Current Projects


Smart, Dynamic, and Adaptable Network Design over Software Defined Networks 


The importance of reliable and adaptable networks has become increasingly relevant with the escalation of connectivity in our lives. The growth of streaming of entertainment and development of always online software has created an environment of large data flows that need to be handled efficiently.

Historically this problem has been solved with hardware-based load balancers. Although these hardware-based load balancers provide a solution, they often times are expensive and lack flexibility and scalability. With the use of Software-Defined Networking (SDN), a more dynamic solution can be created to meet network load balancing needs. By separating the control and data planes of a network, SDNs allow for high programmability and adaptability. We, in this work, propose a smart, dynamic and adaptable scheme seeking to utilize network resources more efficiently by identifying traffic patterns and analyzing network metric to dynamically build virtual slices. Our results show that with this approach, we were able to solve the issues with current load balancing techniques by minimizing packet loss, maximizing network link utilization, and efficiently reduce the load on the controller.

Emergency Crowdsourced Automated Adaptive Networking System


Safety and security are major requirements for any community. These criteria in the past have been granted through the work of emergency first respondents whose work relies on accuracy and efficiency. A misevaluated incident could escalate into a larger situation leading to critical events with greater consequence. Therefore to combat this problem accurate information needs to be passed to the right personnel to ensure appropriate measures are taken. This is in fact the weakness of the first response system. Citizens who are often caught in crisis situations are required to relay information to emergency officials to allow for aid. The issue being that under increased stress and danger citizens often struggle to give an accurate account of the situation.

Our Emergency Automated Adaptive Networking System allows for simple live streaming from a citizen's smart device directly to a first response system. This serves to alleviate the pressure from the individual allowing for a more accurate solution for information transfer. Once information is received along with relevant GPS information, respondents can then act accordingly. 

Traffic Pattern Recognition over Software-defined Networks


Recent trends have shown a migration of software from local machines to server-based services. These service-based networks depend on high uptimes and heavy resistance in order to compete in the market. Services such as Netflix and other media streaming outlets cause huge flows of traffic across networks on a daily basis. An outage in one of these services could lead to thousands of customers using the service to drop their plans. Along with this growth of network services, denial of service attacks have equally grown. With a simple set of tools, attackers could bring down one of these services. Defending against these attacks has become increasingly difficult with the growth of Internet of Things and the different varieties of denial of service attacks. For this, our research offers a solution using software-defined networking and real-time traffic pattern recognition using metric based techniques to mitigate a denial of service attack within a smaller time window than other comparable solutions. The use of our method offers both efficient attack handling and also flexibility to fit a variety of implementations. The end result being a network that can automatically adapt against new attacks based on network activity.

 Community Trust Distribution in Vehicle Ad-hoc Networks


Vehicular Ad-hoc Networks (VANETs) have attracted much research and community attention recently due to the benefits that can be provided though this technology. For instance, communication between vehicles can help make the roadway a safer place, where vehicles can work together to share information about road conditions.

As vehicles are equipped with the ability to communicate, the need to develop trust between vehicles is crucial. Trust is referred to as the ability of an agent to accept, with confidence, information from another vehicle (agent). If messages are not trustworthy, then communication between any given agents is futile.

 Cluster-based Dynamic Backup in Cognitive Radio Networks


Recent years have shown an increase demand for wireless communication which creates new challenges in the field where intelligent spectrum utilization is necessary to maintain high network performance. The emergence of Cognitive Radios which provide opportunistic access to the spectrum, introduce an improvement in the spectrum utilization and the communication quality.

Through dynamic spectrum access (DSA), secondary (unlicensed) cognitive users can opportunistically access the primary users’ spectrum when it is free, but have to release the undesignated spectrum when the primary users emerge to provide reliable and efficient communications through the network. To maintain the highest level of quality of service among all the network users (primary and secondary) and to avoid any interruptions occurred by the primary users needed to use their licensed spectrum, we, in this work, aim to introduce a hybrid scheme in which the secondary users find options for a backup path, but they only solidify that path if they must vacate their original path. Our numerical results showed that through our proposed scheme, we were able to enhance the goodput and provide a higher satisfied ratio through dynamically backing up secondary users requests.