Abstract-Achieving ever-growing Quality of Service (QoS) requirements for business customers is a major concern over the current Internet. However, presently, its architecture and infrastructures are inflexible to meet the demand of increased QoS requirements. OpenFlow, OF-Config (OpenFlow Configuration and Management protocol), and OVSDB (Open vSwitch Database Management protocol) protocols are well-known software defined networking (SDN) technologies for the Future Internet, enabling flexibility by decoupling the control plane from networking devices. In this paper, we propose a QoS framework using the SDN technologies and test the framework in failure-conditions using single and multiple autonomous system scenarios of the current Internet. We show that an effectively high QoS can be achieved for business customers using our framework.

Improving inventory management is essential to retailer profitability. This paper proposes a supervised learning approach for Out-of-Stock (OOS) detection by Texture, Color and Geometry features in high-resolution panoramic images of grocery retail shelves. Cascade classifiers are used to detect labels that can potentially be used to confirm the presence of the OOS cases. The image acquisition setup includes a camera cart that shoots from multi-viewpoints aiming a parallel motion to the shelf. The correction of perspective distortion is applied to handle the different camera translation motions while stitching together images with a high-level of similarity. From the generated panoramas, the proposed OOS detection is followed by classification with Support Vector Machines. The experimental tests were performed throughout the retail environment with real data obtained from supermarket shelves containing labels near the visible ruptures. Results show a detection accuracy of 84.5% for OOS and a sensitivity of 86.6% for label detection

Abstract-The evolution of Software-Defined Networking and the overall acceptance of protocols such as OpenFlow, demonstrates the added value of decoupling the data plane from the control plane. Existing SDN Controllers enable the expected flexibility from such networks by dynamically providing a finegrained control of each flow. However, hardware-specific configurations, such as the creation of queues or other mechanisms is out of the scope of these controllers. This work presents an extension to a well known OpenFlow controller (Floodlight) to efficiently handle the management of Traffic Control Queues in OpenFlow switches, resorting to a RESTful northbound interface. The obtained results demonstrate further possibility of developing innovative on-demand resource reservation mechanisms in SDN without adding unbearable overheads.

Abstract-Software defined Networking (SDN) such as OpenFlow decouples the control plane from forwarding devices and embeds it into one or more external entities called controllers. We implemented a framework in OpenFlow through which business customers receive higher Quality of Service (QoS) than besteffort customers in all conditions (e.g. failure conditions). In the demonstration, we stream video clips (business and best-effort customer's traffic) through an emulated OpenFlow topology. During the demonstration, we trigger a failure in the paths of video clips and show an effectively higher QoS for business customers when compared against best-effort customers. This is demonstrated by simply watching the video clips at the receiver.

No abstract