Initials Areas of Research
Network Congestion Overview
Industry is moving towards Internet Protocol (IP) technology for all its telecommunications applications. When IP networks become overloaded, packets get dropped and other Quality of Service (QoS) measures like packet latency and jitter are significantly degraded. At some point, the quality of service for voice and video packets becomes poor enough that their communication is lost. As Voice over IP (VoIP) and Video becomes more prevalent in these networks, these issues become increasingly important – particularly, since there is no dedicated communication path for a voice or video call. There are three basic queue disciplines being used to ensure packets from various sources receive their desired QoS. They are First Come First Served (FCFS), Priority Queueing (PQ), and Class Based Weighted Fair Queueing (CBWFQ). Low Latency Queueing is a combination of Priority Queueing and Class Based Weighted Fair Queueing and is currently being used to ensure voice and video traffic receives the required QoS, while allowing the non-real time traffic to also access the network bandwidth. The packets that require near real time service, typically Voice and Video, are served using Priority Queueing since these have very stringent QoS requirements. Additional classes of traffic are being proposed in the Internet Engineering Task Force for emergency traffic, and these would likely be treated in the priority queues. The remaining classes are serviced via CBWFQ. In the near term our research activity will focus on the development of modeling and simulation capabilities to determine the performance of voice, video, data, and proposed emergency traffic streams under the LLQ discipline.