Cognitive Networks: Towards Self-Aware Networks

9.1. Introduction

The emerging research area of cognitive networks offers a potential solution for dealing with the increasing complexity of communications networks by empowering networks with decision-making capabilities. A key feature of a cognitive network as described in ^[] is the cognitive process, which is responsible for the learning and reasoning that occurs in the cognitive network. There has been little progress in establishing the underlying mechanisms for the cognitive process and the trade-offs involved in selecting and implementing these mechanisms. This chapter attempts to reveal the learning and reasoning mechanisms within the cognitive process as well as some of the design decisions involved. With respect to reasoning and learning methods, we restrict our focus to methods that appear to be applicable to cognitive networks and do not attempt to evaluate all possible methods for reasoning and learning.

Since the same term may have different meanings depending upon who is using it, we first establish our definition of a cognitive network. We adopt the following definition from ^[]: 'a cognitive network is a network with a cognitive process that can perceive current network conditions, and then plan, decide, and act on those conditions. The network can learn from these adaptations and use them to make future decisions, while taking into account end-to-end goals.' The perception, decision, action and learning aspects of this definition are all elements of learning and ...