Cognitive Networks: Towards Self-Aware Networks

2.7. Conclusions

This chapter has provided a brief introduction to, and motivation for, cognitive wireless networking from a network management viewpoint. Autonomic computing principles were described by first examining IBM's popular autonomic computing architecture. This led to a more in-depth understanding of why existing autonomic approaches were not completely suitable for network management. We then focused on enhancements to the IBM architecture. This built up into an examination of a novel autonomic networking architecture called FOCALE. This architecture was designed specifically to support network management. Specific attention was given to its set of novel features, including the use of multiple control loops, information and data models, ontologies, context awareness and policy management.

FOCALE is a closed loop system, in which the current state of the managed element is calculated, compared to the desired state, and then action taken if the two states aren't equal. Hence, FOCALE uses separate control loops – one for when the state of the managed element is equal to (or converging towards) the desired state, and one for when it isn't. This protects the business goals and objectives of the users as well as the network operators.

FOCALE uses knowledge engineering techniques to manage legacy as well as new (as in inherently autonomic) components. FOCALE orchestrates the behavior of systems and components by using predefined finite state machines. The finite state machines ...