SBIR-STTR Award

We are proposing novel approaches for implementing a flexible, feasible and interoperable multicast admission control mechanism (McastAC) that is located at the red side of the typical Army black and red networks environment to provide end-to-end QoS assurance: • McastAC decisions are based on resource-friendly network performance estimations, which enable quick reaction to the status change of the network caused by radio dynamics and traffic congestion. • McastAC employs content filtering, where the admission/rejection decisions are based on utility and proximity. As such, together with DiffServ, McastAC ensures the high-priority mission critical applications are protected. The role of McastAC will be presented in the context of real-video applications with strict QoS requirements, in terms of packet delay and losses. In this architecture, we are proposing an innovative Multistream Multicast which ensures fairness between receivers with different capacities. A limited number of streams of the original content are sent over different multicast groups. The video receivers can switch between active multicast sessions by requesting admission to a different multicast group, in a manner that minimizes the disruption of the received video quality. Infocom Technology will leverage their extensive experience in video multi-streaming and multicasting to support this effort.

Keywords:
Multicast Admission Control, Mutistreaming Multicast, Join/Leave Policy, Qos

Our SBIR Phase I effort has focused on formulating an innovative design for the multicast admission control architecture (McastAC) that considers novel approaches for implementing a red side network mechanism to provide end-to-end QoS assurance. The McastAC follows an edge-based admission control architecture that is naturally compatible with tree-based multicast routing, such as Protocol Independent Multicast (PIM) routing protocols and its functionality is largely independent of whether the intermediate network is homogeneous or whether it comprises a single or multiple routing/radio domains (or autonomous systems). McastAC employs content filtering, where the admission/rejection decisions are based on utility and proximity, as well as Multicast multistreaming that maximizes the network utilization and adapts the stream quality to the clients’ capabilities and network conditions. In Phase II, we will develop a detailed specification of the Multicast Admission Decision (MAD) Logic for various operations scenarios. We will also design and verify the network resource performance estimation (NRPE) flow correlation procedure that provides critical inputs to MAD logic and study distributed techniques such as sharing information of local flows with network peers to enhance estimation accuracy. We will iteratively vet the design using a combination of simulation and modeling, emulation, and testbed experiments.

Keywords:
Multicast admission control, tactical red networks, situation awareness