor Imagery via Von Neumann Entropy for Brain Computer Interface
I collaborated with Biomedical Functional Imaging and Computation laboratory at the University of Minnesota in developing algorithm for BCI. We use scalp EEG recordings of subjects where they are asked to imagine left or right hand movement. The goal is to classify left and right hand motor imagery tasks from EEG signals. After some standard preprocessing by surface Laplacian filtering and independent component analysis, the cortical imaging technique is used to solve the EEG inverse problem. Then cortical current density distributions of left and right hand trials are classified by exploiting distance measure derived from Von Neumann entropy. The proposed method was tested on three human subjects (180 trials each) and maximum accuracy of 91.5% and average accuracy of 88% were obtained. The results confirm the hypothesis that source analysis methods may improve accuracy for classification of motor imagery tasks.
Research Assistant 09/1998 – 08/2000
Sharif University of Technology, Tehran, Iran,
Advisors: Prof. Alireza Karimi, Prof. Mohamad Haeri;
I designed and implemented a plant simulator and a PID controller for Sharif University control systems lab. This device is used for educational purposes in undergraduate courses on control systems.
Teaching Assistant 09/2003 – 06/2005
University of Minnesota, Minneapolis, Minnesota
Courses: Linear systems and optimal control, Robust control design.
Teaching Assistant 02/1998 – 08/2000
Sharif University of Technology, Tehran, Iran
Courses: Linear Control Systems, Control System Laboratory.
Journals
1. A. Nasiri Amini, E. Ebbini, and T. Georgiou, “Noninvasive Tissue Temperature Estimation Using High-resolution Spectral Analysis,” IEEE Trans. on Biomedical Engineering, vol. 52, no. 2, Feb 2005.
2. A. Nasiri Amini, and T. Georgiou, “Avoiding Ambiguity in Beamspace Processing,” IEEE Signal Processing Letters, vol. 12, no. 5, May 2005.
3. A. Nasiri Amini, and T. Georgiou, “Tunable Spectral Line Estimators Based on State-Covariance Subspace Analysis,” IEEE Trans.on Signal processing, in press.
4. B. Kamousi, A. Nasiri Amini, and B. He, “Classification of Motor Imagery by Means of Cortical Current Density Estimation and Von Neumann Entropy for Brain Computer Interface Applications,” submitted to Journal of Clinical Neurophysiology, Nov 2005.
5. A. Nasiri Amini, and T. Georgiou, “SAR Imaging via State-Covariance Subspace-Estimation Methods,” submitted to IEEE Trans. on Aerospace and Electronic Systems .
6. A. Nasiri Amini, and T. Georgiou, “Multidimensional Multirate Spectral Analysis,” in preparation to be submitted to IEEE Trans. on Signal processing.
Refereed Conferences and Workshops
1. A. Nasiri Amini, and T. Georgiou, “Statistical Analysis of State-covariance Subspace Methods,” Proc. of the 41st Conference on Decision and Control, Las Vegas, NV, Dec. 2002.
2. A. Nasiri Amini, E. Ebbini, and T. Georgiou, “Noninvasive Tissue Temperature Estimation via State-covariance Spectral Estimation,” Proc. of the 10th IEEE DSP workshop, NM, Aug. 2004.
3. A. Nasiri Amini, M. S. Takyar, and T. Georgiou, “A Homotopy Approach for Multirate Spectral Estimation,” to appear in Proc. of IEEE conf. on Acoustic, Speech, and Signal processing, France, May 2006.
4. M. S. Takyar, A. Nasiri Amini, and T. Georgiou, “Sensitivity shaping with complexity constraints using convex optimization,” to appear in Proc. of 25th American Control Conference, Jun 2006.
5. A. Nasiri Amini, M. S. Takyar, and T. Georgiou, “Spectral Analysis from Multi-Rate Observations: Itakura-Saito based Approximation of Power Spectra and Capon-Like Spectral Envelopes,” submitted to 4th IEEE workshop. on sensor arrays and multi-channel processing, Boston, July 2006.
6. M. S. Takyar, A. Nasiri Amini, and M. Massoumnia, “Improving Vertical Gyro Accuracy by Exploiting Moving Constraints in Marine Applications,” Proc. of the Institute of Navigation National Technical Meeting, San Diego, CA, 2005.
Conference Referee: IEEE Conference on Decision and Control, American Control Conference.
Member of IEEE: Signal Processing, Control Systems, and Engineering in Medicine and Biology Societies.