Intelligent Wireless Communications
2: Department of Computer Science, University of Nicosia, Nicosia, Cyprus
3: Warsaw University of Technology, Warsaw, Poland
4: Department of Electrical and Computer Engineering, Hellenic Mediterranean University, Heraklion, Greece
The incredible growth in the development and use of wireless communication technologies has led to research in both academia and industry on artificial intelligence (AI) methods that enable intelligent technologies, smarter services and applications, business processes and social interactions to satisfy future requirements. AI mechanisms are being exploited in smart intelligent network applications to provide insights from collected data by identifying patterns and allowing operational predictions with higher accuracy in smaller time periods. The future of AI-powered wireless networking infrastructures depends on finding effective solutions to a number of technical challenges that such paradigms introduce, including better intelligent sensor capabilities, smarter big data analytics, automated remote data management, as well as open and secure processes. This book presents innovative research in emerging artificial intelligence methods for the processing, storing and analysing of large date sets generated by wireless communication infrastructures. It addresses major new technological developments, reflects on industry needs, current research trends and future directions. The authors focus on the development of AI-powered mechanisms for future wireless networking applications and architectures which will lead to more performant, resilient and valuable ecosystems and automated services. The book is aimed at researchers, engineers and scientists involved in the design and development of protocols and AI applications for wireless communication devices, and wireless networking technologies.
Inspec keywords: intelligent networks; Internet of Things; radio networks; telecommunication computing
Other keywords: QoS; IoMT; VANETs; smart communication; artificial intelligence; underwater communications; medical information communication; nonorthogonal multiple access; ARIMA; machine learning; discrete wavelet transform; jamming signal; IoT; security vulnerabilities; gain vector-based recursive least squares; MPLS protocol; intelligent agents system; smart interoperability public safety wireless network; downlink; electricity consumption; green energy harvesting protocols; damaged critical infrastructure; neural networks; security smart antenna; intelligent big data analytics; Intelligent Internet of things; intelligent wireless communication systems; blockchain systems; optical wireless communication
Subjects: General and management topics; Communications computing; General electrical engineering topics; Radio links and equipment; Computer networks and techniques; Intelligent networks; Computer communications
- Book DOI: 10.1049/PBTE094E
- Chapter DOI: 10.1049/PBTE094E
- ISBN: 9781839530951
- e-ISBN: 9781839530968
- Page count: 453
- Format: PDF
-
Front Matter
- + Show details - Hide details
-
p.
(1)
-
1 An overview of the intelligent big data analytics and their technological presence in the modern digital age
- + Show details - Hide details
-
p.
1
–31
(31)
Big data (BD) refers to the huge volume of data, both structured and unstructured, generated by our society as a large number of people are connected through information technology and the Internet. This is essential and necessary to have technologies capable of monitoring and interpreting this great flow of information that travels through computerized environments. So, companies optimize their solutions and manage to improve the daily lives of the population through the intelligent and efficient processing of these data. This is where the BD and big data analytics (BDA) come into the picture, which together with artificial intelligence (AI) techniques, analyze this large volume of data in realtime and extract information and knowledge. Big data is a process of analyzing and interpreting a large volume of data stored remotely, integrating any data collected on a subject or a company, such as purchase and sale records and even nondigital interaction channels. Big data analytics is a data analysis process with a specific purpose, forming analysis strategies aiming at a large number of data enabling the study of consumer behaviors and expectations, in addition to observing market trends. Artificial intelligence, as its main focus, is the processing of data in order to make the device or technology more intelligent and capable of reproducing human abilities. Therefore, this chapter aims to provide an overview of intelligent BDA, showing its relationship and technological integrations, approaching its success relationship, with a concise bibliographic background, categorizing and synthesizing the potential of both technologies.
-
2 Artificial intelligence in IoT and its applications
- + Show details - Hide details
-
p.
33
–58
(26)
Artificial intelligence (AI) and Internet of things (IoT) are both become most amazing technologies nowadays. Artificial intelligence deals with intelligent reasoning and speedy data analysis which would cover the smartest future applications. Internet of things promises the deployment of these smart applications everywhere in the world. Both AI and the IoT interact in a unique way to increase increased human-machine interaction efficiency, increased operational efficiency, and true digital transformation. In IoT, the AI can ensure to collect only adequate data in order to reduce the processing time of big data amount sent by sensors. Artificial intelligence processes such as data integration, data selection, data cleaning, data transformation, data mining, and pattern evaluation all will provide the best solution to manage huge data flows and storage in the IoT network. The use of AI in IoT will extract the unique capability in different intelligent aspects such as smart decisions, smart metering, and forecasting and these will generate a new intelligent application in industrial security, health care, and smart homes. In this chapter, we will provide a brief concept about the AI and its relationship to the IoT, in addition to the benefits of AI to solve many challenges in IoT operations such as sensing, computing, energy management, and security. The chapter will also provide different types of AI methodologies related to the IoT applications.
-
3 Green energy harvesting protocols for intelligent wireless communication systems
- + Show details - Hide details
-
p.
59
–85
(27)
With the exponential growth of economical and sophisticated user devices in the digitized world, the competence of communicating high volume of data such as high-definition videos or high-quality multimedia has become users' requirements nowadays. However, spectral-efficient networking for high-speed data and self-sustainability in energy are two major trade-offs in design of energy-aware next-generation wireless communications. Efficient utilization of energy resources reduces the cost of power consumption that can improve the economic characteristics of the network. In this chapter, we investigate the energy and quality-ofservice (QoS) aware smart services using artificial intelligence (AI) assisted automated wireless communications, describes the novel energy state prediction model, online optimization algorithms for energy-harvesting (EH), and enhancement of autonomy by maximizing the throughput based on self-sustainability. The intelligent power management technological developments while satisfying the desired QoS paving the way for AI-powered mechanisms in future wireless networking applications.
-
4 Discrete wavelet transform applications in the IoMT
- + Show details - Hide details
-
p.
87
–105
(19)
Recent research efforts have explored methods to achieve the wavelet transform as the most significant tool in medical image enhancement and processing. The problems of image fusion, compression, edge detection, denoising, and contrast enhancement can be handled by discrete wavelet transform (DWT) in the Internet of medical things (IoMT) framework. In this chapter, we present the novel DWT with orthogonal and biorthogonal wavelets application. Multiple applications of the wavelet transform in medical images have been submitted. These applications demonstrate the successful impact of applying DWT. The DWT has the ability to enhance the medical image and remove noise. The DWT in image compression can separately reduce the computational complexity into high and low frequency. This process reduces the image data in order to be able to store or transmit data in an efficient form. There are some advantages in using fusion based on DWT during other traditional methods, for example, reduced features and energy compaction. In digital watermarking, DWT technique is used for embedding and extraction of watermark in the original image.
-
5 Intelligent agents system for medical information communication
- + Show details - Hide details
-
p.
107
–134
(28)
This chapter describes the importance of the agents when making an intelligent web communication system for medical information. The state-of-the-art of the agents is performed and the implementation of mobile agents is highlighted with their positive and negative features when gathering information in the Internet. In this respect, an intelligent system with a mobile agent model for communication of medical information is proposed that could be implemented in the wireless structure. The platform for the creation of the medical communication information system that is proposed is a part of intelligent wireless communication. The result from using mobile agent is generated which is also important when working with big data.
-
6 Intelligent Internet of things in wireless networks
- + Show details - Hide details
-
p.
135
–162
(28)
Nowadays, intelligence Internet of things (IoT) wireless networks are the most promising technologies for intelligent future applications, since it takes the advantages of artificial intelligence for sensing and data analysis to develop a new generation of smarter IoT applications around everywhere. Next-generation wireless networks are expected to support extremely high data rates and radically new applications, which require a new wireless radio technology paradigm. The challenge is that of assisting the radio in intelligent adaptive learning and decisionmaking so that the diverse requirements of next-generation wireless networks can be satisfied. Machine learning is one of the most promising artificial intelligence tools, conceived to support smart radio terminals. The main challenge for future IoT networks is how to cope with such complex systems, where a huge number of devices compete for limited wireless resources and where heterogeneity is everincreasing. There is an urgent need for more intelligent networks that lead to more interoperable solutions and that can make autonomous decisions on optimal operation modes and configurations. In IoT wireless networks, the AI can insure to collect only adequate data in order to reduce the processing time of big data amount sent by sensors. To manage big data and store it in IoT networks, AI processes play a major role in data integration, selection, classification, and mining, as well as assessment of information patterns. It offers unique solutions in decision-making, measurement, and contribution to the construction of IoT networks with a new, fast, and intelligent character. In this chapter, we will provide intelligence IoT wireless networks in addition to the AI contribution in programming and configuring of IoT network devices. This chapter will also introduce the algorithms and strategy of intelligence related to cognitive IoT networks and quality of service, in addition to the benefits of cognitive and SDN to IoT operations in heterogeneous networks.
-
7 Impact of jamming signal on system performance in downlink of IoT network relying on nonorthogonal multiple access
- + Show details - Hide details
-
p.
163
–176
(14)
This chapter analyzed a downlink NOMA network with impacts from jamming signal. With regard to impairment of SIC, that is, imperfect SIC and perfect SIC were used in the receiver to provide a comparative study. To examine system per-formance, the outage probability was considered in this chapter under various para-meters. Specially, we derived the closed-form expressions for the outage probability. In order to evaluate other metrics, the throughput of the system is studied and system performance remains stable at high SNRs. These results can be achieved and verified by the simulation results and they were also further evaluated under varying the number of jamming sources. In future work, more users and more pairs of the group need to be investigated to implement such scenarios in real IoT systems.
-
8 QoS of communication networks using MPLS protocol
- + Show details - Hide details
-
p.
177
–220
(44)
This chapter describes the place of multiprotocol label switching (MPLS) in current state-of-the-art computer networking systems which emphasize the use of MPLS as a quality of service (QoS) technique in different networking environments such as data centers and multimedia networking as backbone networks. The chapter also includes practical experiments, which are applied using the OPNET network simulation tool for transmitting multimedia over IP/MPLS networks and a fat tree data center architecture employing MPLS. The examples show the feasibility of MPLS as a QoS tool compared to best-effort IP networks. In addition, this chapter describes an effort of designing a network-on-chip system engaging MPLS mechanism as on-chip communication method which has been completely developed in Cþþ. However, all results clearly confirm that MPLS is an efficient QoS tool.
-
9 Damaged critical infrastructure for VANETs
- + Show details - Hide details
-
p.
221
–238
(18)
Vehicular ad hoc network (VANET) is a spontaneous creation of wireless network among vehicles to exchange data. Breakdown or interruption of communication is one of the most immediate and significant impacts of natural disasters. Clustering is one of the most common networking protocols for data propagation in those networks. The application of the clustering algorithm is effective in VANET because the algorithm makes it a more robust and scalable network. For stable cluster formation in VANET, some constraints such as vehicles' velocity and vehicles' separation distance must be considered while selecting the cluster head which is the node responsible for data propagation to the infrastructure. In damaged VANET, the cluster head might be unable to successfully transmit the packets to the infrastructure. In such cases, overlapping clusters with double cluster heads is one of the best solutions to guarantee successful packet delivery to the destination. One of the major challenges in clustering is the cluster head (CH) election since the CH has a critical role in data routing. This chapter discusses a new clustering algorithm based on a weighted formula for cluster head election. The weighted formula is based on three parameters: the trust, the distance, and the velocity. This protocol will elect two cluster heads to ensure an end-to-end communication path between source and destination.
-
10 Artificial intelligence-enabled optical wireless communication links: a revolutionary approach toward smart communication model
- + Show details - Hide details
-
p.
239
–269
(31)
Optical wireless communication links with transmission and bandwidth properties vary the same as that of conventional fiber links that can be installed at 1/10th of the cost of the latter. This means OWC links can be the new face of cost-effective network infra-structure, which can conveniently converge between backbone networks and end-users. OWC links are, however, extremely vulnerable to atmospheric adversities, which can deplete the transmission quality and operational link range. Although various mitigation strategies like spatial diversity, relay transmission, and coherent detection have been proposed in the past, each of them possesses certain inherent limitations, which ultimately prohibits large-scale deployment.
-
11 Intelligent underwater wireless communications
- + Show details - Hide details
-
p.
271
–305
(35)
Underwater wireless communications play a vital role in marine applications such as oceans' monitoring, exploration, and scientific data collection. Underwater wireless network (UWN) is an important area for the research due to the unique and harsh underwater environment and application which are quite challenging and problematics issues still remain to be addressed. Hence, intelligent solutions are needed for various levels of network architecture. Recently, many academic and industrial researchers have paid attention to the development of state-of-the-art intelligent solutions for future underwater wireless communications and networks. This chapter enlightens and guides the potential research communities about the recent progress in the area of intelligent UWC. This chapter also reviews underwater wireless networks based on acoustics, even electromagnetic, and optical signals. UWN systems also are reviewed for data gathering and recharge docking systems. Intelligent methodologies and modelings are the key drivers for future intelligent underwater communication.
-
12 Machine learning algorithms for smart data analysis in the Internet of things: an overview
- + Show details - Hide details
-
p.
307
–327
(21)
Machine learning (ML) techniques will benefit immensely from the avalanche of data readily available from various (IoT) applications considered as the major contributor of new data for future intelligent network. Based on this new concept, network systems will further magnify their capacity to exploit variety of experimental data across a plethora of network devices, study the data information, obtain knowledge and make informed decisions based on the dataset at their disposal. Smart IoT data analysis are performed utilizing supervised learning, unsupervised learning and reinforced learning. This study is limited to supervised and unsupervised ML techniques. In other to achieve the set objectives, reviews and discussions of substantial issues related to supervised or unsupervised machine learning techniques were executed, highlighting the advantages and limitations of each algorithm as well presenting the recent research trends and recommendations for future study.
-
13 Artificial intelligence and machine learning aided blockchain systems to address security vulnerabilities and threats in the industrial Internet of things
- + Show details - Hide details
-
p.
329
–361
(33)
Advent of digital sensors and machines led to a significant acceleration in industrial evolution. The desire to automate industrial processes with minimum human intervention paved the way for the onset of a new era of technological nomenclature called the industrial Internet of things (IIoT). A remarkable feature of IIoT is its underlying architecture which allows the managers/engineers/supervisors to remotely operate and access the performance of their machines. Industries ranging from healthcare, finance, logistics, and power have witnessed a major performance increment and quality stabilization by transforming themselves into an IIoT empowered smart environment. However, this transformation has brought with itself a whole new set of challenges with cybersecurity being the paramount. The vulnerabilities like bugs and broken processes can lead to a serious compromise or even collapse of security mechanisms of IIoT networks. Such a situation will have a devastating impact on the financial health, reputation, and credibility of companies. After an extensive review of existing technologies, we believe that blockchain, artificial intelligence (AI), and machine learning (ML) can complement each other in building a revolutionary deterrent to negate malicious activities that in any form intend to harm the system. While, blockchain offers public/private/consortium relationships, ML and AI, on the other hand, follow the principle of supervised/ unsupervised/reinforcement learning and reactive/memory approaches, respectively. Based on the distributed ledger system, blockchain mechanisms can be aided with self-learning algorithms which will update and strengthen the database by learning each time the system suffers new forms of network attacks and intrusions. This process of learning will help build a robust system which can learn to optimize its deterrence procedures against different forms of attacks. It is due to these overwhelming benefits, blockchain, AI, and ML find applications in smart logistics, predictive maintenance, autonomous vehicles, intelligent manufacturing, and smart grid maintenance.
-
14 Improved gain vector-based recursive least squares for smart antenna applications
- + Show details - Hide details
-
p.
363
–380
(18)
In future, the improvement of the utilization of the frequency spectrum for a more desirable quality of service (QoS) hinges upon the use of smart antenna (SA) systems in wireless and cellular networks. A signal's output power can be enhanced by using SA techniques. These techniques enhance the signal power toward expected directions within the network. However, long-range communication applications still face some unsolved problems such as signal fading and cochannel interference. This chapter provides solutions to the problems of signal fading and cochannel interference by enhancing the already conventional recursive least squares (RLS) algorithms in SA arrays over long-range communication channels. This chapter further provides performance-based comparisons between the enhanced RLS algorithm and other beamforming algorithms such as the conventional RLS algorithm and the least mean squares (LMS) algorithm. In the conventional RLS algorithm, matrix inversion computations are not required. This is because the conventional RLS algorithm already determines the inverse correlation matrix directly. For this reason, the RLS algorithm can save computational power. The RLS algorithm is enhanced through the introduction of a constant denoted as m to the gain factor of the algorithm and the reason for this introduction is to yield an improved gain vector. Results from our simulation results indicate that the enhanced RLS decreases mean square error (MSE) of the signals, which makes the output of the filter smoother. These results also show that our enhanced RLS improves SNR when compared to the conventional methods. The results of this study show the benefits of improving the gain of the SA, which yields an increase in the sharpness and range of the SA over a long-range communication channel.
-
15 Forecast of electricity consumption: a comparison of ARIMA and neural networks
- + Show details - Hide details
-
p.
381
–397
(17)
Electricity consumption is a critical factor in the climate change problem. The in time and reliable prediction of future consumption can help experts take the appropriate measures to eliminate electricity production side effects on the planet. Experts also can use forecasts to design suitable renewable energy systems. In this chapter, we analyze two well-known forecasting models. The first is the autoregressive integrated moving average (ARIMA), which has been used in many real-life cases in the previous years, and the second one is the neural network forecasting method which, is based on human's brain function. Each method is analyzed with its implementation and steps. The last section is a head-to-head comparison of the two methods.
-
16 Smart interoperability public safety wireless network
- + Show details - Hide details
-
p.
399
–418
(20)
Smart public safety network is one of the hot topics in wireless communication. This chapter presents the design and development of first responders network (First Net) which provides broadband spectrum to the emergency service personnel at the time of disaster in a national level. In the same direction, the proposed design solves the problems of radio networks' interoperability. The smart public safety network uses incompatible radio systems and technologies in different frequencies to coordinate and collaborate works without any interferences. Smart radio network technologies aim to optimize the connections between different public safety organizations. An adaptive controlled system for different two-way radio networks interoperability is designed to increase the trust and reliability at the disaster and crisis times. The design usually depends upon the infrastructure and environmental condition of radio networks that already exist in the country. For simplicity, the presented design is focused on voice communications by accessing audio through commonly featured radio sockets. On the other hand, the existence of the third parties in the system such as the Internet, cellular, or even a console in large twoway radio systems reduces the reliability of the whole design and increases the cost and lack of time.
-
Back Matter
- + Show details - Hide details
-
p.
(1)