Engineering and the Environment, University of Southampton

Knowledge and understanding
Having successfully completed the module, you will be able to demonstrate knowledge and understanding of:

• the theory and uses of the z-transform
• a range of digital filter design methods
• high resolution spectral analysis methods
• the concept of nonstationarity and time-frequency analysis

Cognitive (thinking) skills
Having successfully completed the module, you will be able to:

• read, understand and interpret the literature relating to a broader range of DSP techniques
• better appreciate the breadth of DSP algorithms available.

Practical, subject specific skills
Having successfully completed the module, you will be able to

• analyse a digital system's performance
• appreciate the factors tht need to be considered when designing a digital filter
• understand the different principles used to construct non-parametric spectral estimates.

Key transferable skills
Having successfully completed the module, you will be able to comprehend the power and limitations of DSP so that you appreciate where it can be used to enhance system performance.

The aims of this module are to provide the rationale and conceptual bases of some advanced signal processing topics.

• To introduce the student to the analytical and computational methods required for advanced analysis.
• To relate the analysis to applications and interpretation of the results.
• To ensure the student is equipped to understand and use current literature on advanced digital signal processing (DSP).

• Digital Systems
  Auto-regressive (AR) modelling (through population models)
  Moving average (MA) systems
  ARMA processes
• The Z-Transform
  Definition
  Representing the z-plane
  Poles and zeros
  Regions of convergence
  Inversion using, power series, partial fractions and contour integration
• FIR Filter Design
  General filter design issues
  The windowing design method
  Frequency sampling
• IIR Filter Design
  General comments on how to map analogue to digital systems
  Analogue filter designs
  Method of mapping differentials
  Impulse variance
  Bilinear transforms
• High Resolution Spectral Estimation
  Time series models and the contrast between parametric and non-parametric estimation.

Auto-regressive moving-average models and parameter estimation
Auto-regressive models and the Yule-Walker equations; maximum entropy methods; model order determination.
Capon's method.
Eigen-based methods; the MUSIC algorithm.

2 lectures/week
2 x 2h lab classes.

Examples are provided to students in order to practise their analytical skills and these are backed up with interactive tutorial sessions. Students are encouraged to read supporting texts and a booklist is provided.