Overload control strategies for distributed communication networks

Abstract: In distributed communication networks several nodes interact to carry out services. New telecommunication networks support a variety of services, much different from traditional telephony. Each service has its own so called load profile, which means that the amount of capacity needed to complete a service varies. The services may also have different priorities and quality of service requirements. This thesis discusses overload control strategies for distributed communication networks. Further, a number of overload control schemes that carry out a certain strategy are developed and examined. The main objectives of the overload control schemes are to efficiently protect a resource from overload, to maintain a high throughput and to ensure a fair distribution of the resource's capacity among the services. Two network architectures are examined; the Intelligent Network (IN) and the Telecommunication Information Networking Architecture (TINA). For IN, the thesis concentrates on the overload control of Service Control Points (SCPs), Service Switching Control Points (SSCPs) and Intelligent Peripherals (IPs). These nodes are examined with both simulation and analytical methods. For TINA, a performance simulation model is developed. Further, the overload control of TINA networks is discussed. The thesis also analytically analyses a feedback queuing system with control theoretic methods. A transfer function for the system is developed when a modified PD controller is used in the overload control.