5 ECTS credits / 135 hours of work
|Language of instruction
The course is held in the autumn semester, during period II. For master students of Computer Science and Engineering specializing in Computer Engineering, it is recommended to complete the course at the first autumn semester.
Learning outcomes of the course are:
- The student understands common real number formats used in digital signal processing.
- The student can implement a digital filter using fixed-point computations. He can optimize word lengths so that the required performance goals are fulfilled.
- The student knows the CORDIC algorithm and can utilize it in the implementation of function and transform (e.g. DCT) computations.
- The student knows the principles, which allow computationally efficient implementation of decimation and interpolation operations. Related to this, he can implement narrow-band digital filters.
- The student can explain how a modulated filter bank works and knows its polyphase decomposition based implementation.
- The student can implement convolution for long data sequences and filters. He also knows, how the same principles are used in the implementation of correlation.
- The student can explain the general operational principles of adaptive filters and knows some of their applications. He knows operation of some common adaptive algorithms. He can study behaviour of adaptive filters with simulation.
Some exercise tasks of the course are done in the Matlab environment utilizing also its Simulink tool. The student learns how it can be used in the modelling of signal processing systems.
Fixed-point and floating-point arithmetics, fixed-point filter implementation, CORDIC, DCT, FFT, polyphase decomposition, multirate signal processing, modulated filter banks, sectioning, adaptive filters and algorithms, Matlab and Simulink tools in DSP modelling.
|Mode of delivery
The tuition will be implemented as face-to-face teaching and web-based teaching. Moodle is used as the learning environment.
Due to Covid-19 pandemic, teaching in Autumn 2020 will be implemented remotely. Details of arrangement can be found from the course web page, which will be available from October 16 in Moodle.
|Learning activities and teaching methods
Lectures 28 h / Group work 42 h / Self-study 65 h. The group work consists of six weekly design tasks.
The course is primarily targeted to the students of Computer Science and Engineering specializing to Computer Engineering.
|Prerequisites and co-requisites
A recommended prerequisite is the completion of "521337A Digital Filters".
|Recommended optional programme components
The course is an independent entity and does not require additional studies carried out at the same time.
|Recommended or required reading
The course provides lecture notes for reading. In addition, the following books provide useful information:
E.C. Ifeachor, B.W. Jervis. Digital Signal Processing - A Practical Approach. Second Edition. Prentice-Hall, 2002.
W.T. Padgett, D.V. Anderson. Fixed-Point Signal Processing. Morgan&Claypool Publishers, 2009.
|Assessment methods and criteria
The course uses continuous assessment, which is based on evaluation of the weekly group works and exams arranged during lectures.
The course utilizes a numerical grading scale 0-5. In the numerical scale zero stands for a fail.
|Working life cooperation
The course does not contain working life cooperation. There may be guest lectures.
The web page of the course arranged at Autumn 2020 will be target=_blank>https://moodle.oulu.fi/course/view.php?id=3212