Improving Next Generation Wireless Network Performance And Reliability With Deep Learning

A rudimentary question whether machine learning in general, or deep learning in particular, could add to the well-established field of wireless communications, which has been evolving for close to a century, is often raised. While the use of deep learning based methods is likely to help build intelligent wireless solutions, this use becomes particularly challenging for the lower layers in the wireless communication stack. The introduction of the fifth generation of wireless communications (5G) has triggered the demand for “network intelligence” to support its promises for very high data rates and extremely low latency. Consequently, 5G wireless operators are faced with the challenges of network complexity, diversification of services, and personalized user experience. Industry standards have created enablers (such as the network data analytics function), but these enablers focus on post-mortem analysis at higher stack layers and have a periodicity in the time scale of seconds (or larger). The goal of this dissertation is to show a solution for these challenges and how a data-driven approach using deep learning could add to the field of wireless communications. In particular, I propose intelligent predictive and prescriptive abilities to boost reliability and eliminate performance bottlenecks in 5G cellular networks and beyond, show contributions that justify the value of deep learning in wireless communications across several different layers, and offer in-depth analysis and comparisons with baselines and industry standards. First, to improve multi-antenna network reliability against wireless impairments with power control and interference coordination for both packetized voice and beamformed data bearers, I propose the use of a joint beamforming, power control, and interference coordination algorithm based on deep reinforcement learning. This algorithm uses a string of bits and logic operations to enable simultaneous actions to be performed by the reinforcement learning agent. Consequently, a joint reward function is also proposed. I compare the performance of my proposed algorithm with the brute force approach and show that similar performance is achievable but with faster run-time as the number of transmit antennas increases. Second, in enhancing the performance of coordinated multipoint, I propose the use of deep learning binary classification to learn a surrogate function to trigger a second transmission stream instead of depending on the popular signal to interference plus noise measurement quantity. This surrogate function improves the users' sum-rate through focusing on pre-logarithmic terms in the sum-rate formula, which have larger impact on this rate. Third, performance of band switching can be improved without the need for a full channel estimation. My proposal of using deep learning to classify the quality of two frequency bands prior to granting the band switching leads to a significant improvement in users' throughput. This is due to the elimination of the industry standard measurement gap requirement—a period of silence where no data is sent to the users so they could measure the frequency bands before switching. In this dissertation, a group of algorithms for wireless network performance and reliability for downlink are proposed. My results show that the introduction of user coordinates enhance the accuracy of the predictions made with deep learning. Also, the choice of signal to interference plus noise ratio as the optimization objective may not always be the best choice to improve user throughput rates. Further, exploiting the spatial correlation of channels in different frequency bands can improve certain network procedures without the need for perfect knowledge of the per-band channel state information. Hence, an understanding of these results help develop novel solutions to enhancing these wireless networks at a much smaller time scale compared to the industry standards today

The ever-evolving wireless technology industry is demanding new technologies and standards to ensure a higher quality of experience for global end-users. This developing challenge has enabled researchers to identify the present trend of machine learning as a possible solution, but will it meet business velocity demand? Next-Generation Wireless Networks Meet Advanced Machine Learning Applications is a pivotal reference source that provides emerging trends and insights into various technologies of next-generation wireless networks to enable the dynamic optimization of system configuration and applications within the fields of wireless networks, broadband networks, and wireless communication. Featuring coverage on a broad range of topics such as machine learning, hybrid network environments, wireless communications, and the internet of things; this publication is ideally designed for industry experts, researchers, students, academicians, and practitioners seeking current research on various technologies of next-generation wireless networks.

A new era of wireless networks is evolving, thanks to the significant advances in communications and networking technologies. In parallel, wireless services are witnessing a tremendous change due to increasingly heterogeneous and stringent demands, whose quality of service requirements are expanding in several dimensions, putting pressure on mobile networks. Examples of those services are augmented and virtual reality, as well as self-driving cars. Furthermore, many physical systems are witnessing a dramatic shift into autonomy by enabling the devices of those systems to communicate and transfer control and data information among themselves. Examples of those systems are microgrids, vehicles, etc. As such, the mobile network indeed requires a revolutionary shift in the way radio resources are assigned to those services, i.e., RRM. In RRM, radio resources such as spectrum and power are assigned to users of the network according to various metrics such as throughput, latency, and reliability. Several methods have been adopted for RRM such as optimization-based methods, heuristics and so on. However, these methods are facing several challenges such as complexity, scalability, optimality, ability to learn dynamic environments. In particular, a common problem in conventional RRM methods is the failure to adapt to the changing situations. For example, optimization-based methods perform well under static network conditions, where an optimal solution is obtained for a snapshot of the network. This leads to higher complexity as the network is required to solve the optimization at every time slot. Machine learning constitutes a promising tool for RRM with the aim to address the conflicting objectives, i.e., KPIs, complexity, scalability, etc. In this thesis, we study the use of reinforcement learning and its derivatives for improving network KPIs. We highlight the advantages of each reinforcement learning method under the studied network scenarios. In addition, we highlight the gains and trade-offs among the proposed learning techniques as well as the baseline methods that rely on either optimization or heuristics. Finally, we present the challenges facing the application of reinforcement learning to wireless networks and propose some future directions and open problems toward an autonomous wireless network. The contributions of this thesis can be summarized as follows. First, reinforcement learning methods, and in particular model-free Q-learning, experience large convergence time due to the large state-action space. As such, deep reinforcement learning was employed to improve generalization and speed up the convergence. Second, the design of the state and reward functions impact the performance of the wireless network. Despite the simplicity of this observation, it turns out to be a key one for designing autonomous wireless systems. In particular, in order to facilitate autonomy, agents need to have the ability to learn/adjust their goals. In this thesis, we propose transfer in reinforcement learning to address this point, where knowledge is transferred between expert and learner agents with simple and complex tasks, respectively. As such, the learner agent aims to learn a more complex task using the knowledge transferred from an expert performing a simpler (partial) task.

Recently machine learning schemes have attained significant attention as key enablers for next-generation wireless systems. Currently, wireless systems are mostly using machine learning schemes that are based on centralizing the training and inference processes by migrating the end-devices data to a third party centralized location. However, these schemes lead to end-devices privacy leakage. To address these issues, one can use a distributed machine learning at network edge. In this context, federated learning (FL) is one of most important distributed learning algorithm, allowing devices to train a shared machine learning model while keeping data locally. However, applying FL in wireless networks and optimizing the performance involves a range of research topics. For example, in FL, training machine learning models require communication between wireless devices and edge servers via wireless links. Therefore, wireless impairments such as uncertainties among wireless channel states, interference, and noise significantly affect the performance of FL. On the other hand, federated-reinforcement learning leverages distributed computation power and data to solve complex optimization problems that arise in various use cases, such as interference alignment, resource management, clustering, and network control. Traditionally, FL makes the assumption that edge devices will unconditionally participate in the tasks when invited, which is not practical in reality due to the cost of model training. As such, building incentive mechanisms is indispensable for FL networks. This book provides a comprehensive overview of FL for wireless networks. It is divided into three main parts: The first part briefly discusses the fundamentals of FL for wireless networks, while the second part comprehensively examines the design and analysis of wireless FL, covering resource optimization, incentive mechanism, security and privacy. It also presents several solutions based on optimization theory, graph theory, and game theory to optimize the performance of federated learning in wireless networks. Lastly, the third part describes several applications of FL in wireless networks.

A paradigm shift in connectivity is being experienced. Traditional demarcations among different types of networks are getting blurred. This organic development inevitably leads to a unified Internet architecture. Such an architecture would connect different types of networks, which also implies an integration of broad range of entities and services, by adopting Internet Protocol (IP). Undoubtedly, this architecture would invoke a proliferation of volume and diversity of network traffic. To address these challenges, it is of paramount importance to innovate proper architecture and algorithms for different segments of such a unified network. A robust next generation Internet should seamlessly accommodate different types of traffic, including Machine-to-Machine (M2M) type. Primary source of M2M traffic is the Internet of Things (IoT). Wireless network is a crucial component of such a system. In addition, it is evident that 5G is the technology of choice to connect a multitude of networks. Therefore, an improved user experience demands improvement in 5G networks and IoT networks. In this dissertation, innovative architectures and algorithms for both 5G networks and IoT networks are proposed. The first chapter describes the problem of scalability in typical cellular networks. A novel software defined unified architecture based on virtualized core network, which integrates both WLAN and LTE traffic, is proposed to reduce the signaling load on the core network. The second chapter focuses on the need of low-latency (or high throughput) in certain IoT networks. A modified scheduling algorithm based on multi-radio nodes and multiple sink nodes is proposed to address this problem. In addition to this, two novel retransmission strategies are proposed in this chapter. In the third chapter, the problem of stability in the IoT wireless networks caused by inefficient frequency hopping is described, which is followed by a channel blacklisting algorithm to improve the network stability.

Today, Artificial Intelligence (AI) and Machine Learning/ Deep Learning (ML/DL) have become the hottest areas in information technology. In our society, many intelligent devices rely on AI/ML/DL algorithms/tools for smart operations. Although AI/ML/DL algorithms and tools have been used in many internet applications and electronic devices, they are also vulnerable to various attacks and threats. AI parameters may be distorted by the internal attacker; the DL input samples may be polluted by adversaries; the ML model may be misled by changing the classification boundary, among many other attacks and threats. Such attacks can make AI products dangerous to use. While this discussion focuses on security issues in AI/ML/DL-based systems (i.e., securing the intelligent systems themselves), AI/ML/DL models and algorithms can actually also be used for cyber security (i.e., the use of AI to achieve security). Since AI/ML/DL security is a newly emergent field, many researchers and industry professionals cannot yet obtain a detailed, comprehensive understanding of this area. This book aims to provide a complete picture of the challenges and solutions to related security issues in various applications. It explains how different attacks can occur in advanced AI tools and the challenges of overcoming those attacks. Then, the book describes many sets of promising solutions to achieve AI security and privacy. The features of this book have seven aspects: This is the first book to explain various practical attacks and countermeasures to AI systems Both quantitative math models and practical security implementations are provided It covers both "securing the AI system itself" and "using AI to achieve security" It covers all the advanced AI attacks and threats with detailed attack models It provides multiple solution spaces to the security and privacy issues in AI tools The differences among ML and DL security and privacy issues are explained Many practical security applications are covered

In recent years, wireless networks communication has become the fundamental basis of our work, leisure, and communication life from the early GSM mobile phones to the Internet of Things and Internet of Everything communications. All wireless communications technologies such as Bluetooth, NFC, wireless sensors, wireless LANs, ZigBee, GSM, and others have their own challenges and security threats. This book addresses some of these challenges focusing on the implication, impact, and mitigations of the stated issues. The book provides a comprehensive coverage of not only the technical and ethical issues presented by the use of wireless networks but also the adversarial application of wireless networks and its associated implications. The authors recommend a number of novel approaches to assist in better detecting, thwarting, and addressing wireless challenges and threats. The book also looks ahead and forecasts what attacks can be carried out in the future through the malicious use of the wireless networks if sufficient defenses are not implemented. The research contained in the book fits well into the larger body of work on various aspects of wireless networks and cyber-security. The book provides a valuable reference for cyber-security experts, practitioners, and network security professionals, particularly those interested in the security of the various wireless networks. It is also aimed at researchers seeking to obtain a more profound knowledge in various types of wireless networks in the context of cyber-security, wireless networks, and cybercrime. Furthermore, the book is an exceptional advanced text for Ph.D. and master’s degree programs in cyber-security, network security, cyber-terrorism, and computer science who are investigating or evaluating a security of a specific wireless network. Each chapter is written by an internationally-renowned expert who has extensive experience in law enforcement, industry, or academia. Furthermore, this book blends advanced research findings with practice-based methods to provide the reader with advanced understanding and relevant skills.

- provides some fundamental concepts related to 5G networks and the 5G NR signal processing. A review of AI and state of the art machine learning techniques is also given. - deals with the 5G/6G and AI enabled applications such as AR/VR, autonomous vehicles, mobile multimedia services, context aware communications, Industrial IoT and security. -elaborates on how AI techniques can enhance network and traffic management in 5G/6G networks. These include AI based mobility management, routing, scheduling, network performance optimization and even energy efficiency. -discusses the application of AI to 5G/6G NR signal processing and also the air interface. AI and deep learning techniques for channel coding, automatic modulation detection, channel estimation and equalization as well as spectrum management are presented with a view to highlight the benefits of using AI as compared to traditional techniques.

The ever-increasing demand for data-hungry wireless services and rapid proliferation of wireless devices in sub-6 GHz band have pushed the current wireless technologies to a breaking point, necessitating efficient and intelligent strategies to utilize scarce communication resources. This thesis aims at leveraging novel communication frameworks, artificial intelligence techniques, and synergies between them in bringing efficiency and intelligence to the next generation of wireless networks.In the first chapter of this thesis, we propose a novel spectrum sharing scheme to address spectrum shortage, a fundamental issue in current and future wireless networks. Our proposed scheme enables transparent spectrum utilization for a small cognitive radio network by leveraging two interference management techniques that are not reliant on inter-network coordination, fine-grained synchronization, and knowledge about other occupants of the spectrum. We further extend this idea in the second chapter of this thesis and enable concurrent device-to-device and cellular communications in cellular networks where the base station and wireless devices exploit interference management techniques to avoid causing interference to each other, making concurrent spectrum utilization possible for both cellular and device-to-device communications. In the third chapter, to enhance spectral efficiency, connectivity, and throughput of Wireless Local Area Networks (WLAN), we propose a non-orthogonal multiplexing scheme (NOMA). In our proposed scheme, the access point (AP) is equipped with a novel precoder design and user grouping which are tailored based on the requirements of power-domain NOMA. Also, a novel successive interference cancellation technique is designed for users which does not require channel estimation to decode the signals and is more resilient to interference compared to the existing techniques. The second part of this thesis focuses on taking advantage of artificial intelligence for solving communication and networking challenges and also taking advantage of novel communication frameworks to let future wireless networks indulge intelligence-oriented networking and resource management. In the fourth chapter, we propose a new solution to solve a long-standing issue ahead of multi-user multiple-input multiple-output (MU-MIMO) communications in WLANs, which is the large sounding overhead for acquiring the channel state information (CSI). Our learning-based solution includes an automated mechanism that enables access points to collect, clear, and balance dataset, and also deep neural networks to compress CSI and reduce the airtime overhead for channel acquisition. However, with provisioning concurrent MU-MIMO and orthogonal frequency division multiple access (OFDMA) in the new generation of WLANs, not only the sounding overhead problem becomes more acute, but it also marries with a complex resource allocation problem which makes designing a practical enabler of MU-MIMO-OFDMA transmissions necessary for WLANs. In the fifth chapter of this thesis, we propose DeepMux, which comprises a deep-learning-based channel sounding and a deep-learning-based resource allocation both of which reside in access points and impose no computational/communication burden on users, enabling efficient downlink MU-MIMO-OFDMA transmissions in WLANs. We finally design a communication framework for accelerating federated learning in future intelligent transportation systems, where heterogeneous capabilities and mobility of users along with limited available bandwidth for communications are huge obstacles toward making the network intelligent in a distributed manner. With the aid of a deadline-driven scheduler and asynchronous uplink multi-user MIMO, our proposed solution reduces data loss at vehicles in a dynamic vehicular environment, making a concrete step toward the practical adoption of federated learning in future transportation systems.

This conference proceeding is a collection of the papers accepted by the CENet2022 – the 12th International Conference on Computer Engineering and Networks held on November 4-7, 2022 in Haikou, China. The topics focus but are not limited to Internet of Things and Smart Systems, Artificial Intelligence and Applications, Communication System Detection, Analysis and Application, and Medical Engineering and Information Systems. Each part can be used as an excellent reference by industry practitioners, university faculties, research fellows and undergraduates as well as graduate students who need to build a knowledge base of the most current advances and state-of-practice in the topics covered by this conference proceedings. This will enable them to produce, maintain, and manage systems with high levels of trustworthiness and complexity.