A Time-Vertex Signal Processing Framework: Scalable Processing and Meaningful Representations for Time-Series on Graphs

A Time-Vertex Signal Processing Framework

An emerging way to deal with highdimensional non-euclidean data is to assume that the underlying structure can be captured by a graph. Recently, ideas have begun to emerge related to the analysis of time-varying graph signals. This work aims to elevate the notion of joint harmonic analysis to a full-fledged framework denoted as Time-Vertex Signal Processing, that links together the time-domain ...

Stationary time-vertex signal processing

The goal of this paper is to improve learning for multivariate processes whose structure is dependent on some known graph topology. Typically, the graph information is incorporated to the learning process via a smoothness assumption postulating that the values supported on well connected vertices exhibit small variations. We argue that smoothness is not enough. To capture the behavior of comple...

A Scalable Multiprocessor for Real-Time Signal Processing

Multimedia applications that operate on data such as audio, video or graphical data are showing a growing need for processor power. To meet the required performance, often specialized processors and custom hardware architectures are used. Our goal was to design a scalable multiprocessor that is versatile enough to replace many of these custom solutions for applications that perform signal proce...

Windowing Effects of Short Time Fourier Transform on Wideband Array Signal Processing Using Maximum Likelihood Estimation

During the last two decades, Maximum Likelihood estimation (ML) has been used to determine Direction Of Arrival (DOA) and signals propagated by the sources, using narrowband array signals. The algorithm fails in the case of wideband signals. As an attempt by the present study to overcome the problem, the array outputs are transformed into narrowband frequency bins, using short time Fourier tran...

Signal processing and neural network models for economic time series