EE381K-6 Estimation Theory: Final Projects

The University of Texas at Austin
Department of Electrical and Computer Engineering

EE381K-6 Estimation Theory: Final Projects

Types of final projects (students may choose either one):

Survey Paper: A paper in the format of a journal or a conference paper that provides a survey of a chosen topic. This project should have a detailed survey component and a simulation component for full credit. Survey papers should be unbiased and well written.
Research Paper: A paper in the format of a journal or a conference paper, containing some element of innovation, even if it is small or incremental.

Project Proposal (Due: on 11/07/17, in class)

The research paper proposal should state the problem that you are trying to solve, including setup, key assumptions, and methods to be used. For a survey project, the goal is a more detailed overview of the area.
Please include realistic plan of action proposed, including action items and a timeline. At least 2-3 papers that you have read should be cited. Proposal format: 1-2 pages in length (2 pages preferred) with 11pt or 12pt font, single spacing, and standard 1 inch margins.

Evaluating final reports (the reports are due 12/11/17, 4:30pm, EER 7.810 or by e-mail):

Survey Article

(15 points) In the Introduction section, the article should provide background on the general area and motivate the survey.

(10 points) The references should be relevant to the topic of the survey. Journal papers are strongly preferred. Including references which present different approaches to the solution of the same problem is desirable.

(65 points) The main part of the article -- survey of the area -- should provide details about the area/problem being surveyed; give a thorough description of the contributions in the cited papers; compare and contrast different contributions, including numerical/simulation illustrations; and give some insight and provide suggestions for future work.

(10 points) Since this is a report, please take care of clarity and style thereof. Please use 11pt or 12pt font (references may be 10pt), double spaced text, standard 1 inch margins. Preferred length (not including title, abstract, figures, and table-of-contents) is 10-12 pages.
Research Paper

(15 points) In the Introduction section, the article should provide background on the general area and motivate the research project.

(10 points) In the Introduction (or a Problem Statement) section, clearly describe the objectives of the project. (Ideally, they should be similar to the objectives outlined in the project proposal.)

(65 points) The main part of the paper should provide concise problem statement, setup, and key assumptions; description of methods (any derivations, algorithms employed, etc.); explanation of the project contributions illustrated with analytical and/or simulation results; and give some insight and provide suggestions for future work.

(10 points) Since this is a report, please take care of clarity and style thereof. Please use 11pt or 12pt font (references may be 10pt), double spaced text, standard 1 inch margins. Preferred length (not including title, abstract, figures, and table-of-contents) is 10-12 pages.

An incomplete list of potential projects, with some interesting papers (last updated: 09/26/17):

Kalman filtering for sparse signals

N. Vaswani, ``Kalman filtered compressed sensing," Image Processing, 2008. ICIP 2008. 15th IEEE International Conference on.
N. Vaswani and W. Lu, ``Modified-CS: Modifying compressive sensing for problems with partially known support," IEEE Transactions on Signal Processing 58 (9), Sept. 2010, 4595-4607.
A. Carmi , P. Gurfil and D. Kanevsky, ``Methods for sparse signal recovery using Kalman filtering with embedded pseudo-measurement norms and quasi-norms," IEEE Trans. Signal Process., vol. 58, no. 4, pp. 2405-2409, 2010.

Distributed Kalman filtering

R. Olfati-Saber, ``Distributed Kalman filter with embedded concensus filter," CDC, 2005.

Deep tracking and deep Kalman filter

P. Ondruska and I. Posner, ``Deep Tracking: Seeing Beyond Seeing Using Recurrent Neural Networks," AAAI, February 12-17, 20016, Phoenix, AZ.
R. G. Krishnan, U. Shalit and D. Sontag, ``Deep Kalman Filters," Arxiv, 2015.

Cognitive radio

Seung-Jun Kim, Dall'Anese, E., Giannakis, G.B., ``Cooperative Spectrum Sensing for Cognitive Radios Using Kriged Kalman Filtering," Selected Topics in Signal Processing, IEEE Journal of (Volume:5, Issue: 1), 2011.

Applications of Cramer-Rao bounds

Montorsi, F. and Vitetta, G.M., ``On the Performance Limits of Pilot-Based Estimation of Bandlimited Frequency-Selective Communication Channels," Communications, IEEE Transactions on (Volume:59, Issue: 11), 2011.

Total least-squares

R. Roy and T. Kailath, ``ESPRIT-estimation of signal parameters via rotational invariance techniques," Acoustics, Speech and Signal Processing, IEEE Transactions on (Volume: 37, Issue: 7), 1989.

Resolving mixtures of exponential functions

M. R. Osborne and G. K. Smyth, ``A modified Prony algorithm for fitting sums of exponential functions," SIAM J. Sci. Statist. Comput., 16, 1995, pp. 119-138.
E. M. Dowling et. al., ``Exponential parameter estimation in the presence of known components and noise," IEEE Transactions on Antennas and Propagation, vol. 42, no. 5, May 1994.

Sensor selection and scheduling

A. Bertrand et al., ``Efficient calculation of sensor utility and sensor removal in wireless sensor networks for adaptive signal estimation and beamforming," Signal Processing, IEEE Transactions on (Volume:60, Issue: 11), 2012.

Estimating parameters of diffusion processes

H. Sorensen, ``Parametric inference for diffusion processes observed at discrete points in time: A survey," Int. Statist. Rev., vol. 72, no. 3, pp. 337-354, 2004.

Neural decoding with Kalman, particle filters

W. Wu et al., ``Bayesian population decoding of motor cortical activity using a Kalman filter," Neural Comput. 2006 Jan;18(1):80-118.
A. E. Brockwell et al., ``Recursive bayesian decoding of motor cortical signals by particle filtering," J Neurophysiol. 2004 Apr;91(4):1899-907.

Channel estimation

C. Komninakis et al., ``Multi-Input Multi-Output Fading Channel Tracking and Equalization Using Kalman Estimation," Signal Processing, IEEE Transactions on (Volume:50, Issue: 5).
B. Balakumar et al., ``Joint MIMO Channel Tracking and Symbol Decoding Using Kalman Filtering," Signal Processing, IEEE Transactions on (Volume:55, Issue: 12).