SCALABLE AND QoS NETWORKING SOLUTIONS FOR TELEMEDICINE

Abstract

Retrieving data from a patient in real-time is a challenging operation, especially when requiring information from the network to support the patient’s health. A real-time healthcare system process is conducted with a continual input, processing, and output of data. It needs to have the ability to provide different priorities to different applications, users, or data flows, or to guarantee a certain level of performance to a data flow. The current Internet does not allow applications to request any special treatment. Every packet, including delay-sensitive audio and video packets, is treated equally at the routers. This simplest type service of network is often referred to as best effort, a network service in which the network does not provide any guarantees that data is delivered or that a user is given a guaranteed QoS level or a certain priority. Providing guaranteed services requires routers to manage per-flow states and perform per-flow operations. Such network architecture requires each router to maintain and manage perflow state on the control path, and to perform per-flow classification, scheduling, and buffer management on the data path. This complicated and expensive network architecture is less scalable and robust than today’s modern stateless network architectures such as Random Early Dropping (RED) for congestion control, DiffServ for QoS, and the original IP network. This thesis introduces a new DiffServ-based scheme of IP bandwidth allocation during congestion, called Proportional Allocation of Bandwidth (PAB) which can be used in all networks. In PAB scheme, the bandwidth is allocated in proportion to Subscripted Information Rate (SIR) of the competing flows. PAB implementation uses multiple token buckets to label the packets at the edge of the network and multilevel threshold queue at the IP routers to discard packets during congestion.