engleski

An agent-based model for the adaptation of processing efficiency for prioritized traffic

The increasing number and variety of telecommunication services being offered by networks have emphasized the demand for optimized load management strategies. Obtainable processing efficiency is one of the key elements for all services and has to be dominantly supervised. Different priority levels and classifications can co-exist to support services, such fine regulation of precedence mixed with hard and destructive priority for critical service classes. Intelligent processing efficiency management with adaptive protection of predefined processing power can support such requests. This paper deals with the regulation of packets moving towards service- processing resources. Regulation is based on service priorities and processing load in the agent controlled part of the network.

Intelligent agent; reinforcement learning; Q-learning

nije evidentirano