Signal Processing - January 2017 - 83

Nodal
Time Series
Time
Windows
r = 0.62 r = 0.35 r = 0.28 r = 0.31 Pairwise
Coherence
Connectivity
Network
Brain Network
Nodes

Averaged
Network
(a)

Average
Network

Graph Signal Representation of Nodal Time Series
t
t
t +1
t +1
(b)

Figure 1. (a) A brain network representation using average functional coherence values. (b) A brain
graph signal using regional fMRI data for each time point t. See the original figure and report at
https://sigport.org/documents/graph-frequency-analysis-brain-signals.

(a)

effective means to introduce the relationship between data and information,
commonly achieved through the use of
alternative representation of data, the
project is designed to explore the relationship between neuroscience and
graph Fourier transform (GFT). The notion of the GFT and graph filters is presented to decompose a given subject's
brain signal into sections that represent

One of these projects, "Graph Frequency Analysis of Brain Signals," by
Leah Goldsberry, Weiyu Huang, and Dr.
Alejandro Ribeiro with the Department
of Electrical and Systems Engineering in
the University of Pennsylvania, Philadelphia (http://arxiv.org/pdf/1512.00037v2
.pdf), is evaluating the practice of graph
signal processing methods in neuroscience. Motivated by the need to identify

be instrumental in enhancing students'
hands‐on experience during undergraduate studies while building solid theoretical foundations. A new article series and
contributions platform has been developed by a team of guest editors for IEEE
Signal Processing Magazine who wish
to promote the sharing of experiences
and best practices with undergraduate
studies, engineering projects, research,
and innovation in diverse signal and information processing applications. Students and advisers have been invited to
contribute information and practices in
engineering design projects in SigPort
(https://sigport.org/events/spm-studentdesign-project-series#citation-ieee). The
goal of this initiative is to start a discussion on the role of experimental and
project-based practices in modern signal and data processing education and
support fast-track progress by sharing
"know-how" experience. The SigPort
submissions give a first glimpse into the
potential of this idea. With more than
400 downloads within a time span of
just a few weeks, it seems that there is a
need within the signal and information
processing community that should be
further explored and advanced.
The projects submitted as the first
wave of response to the call for contributions articulate the field's diversity
and the multitude of applications, methods, and hardware and software tools
explored by students and faculty.

0.08

(b)

(c)

Figure 2. An example distribution of decomposed signals across all brain regions for the first experiment. Average energy with respect to (a) xL, (b) xM, and
(c) xH. A thresholding is applied.
IEEE Signal Processing Magazine

January 2017

http://www.arxiv.org/pdf/1512.00037v2 https://www.sigport.org/events/spm-student https://www.sigport.org/documents/graph-frequency-analysis-brain-signals

Table of Contents for the Digital Edition of Signal Processing - January 2017