WiSense Seminar #91″Local Control Based Cognitive Radio Ad Hoc Networks”, Peng Hu, “Using Cognitive Networking” Amr El-Mougy 24 Jan 2012
Title: WiSense Seminar #91″Local Control Based Cognitive Radio Ad Hoc Networks”, Peng Hu, “Using Cognitive Networking” Amr El-Mougy 24 Jan 2012
Location: uOttawa, SITE Building (800 King Edward) Room: 5084
Description: Title: Local Control Based Cognitive Radio Ad Hoc Networks
Abstract: As a result of the development of cognitive radio technology, the concept of cognitive radio ad hoc networks (CRAHNs) has recently been proposed in the literature, which creates more research challenges than those in classical cognitive radio networks (CRNs). These research challenges in CRAHNs are due to the variable radio environments caused by spectrum-dependent communication links, hop-by-hop transmission, and changing topology. This study focuses on important research topics in spectrum management in scalable CRAHNs driven by local control schemes, such as spectrum sharing, allocation, and mobility. To conduct our study, a local control approach is proposed to enable system-level analysis and protocol-level design with distributed protocols for spectrum sharing. In the local control approach, we can evaluate the system dynamics caused by either protocol-specific parameters or application-specific parameters in CRAHNs, which is difficult to explore using the existing methods. Moreover, combining the performance analysis based on local control concept, we can design new distributed protocols based on the features of the medium access control layer and the physical layer. The proposed research themes and related research topics surrounding local control approach are also discussed.
BIO Notes: Peng Hu received the B.S. degree in information engineering and the M.S. degree in communication and information systems, both from Wuhan University of Technology, Wuhan, China, in 2004 and 2007, respectively. He is currently working towards his Ph.D. degree at the Department of Electrical and Communication Engineering, Queen’s University, Kingston, Canada. His current research interest is in the field of cognitive radio ad hoc and sensor networks.
Speaker 2: Amr El-Mougy, Queen’s University
Title: Using Cognitive Networking to Achieve the End-to-End Goals of Wireless Sensor Networks
Abstract: Cognitive networking is a paradigm that promises to provide a holistic perspective to achieving the end-to-end goals of Wireless Sensor Networks (WSNs). Using aspects of human intelligence such as reasoning and learning, cognitive networking enables efficient resource utilization and adaptive and dynamic network operation. Thus, new challenging applications can be supported in WSNs that were hitherto not feasible. In this talk, a novel framework for cognitive networking is discussed. This framework consists of a reasoning machine designed using the mathematical tool known as Weighted Cognitive Maps (WCMs) and a learning protocol that utilizes the Q-Learning algorithm.
WCM is a mathematical tool that has powerful inference capabilities. Every end-to-end goal, environment variable, or process in the system is represented as a vertex in the WCM, and edges connect the vertices that are causally related. This enables the WCM to consider conflicting interactions and multiple goals. Furthermore, the inference capabilities allow the WCM to determine the appropriate actions to be taken using simple mathematical operations, requiring only knowledge of the causal relationships between concepts. Two case studies are proposed that illustrate the capabilities of WCM and their flexibility in supporting different application requirements and network types. In addition, an elaborate theoretical model based on several interacting Markov Chains (MC) is proposed to analyze the operation of the WCM system. Extensive computer simulations and analytical results show the ability of the WCM system to achieve the end-to-end goals of the network and find compromises between conflicting constraints. The results also show that that the WCM system outperforms its existing counterparts and is able to adapt to changing network conditions and application requirements.
On the other hand, Q-Learning is a well known reinforcement learning algorithm that is used to evaluate the actions taken by an agent over time. Thus, it is used to design a learning protocol that improves the performance of the WCM system. Q-Learning achieves this task by building a knowledge base that the WCM uses to make decisions about the parameters to be used. Furthermore, to ensure that the learning protocol operates efficiently, methods for improving the learning speed and achieve distributed learning across multiple nodes are proposed. Extensive computer simulations show that the learning protocol improves the performance of the WCM system in several metrics.
BIO Notes: Amr El Mougy is a PhD candidate at Queen’s University working at the Wireless Communications and Signal Processing Lab (WISIP), under the supervision of Dr. Mohamed Ibnkahla. He received his M.Sc. degree from Concordia University in 2006. He has authored more than a dozen journal and conference publications as well as two book chapters. He has also co-supervised several graduate and undergraduate students. In addition, he was also the coordinator for project titled “Versatile Wireless Sensor Network for Environmental Monitoring”, which involved the design and implementation of a WSN. He was also a member of the teams working on projects that involved a WSN for highway safety and a WSN for smart grids.
Start Time: 10:00
End Time: 12:00