2.1 Introduction

Current data networking technology limits a network s ability to adapt, often resulting in sub-optimal performance. Limited in state, scope, and response mechanisms, the network elements (consisting of nodes, protocol layers, policies, and behaviors) are unable to make intelligent adaptations. Communication of network state information is stifled by the layered protocol architecture, making individual elements unaware of the network status experienced by other elements. Any response that an element may make to network stimuli can only be made in the context of its limited scope. The adaptations that are performed are typically reactive, taking place after a problem has occurred. In this chapter, we advance the idea of cognitive networks, which have the promise to remove these limitations by allowing networks to observe, act, and learn in order to optimize their performance.

Cognitive networks are motivated by complexity. Particularly in wireless networks, there has been a trend towards increasingly complex, heterogeneous, and dynamic environments. While wired networks can also take on any of these characteristics (and are not excluded from potential cognitive network applications) wireless networks are a natural target because of their inter-node interactions and the size of their system state space. Previous research into cognitive radio and cross-layer design have addressed some of these issues but have shortcomings from the network perspective. Cognitive networks represent a new scope and approach to dealing with this complexity.

This chapter provides the...