Projects Handout

Project report due Wednesday, April 22 by 5pm. You must email me a description of your proposed project by Friday, March 27.

Everyone must complete a project-working in groups of two is welcome as long as work is roughly twice that of a single person project. The project represents 25% of your grade. Your project report should be written as if it were a paper to be submitted to a conference. The main body can be up to 10 pages (up to 20 pages for two person project)and and should contain the following components:

1. A short review about the topic, you should include at least one reference to a paper you have read (not a textbook).
2. A concise description of the problem.
3. A detailed description of your solution to the problem.
4. Matlab simulation results.
5. A discussion of the significance of these results.
6. The appendix should contain complete MATLAB codes, messy derivations and any other information too detailed to keep in the main body.

You are strongly encouraged to come up with your own idea for a project. The best projects relate to something you have already worked on or familiar with. If you have the capability of measuring data of any kind yourself, then this can lead to an excellent project. Otherwise real data can be obtained from sources on the internet. Sometimes synthetic data is appropriate especially if you want to analyze properties of the solution-e.g., how things change with SNR.

Important Dates:

• By Friday March 27, 5pm. Email me a description of the nature of your project (at least one paragraph in length).
• Each Friday until you give your presentation, each of you must email me a short description of your progress for the week. If you are working in a group, only one email need be sent.
• Exam 2 will be on Thursday night, April 9 at 7:20pm.
• Oral Presentations: I would like each of you to give a short presentation on your accomplishments. We will have presentations the last four days of class: 4/15, 4/17, 4/20, and 4/22. The organization and preparation will be considered as part of your grade, but not your raw spoken English ability. Your attendance during the last four days of class will also be considering in grading.
• Final project reports are due on the last day of class by 5pm, Wednesday April 22.

A few suggestions for projects ideas are given below:

1. There are many applications in Clarkson's book that we have not discussed. Any of these will make a good project. Such applications include:
   • Adaptive line enhancement
   • Time delay estimation
   • Echo cancellation in voiced channels
   • Adaptive Differential Pulse Code Modulation (ADPCM)
   • Adaptive Arrays (Chap 8)
   • Control problems (look at Widrow)
   • DSP filter design (look at Widrow)
2. Vocoders using LPC for speech compression
3. Frequency domain implementations of adaptive filters (6.6 in Clarkson)
4. Spectral Estimation (Chapter 7)
5. 2D Wiener filters can be applied in image processing (look at most any image processing text book).
6. Add varying amounts of noise to a signal and use a reference to recover the original signal the best you can. If you know the original signal, you can measure exactly how well your algorithm is working. The original signal can be speech, an eeg trace or any other real signal.
7. Adaptive noise cancellation without a reference signal.
8. Recover a periodic signal from broadband noise.
9. Problems dealing with multiple time series data.
10. How many taps? Use the information theoretic criterion described in Haykin section 2.10 and develop an algorithm for choosing an appropriate number of taps. How well does it work?
11. So far we have assume that all signals are stationary. How quickly can the statistics of the signal change while still allowing LMS to keep up?
12. Linear prediction is a very popular area. There is a whole chapter of Haykin deals with linear prediction. You can predicting a whole host of standard time series data that are available on the internet, e.g. financial time series, chaotic time series.
13. You remember how many assumptions we made to derive properties about LMS. Any theoretical analysis of convergence of LMS or other aspect of adaptive filters makes a good project. Make sure that you are the type of person cut out for this.
14. If you are graduating this semester (and therefore not able to take neural networks in the Fall) or if you already know something about neural networks, you may want to play with some very simple neural networks. You can contrast the linear adaptive filters with simple but nonlinear neural networks.