Role of Artificial Intelligence in Cognitive Radio Networks

The role of artificial intelligence techniques and its impact in context of cognitive radio networks has become immeasurable. Artificial intelligence redefines and empowers the decision making and logical capability of computing machines through the evolutionary process of leaning, adapting, and upgrading its knowledge bank accordingly. Significant functionalities of artificial intelligence include sensing, collaborating, learning, evolving, training, dataset, and performing tasks. Cognitive radio enables learning and evolving through contextual data perceived from its immediate surrounding. Cognitive science aims at acquiring knowledge by observing and recording externalities of environment. It allows self-programming and self-learning with added intelligence and enhanced communicational capabilities over wireless medium. Equipped with cognitive technology, the vision of artificial intelligence gets broadened towards optimizing usage of radio spectrum by accessing spectrum availability, thereby reducing channel interferences while communication among licensed and non-licensed users.

Cognitive Radio Networks

Cognitive radio (CR) offers a novel way for effective usage of wireless spectrum by using dynamic spectrum sensing and allocation. One of the main components of CR is to find a spectrum hole for data transmission. Spectrum hole can be found by using spectrum sensing, a geolocation database, or by using a beacon signal. In this chapter, the authors describe algorithms for spectrum sensing in the presence of both additive white Gaussian and colored Gaussian noise. The algorithms include blind, non-blind, and cooperative sensing-based methods. The authors have compared the performance of various methods for IEEE 802.22 standard (which is the first standard incorporating CR).

Cognitive Internet of Things (C-IOT)

Energy conservation is one of the essential requirements for the betterment of the Earth's future. The main objective of this chapter is to provide an effective framework of smart grid (SG) using the intelligence of cognitive radio network (CRN). The SG connectivity will be enhanced by using the functionality of internet of things (IoT). Using IoT, the devices between the consumers, manufactures, and energy source provider (government) can conserve resources and economy. In this chapter, the requirement of smart grid, the need for IoT technology in SG, and the usage of CRN in SG are all elaborated. The infrastructural design pattern of cognitive radio-IoT-based smart grid is introduced in this chapter.

Cognitive Radio Network for E-Health Systems

The world is witnessing widespread roots of ubiquitous computing across disciplines and industries. It is equipped with the ability to monitor anything from anywhere with efficient, easy, and equitable goods that offer services for everyone. This has become possible through usage of wireless technologies, which possess an extensive scope in healthcare domain. However, despite various advantages, wireless technologies are faced with distinct challenges in hospital environment. For instance, wireless devices often tend to cause electromagnetic interference to critical medical devices resulting in malfunctioning. Further, with ubiquitous computing, sensitive data about health state of patients is constantly being shared remotely from one place to another. Therefore, systems in place must address requirements of data security, and thus privacy. For this purpose, the chapter presents a collaborative study on cognitive-radio-based healthcare system, including advantages, architecture, and challenges related to implementation of cognitive radios in hospital environment.

Improved Spectrum Sensing Based on Polar Codes for Cognitive Radio Networks

One challenge of a sensing technique is reducing sensing time while ensuring good effective data rate. In fact, once compressive sensing based on sub-Nyquist sampling is adopted, sensing time can be reduced by saving number of samples. This increases the probability of missed detection which causes collisions with primary service and worsens channel imperfections. In such case erasures occur in addition to errors. In this chapter, the authors propose a new technique to correct erasures while keeping sensing time at a desired level. Based on polar code and low complexity decoding algorithm, the proposed technique exhibits for high code rates better performance in terms of bit error rate (BER) compared to two existing techniques based on other codes, namely low-density parity check (LDPC) and BCH.

Spectrum Sensing Techniques

Cognitive radio has come a long way in the recent years with the advent of improved algorithms and instrumentation. However, for ease and efficient working of cognitive radio, there is a need to have effective detection of spectrum sensing. The objective of spectrum sensing techniques is to find spectrum holes which can be accessible by the users of cognitive radio. The deployment of suitable sensing techniques reduces undesirable congestion in traffic and enhancement of spectrum usage. All these require sensing techniques whose main goal is oriented towards efficient identification and subsequent deployment of spectrum. This chapter is aimed to give a brief overview of some spectrum sensing techniques. An attempt is made to give the characteristics of the highly deployable sensing schemes. Accordingly, the merits and demerits are comprehensively highlighted. Further, emphasis has been given to relevant future challenges.

Sensing Techniques for Next Generation Cognitive Radio Networks

Cognitive radio is the technology used to solve the problem of spectrum underutilization by performing spectrum sensing, spectrum management, spectrum sharing, and spectrum mobility. The primary goal of cognitive radio is open spectrum sharing. Spectrum is a scarce and valuable natural resource that has to be used very effectively. The static allocation of spectrum to the licensed users will lead to wastage of resources when the spectrum is unused by the licensed user. Spectrum sensing methodology helps in detecting the spectrum holes and enables the unlicensed users to access the unused bands in the licensed spectrum effectively without interfering the licensed users. Cognitive thinking takes wireless communication to the next level by sensing the electromagnetic environment and dynamically adjusts its operating parameters in order to achieve maximum throughput, mitigate interference, facilitate interoperability, etc. The chapter presents the basics of cognitive radio networks, its architecture, its application, and advantages of cognitive radio networks.

Basic Spectrum Sensing Techniques

Cognitive radio network (CRN) is an upcoming networking technology that can utilize both radio spectrum and wireless resources efficiently based on the information gathered from the past experience. There are two types of users in CRN, namely primary and secondary. PUs (PU) have the license to operate in certain spectrum band while the secondary (SU) or cognitive radio (CR) users do not have the license to operate in the desired band. However, they can opportunistically utilize the unused frequency bands. Spectrum sensing, spectrum management, spectrum sharing, and spectrum mobility are the four major functions of cognitive radio systems. The main objective of spectrum sensing is to provide better spectrum access to CR users, without causing any harmful interference to PUs. Sensing accuracy is considered as the most important factor to determine the performance of cognitive radio network. In this chapter, the challenges and requirement involved in spectrum sensing are detailed. Further, various spectrum sensing basic techniques are also discussed in detail.

A Quick Overview of Different Spectrum Sensing Techniques

The next generation of emerging wireless technology is dealing with spectrum shortage. For appropriate and practical implementation of latest wireless technologies, the sufficient amount of frequency is needed. Cognitive radio (CR) is introduced as a proficient nominee to manage spectral undersupply problem, as it rapidly increases the use of underutilize spectrum via spectrum sensing. This chapter introduces brief start about spectrum holes in addition to spectrum sensing framework. Further, the chapter explains the issues in spectrum sensing and how the cooperative sensing technique is fit to overcome these issues like shadow fading and receiver uncertainty. Consequently, the various non-cooperative sensing techniques are also discussed including their test statics. The advantages and disadvantages of different sensing techniques is exhibited at the end.

Multiple Antennas-Based Improved Sensing Detector

5G is one of the newest and most acclaimed engineering technology of wireless era. This technology is also known as CRNs, in which spectrum detection plays a key role. Through this chapter, the authors explore multiple antennas-based improved sensing detector (MA_ISD). In the said scheme by using adaptive threshold and multiple antennas patterns, its easy to mitigate sensing failure issue enriches reliability. The presented scheme uses two detectors (TD) concept. The concept of two detectors (i.e., TD scheme has been applied in which multiple antennas are used for electing the quality signals). The considered model upgrades the detection operation and acquires limited or inferior detection time. The first energy detector uses a single adaptive threshold (ED-SAT) while another energy detector employs two adaptive thresholds (ED-TAT). The threshold value is accommodative as it relays on noise variance (σ_ω^2), the behavior of noise variance transforms in accordance with noise signal. Both the detectors run collaboratively and their gain is then supply to a decision device which operates OR functions. In this research work, results shows that with cooperation of two antennas (Nr=2) in multiple antennas-based improved sensing detector (MA_ISD) technique delivers enhancement in detection performance by 24.6%, 53.4%, 37.9%, and 49.6%, in contrast to existing schemes (i.e., EDT-ASS-2015 scheme, ED and cyclo-2010, adaptive SS-2012, and conventional-ED) scheme at -12dB SNR, respectively. During the time, proposed technique also reduces the sensing time in the order of 47.0 ms, 49.0 ms, and 53.2 ms as compared to existing schemes (EDT-ASS-2015, Adaptive SS-2012, and ED and Cyclo-2010) scheme at - 20 dB SNR, respectively.