This course introduces the concepts and techniques of digital signal processing (DSP) and demonstrates some interesting and useful practical applications of DSP. It also provides practical experience in using Python in analysis and design of DSP systems and algorithms. Taught material is programming language agnostic ensuring that the skills and knowledge acquired remain relevant even as underlying implementation technologies evolve. However, this material is generously supplemented with reference implementations of the algorithms described written in the Python programming language. The course concludes with a free project where students are encouraged to connect sensor modules to a microcontroller board to achieve an aim of their choice, then, using the signal processing techniques acquired in the course, to design, implement and test a digital filter which demonstrably satisfies their stated requirements.

2 hours of flipped classroom teaching per week: Online videos and extensive lab session with problem-based learning.

None.

None.

50% Examination

50% Assignment (3 assignments at 12.5%, 12.5% and 25%)

It is the default expectation that all courses will offer opportunities for reassessment or deferred assessment. Where it is not possible to offer this in some assessment components, the grade achieved at the first attempt will be counted towards the final course grade, and any exceptions for this course are described below.

[No exceptions]

The aims of this course are to:

■ Introduce the concepts and techniques of digital signal processing (DSP);

■ Demonstrate some interesting and useful practical applications of DSP;

■ Provide practical experience in using DSP software in analysis and design of DSP systems and algorithms;

■ Develop energy-efficient implementations of DSP systems by understanding the impact of different implementation choices;

■ Integrate interdisciplinary knowledge and skills into DSP such as audio, medical and communications;

■ Understand safety issues of the interaction between humans and DSP systems and to mitigate risk in this context;

■ Apply the theory, concepts and techniques learned in the lecture material to the solution of incompletely specified ("real world"), complex problems;

■ Evaluate the suitability of various technologies in the acquisition and implementation of DSP solutions;

■ Use empirical techniques to distil a requirement from observed data and by understanding its relationship to the desired outcome;

■ Demonstrate the importance of testing whether system requirements have been satisfied and to introduce methods to perform such testing;

■ Develop the participants capacity to function effectively as part of a team;

■ Communicate outcomes and achievement by both formal, technical, analytical reporting and in a manner appropriate to non-technical audiences.

By the end of this course students will be able to:

■ use the Fourier transform to filter signals, such as audio, biomedical and communication signals

■ critically evaluate literature concerning DSP algorithms and their application

■ design and create FIR filters from a desired frequency response arising from complex interdisciplinary projects

■ design and create IIR filters on the basis of an analogue prototype and apply them in complex low latency applications

■ design and create matched filters for biomedical, communication and other complex interdisciplinary problems

■ create system specifications and requirements analysis of a product in both the time domain (latency) and frequency domain (frequency response)

■ make ethical choices when processing human data

■ make decisions on the safety of laboratory equipment particularly when interacting with human subjects

■ compare and judge different DSP methodologies in the solution of a given "real world" design challenge

■ create optimal filters for best performance in a given problem including due consideration to efficiency and the capacity to implement the solution on the minimal hardware

■ use Python as a filter design tool

■ create object oriented DSP processing code in Python, thus promoting reuse and sustainability

■ critically evaluate the performance of such solutions in terms of satisfying product requirements

■ test and measure the capacity of a DSP system to satisfy latency requirements

■ analyse interdisciplinary problems and to provide a solution to the problem

■ formally report the DSP deliverable using the highest standard of graphical and written content

communicate the benefits of their work to a non-technical audience through multimedia