The University of Southampton

ISVR6140 Applied Digital Signal Processing

Module Overview

Signal processing methods are used in many aspects of acoustics and engineering. In this course, you will study several applications where such tools are commonly used. With these applications in mind, new signal processing concepts are introduced and used to tackle a range of engineering problems. Topics include condition monitoring, ultrasound imaging, microphone array processing, the detection, estimation and classification of signals and adaptive methods. Signal processing methods are used routinely to monitor and assess the condition of engineering parts. For example, the vibrations generated by rotating machines can be analysed to detect failing parts such as ball-bearings and gear-boxes, while ultrasound images can be analysed to detect cracks and defects in manufactured components. In many applications, such as in ultrasound imaging or in underwater sonar, we are typically not using a single sensor to record a signal, but use several sensors in a so-called sensor array. These arrays provide additional flexibility, but special array signal processing methods have be used. Using these techniques it then becomes possible, for example, to build microphone arrays that are sensitive only to sound arriving from a specific direction. In nearly all applications, we have to find a way to deal with noise and uncertainties and this is another aspect you will explore in this course. You will learn about methods that can be used to detect signals embedded in noise (for example, this could be used to detect hidden underwater objects from noisy sonar recordings), methods that can be used to estimate signal properties (for example, to estimate the room response from a recording of a rooms response to an impulsive sound) and tools to classify signal into different categories (a good example here could be the automatic classification of different bat species from recordings made of their calls). In many of these settings, analysis has to be performed under variable environmental conditions. For example, noise properties might change during operation of a sensor so that it becomes important for the signal processing methods to automatically adapt to this change.

Aims and Objectives

Module Aims

• Provide details of the application of signal processing methods to a range of application areas from machine condition monitoring to sonar. • Introduce some advanced signal processing concepts and algorithms and their application, including methods for array data analysis, image analysis, signal detection, estimation and classification, and adaptive methods. • Highlight practical issues associated with specific real-world problems. • Create a broader awareness of the set of available signal processing methods and to understand the practical limitation that occur when idealised mathematical models begin to breakdown.

Learning Outcomes

Disciplinary Specific Learning Outcomes

Having successfully completed this module you will be able to:

  • identify and apply appropriate signal processing techniques to analyse signals for specific realworld applications.
  • argue the advantages and limitations of advanced signal processing techniques for specific applications.
  • select and apply standard signal processing techniques for the detection of faults within gearboxes.
  • identify appropriate signal processing tools for the analysis of ultrasound images
  • illustrate key concepts in array processing, including standard techniques for beamforming, direction of arrival estimation and source separation
  • apply fundamental concepts in statistics to problems in signal detection, estimation and classification
  • select and apply appropriate adaptive algorithms for the analysis of signals under changing conditions


Condition monitoring: detection of faults within gearboxes and bearings using vibration data, describing synchronous averaging, resampling, interpolation, order analysis, envelope extraction and the Hilbert transform. Ultrasound image processing for non-destructive testing: basic principles, A-, B- and C-mode imaging, matched-filtering. Signal processing for sonar and microphone arrays: delay and sum beamforming, estimation of the direction of arrival, source separation. Dealing with uncertainties and noise - detection, estimation and classification: the formulation as hypothesis tests, likelihood ratio test, least squares and maximum likelihood estimators, Gaussian classifiers. Signal Processing in changing environments - adaptive methods: optimal filtering, gradient based adaptation, LMS algorithm, linear prediction, linear least squares filtering, recursive least squares.

Learning and Teaching

Teaching and learning methods

Teaching will be in the form of lectures and computer labs. The course is organised into four blocks of three weeks each. • Weeks 1-2, 4-5, 7-8 and 10-11 consist of 4 times 6 45 minute lectures. • Weeks 3, 6, 9, 12 consist of 4 times 150 minutes of computing labs. • The computing labs are the basis for the four assignments.

Follow-up work54
Completion of assessment task60
Practical classes and workshops12
Total study time150

Resources & Reading list

Computer requirements. This course requires access to a suite of computers on which Phython can be run. The computers should have microphones and headsets.


Assessment Strategy



Coursework assignment(s)


MethodPercentage contribution
Coursework assignment(s) 35%
Coursework assignment(s) 35%
Coursework assignment(s) 30%


MethodPercentage contribution
Choice of lab report or coursework assignment 30%
Coursework assignment(s) 35%
Coursework assignment(s) 35%

Repeat Information

Repeat type: Internal & External

Linked modules

Pre- requisites: ISVR2041 Audio and Signal Processing or ISVR6130 Signal Processing.

Share this module Facebook Google+ Twitter Weibo

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.