Breadcrumb
Wavelet transforms on networks
Supervisor: Guy Nason
Theme: Multiscale Methods
Over the last decade wavelet transforms on 'regular' time series or 'regular' gridded images have proved to be particularly powerful tools for compression and denoising purposes. Recently, through the lifting transform, multiscale methods have been adapted to a wide range of data structures including networks. Several potential projects are available using the 'wavelet transform on a network' with application areas in transport modelling and forecasting, social networks and multivariate networks.