Rasit Eskicioglu

Work place: Department of Computer Science, University of Manitoba, Winnipeg, R3T 2N2, Canada

E-mail: rasit.eskicioglu@cs.umanitoba.ca


Research Interests: Operating Systems


Dr. Rasit Eskicioglu is an Associate Professor at computer science department, University of Manitoba, Canada. His research interests are primarily in Experimental Systems. He is interested in Computer systems; Systems software; Operating systems; Distributed, cluster and grid computing; High speed network interconnects, Mobile networks, and Pervasive computing. Most recently, I am looking at wireless sensor networks (WSNs) and their applications to real world problems; Indoor localization, monitoring, and tracking; and Internet of Things (IoT).

Author Articles
Distributed Intelligence Model for IoT Applications Based on Neural Networks

By Baha Rababah Rasit Eskicioglu

DOI: https://doi.org/10.5815/ijcnis.2021.03.01, Pub. Date: 8 Jun. 2021

Increasing the implication of IoT data puts a focus on extracting the knowledge from sensors’ raw data. The management of sensors’ data is inefficient with current solutions, as studies have generally focused on either providing cloud-based IoT solutions or inefficient predefined rules. Cloud-based IoT solutions have problems with latency, availability, security and privacy, and power consumption. Therefore, Providing IoT gateways with relevant intelligence is essential for gaining knowledge from raw data to make the decision of whether to actuate or offload tasks to the cloud. This work proposes a model that provides an IoT gateway with the intelligence needed to extract the knowledge from sensors’ data in order to make the decision locally without needing to send all raw data to the cloud over the Internet. This speeds up decisions and actions for real-time data and overcomes the limitations of cloud-based IoT solutions. When the gateway is unable to process a task locally, the data and task are offloaded to the cloud.

[...] Read more.
Other Articles