Influence of Artificial Intelligence on Cognitive Radio Applications

Introduction

Cognitive Radio Network (CRN) based technology deployment is a promising new application area of wireless mobile communication to provide many industrial services. Historically CRN technology is deeply rooted in early cellular and personal communication systems. Cognitive radio research and applications were envisioned in the late 1990s (Mitola, 2000). The regulatory bodies (e.g., Office of Federal Communications (OFCOM) Commission in the United States of America) regulate technical communications issues for commercial applications. For example, some well-known industrial applications include public safety systems (e.g., fire control, flood-related disaster management, earthquake relief operation service, transport network management, and healthcare service operations). In these applications, a large portion of certain licensed frequency bands often remain unused. Therefore, to address the new radio spectrum issue, a new policy would allow devices that can sense and adapt to their spectral environment, such as cognitive radio to become secondary users. Such users are wireless devices that opportunistically use the spectrum already licensed to the primary users. The primary users are generally associated with the primary radio spectrum licensed holder and thus have a higher priority right to the spectrum. The intuitive objective behind secondary spectrum licensing is to improve the spectrum use efficiency of the network, depending on the type of licensing, without affecting higher-priority users.

Spectrum sensing, spectrum decision, spectrum sharing, and spectrum mobility are the four main characteristics of cognitive radio systems. Spectrum sensing helps to observe the spectrum occupancy status and recognize the available ability, while cognitive radio users dynamically access the available channels through the regulation processes of spectrum decision, spectrum sharing, and spectrum mobility. In order to mitigate the processing delays required in these four functions and to improve the efficiency of spectrum utilization, spectrum prediction for cognitive radio networks has been extensively studied in the research literature (Neel, 2007) (Haykin, 2005) (Mitola, 2000). Besides, one of the essential characteristics of CRN is opportunistic spectrum access (OSA) to mitigate the mobile communication spectrum scarcity-related issues. In CRNs, opportunistic spectrum access relates to the wireless communications paradigm in which the communicating parties dynamically exploit the radio spectrum band not utilized by the primary service licensed to operate over such a band.

Figure 1.

Communication model for cognitive radio systems

The central component of CRNs, hence OSA, is the cognitive radio transceiver. In a cognitive radio operating environment, a wireless device senses the surrounding radio environment and opportunistically accesses the unutilized spectrum band(s), relying on examining the activities of the primarily licensed networks. In this way, to enhance the usage of the radio range, a cognitive radio hub detects the weather, evaluates the open-air qualities, and then makes certain decisions and distributes the executives' space assets. The cognitive radio works in tandem with artificial intelligence (AI) techniques (e.g., fuzzy logic, neural networks, genetic algorithm, genetic programming, case-based reasoning, Bayesian network-based learning paradigm) to provide a flexible and intelligent allocation of resources for its users. The central objective of this chapter is to highlight the various artificial intelligence techniques used in cognitive radio operations to enhance cognition capabilities in CRNs and present a review of the subject, which includes the typical learning challenges emerging in cognitive radio systems.