IEEE Transactions on Big Data

Expand your horizons with Colloquium, a monthly survey of abstracts from all CS transactions!

From the October-December 2017 issue

Graph Regularized EEG Source Imaging with In-Class Consistency and Out-Class Discrimination

By Feng Liu, Jay Rosenberger, Yifei Lou, Rahilsadat Hosseini, Jianzhong Su, and Shouyi Wang

Featured article thumbnail image EEG source imaging integrates temporal and spatial components of EEG to localize the generating source of electrical potentials based on recorded EEG data on the scalp. As EEG sensors can't directly measure activated brain sources, many approaches were proposed to estimate brain source activation pattern given EEG data. However, since most part of the brain activity is composed of the spontaneous non-task related activations, true task caused activation sources will be corrupted in strong background signal. For decades, the EEG inverse problem was solved in an unsupervised way without any utilization of the label information that represents different brain states. We propose that by leveraging label information, the task related discriminative sources can be much better retrieved among strong spontaneous background signals. A novel model for solving EEG inverse problem called Laplacian Graph Regularized Discriminative Source Reconstruction which aims to explicitly extract the discriminative sources by implicitly coding the label information into the graph regularization term. The proposed model can be generally extended with different assumptions. The extension of our framework is applied to VB-SCCD model which aim to estimate extended brain sources by including a spatial total variation regularization term. Simulated results show the effectiveness of the proposed framework.

download PDF View the PDF of this article      csdl View this issue in the digital library

Editorials and Announcements


  • In order to promote timely publication of regular paper submissions, please note that TBD is not currently accepting proposals for new special issues until the existing publication queue has been cleared.
  • TBD is pleased to participate in a free trial offering of the new IEEE DataPort data repository, which supports authors in hosting and referring to their datasets during the article submission process. Learn more about this exciting opportunity.
  • We're pleased to announce that Qiang Yang, head of the Huawei Noah's Ark Research Lab and a professor at the Hong Kong University of Science and Technology, has accepted the position of inaugural Editor-in-Chief beginning 1 Jan. 2015. Read more.


Guest Editorials

Call for Papers

Special Issue on Big Data from Space

Extended submission deadline: February 28, 2018. View PDF.

Big Data from Space refers to the massive spatio-temporal Earth and Space observation data collected by space-borne sensors, and their use in synergy with data coming from other aerial or ground based sensors or sources. This domain is currently facing sharp development with numerous new initiatives and breakthroughs ranging from computational sensors to space sensor web, covering almost the entire electromagnetic spectrum from Gamma-rays to radiowaves, or from gravitational to quantum principles. The analysis of these data largely contributes to the broad scientific effort to understand the Universe and to enhance life on Earth. The recent multiplication of open access initiatives to Big Data from Space is giving momentum to the field by widening substantially the spectrum of scientific communities and users as well as awareness among the public while offering new benefits at all levels from individual citizens to the whole society.

In this Special Issue, we solicit high-quality scientific research articles, in areas such as, but not limited to, Earth Observation, planetary sciences, Space and Security, deep space exploration, astrophysics, satellite telecommunication, navigation and positioning systems, addressing key challenges and innovative solutions on how Big Data paradigms can improve the space sciences, technologies, and applications.

Special Issue on Edge Analytics in the Internet of Things

Extended submission deadline: March 1, 2018. View PDF.

The cloud-based Internet of Things (IoT) that connects a wide variety of things including sensors, mobile devices, vehicles, manufacturing machines, and industrial equipments, etc. is changing the way we live. IDC forecasts that the IoT will grow to 50 billion connected devices by 2020, and will generate an unprecedented volume and variety of data. However, moving this big volume of data from the network edge to a central data center for processing and analysis not only adds latency but also consumes network bandwidth. Therefore, the cloud-based IoT with a centralized data center may not be able to enable smart environments, such as cities, homes, schools, etc., or smart systems, such as automated vehicles, traffic controls, factories, etc., whose data need to be analyzed and acted on quickly. This is especially true in scenarios such as health monitoring or autopilot, where milliseconds can have fatal consequences. Such demand indicates that data processing and analysis has to be performed where the data are collected or generated instead of waiting for the data to be sent back to the centralized data center. Also, often these smart environments or systems need to be capable of self-monitoring, self-diagnosing, self-healing, and self-directing, and thus the task of edge-based data analytics may need to incorporate the technology of machine learning. Thus, there is a need to find a way to push intelligence from the central data center to the edge of the network. Indeed, IDC also predicts that up to 40% of IoT data will need edge-based analytics for applications that need real-time action. To solve this issue, fog computing, in which a set of interconnected micro data centers, called fog nodes, are deployed in between the things and the cloud data center, has been adopted as a bridge linking IoT devices and their remote data center. Since a fog node can run IoT-enabled applications for real-time data analytics with millisecond response time, fog computing enables application services of the IoT to be performed close to their consumers, and has created an emerging technology { edge analytics. Meanwhile, some IoT things are getting more capable and more powerful, making edge-based analytics possible. On the other hand, for the moment, most of the IoT things still do not have the computing and storage resources to perform intelligent analytics directly. For such IoT things, a nearby fog node or cloudlet may perform the tasks on their behalf. Furthermore, since data sources are widely distributed, some analytics tasks may need to be collaboratively performed by a set of fog nodes working together with some IoT things. As such, orchestrating fog nodes by means of topology control and network function virtualization may leverage the edge analytics performance.

Though edge analytics is in its nascent stage, it is getting more and more popular. The goal of this special issue is to provide a forum for researchers working on IoT and fog computing to present their recent research results in edge analytics

General Call for Papers

TBD Call-for-Papers Flyer Version 1 (PDF)

TBD Call-for-Papers Flyer Version 2 (PDF)

Reviewers List

Annual Index

Access Recently Published TBD Articles

RSS icon Subscribe to the RSS feed of recently published TBD content

mail icon Sign up for e-mail notifications through IEEE Xplore Content Alerts

preprints icon View TBD preprints in the Computer Society Digital Library

TBD is financially cosponsored by:

IEEE Computer SocietyIEEE Communications SocietyIEEE Computational Intelligence SocietyIEEE Sensors CouncilIEEE Consumer Electronics Society


IEEE Signal Processing SocietyIEEE Systems, Man, & Cybernetics SocietyIEEE Systems CouncilIEEE Vehicular Technology Society


TBD is technically cosponsored by:

IEEE Control Systems SocietyIEEE Photonics SocietyIEEE Engineering in Medicine & Biology SocietyIEEE Power & Energy SocietyIEEE Biometrics Council