Internet of Unmanned Things (IoUT) and Mission-based Networking

Unmanned aerial vehicles (UAVs), also known as drones, are rapidly growing in popularity in different sectors. Although they are still in their infancy stage in terms of mass adoption, usage, and regulations, they have already broken through industry rigid barriers, which otherwise seemed impenetrable by similar technological innovations. Most recently, drone technology uses have spread from commercial, industrial up to military applications and are expected to play a relevant role also as future communication technology, acting as support technology in next generation wireless networks (i.e., 6G network). Nowadays, UAVs are adopted into the main functions of various businesses and governmental organizations. UAVs are largely adopted in all those fields where men cannot reach or are unable to perform in a timely and efficient manner. For instance, quick deliveries at rush hour, scanning an unreachable military base or monitoring of harsh environments, are just a few examples of their main applications. Recently, drones are largely used for military offensive as flying weapons. In general, UAVs are adopted by industries to increase work efficiency and productivity, while decreasing workload and production costs and improving accuracy. In this chapter, we will provide an overview of main UAV applications, starting from military and commercial ones, till investigating their use for smart farming and agriculture. An overview of different machine-learning-based solutions for UAV applications, as well as Blockchain techniques for security purposes in UAV networks, is also included.

Cities are steadily upgrading the planning, development, and operation of their multimodal transportation networks in accordance to rising and shifting travel demand, necessitating the development of a multimodal mobility management system. Private sector technology-driven innovation in shared mobility services, vehicles, and networks is happening at a quick, accelerating, and opportunity-rich pace. City streets, however, are a limited and scarce resource. In this chapter, we evaluate different perspectives of AI technologies for Internet of Unmanned Things (IoUT).

The intelligent decision-making and communication among devices, allowing direct transmission among various entities and human, is the key component of Internet of Unmanned Things (IoUT). By gathering various smart devices (SDs) into clusters (Cls), the SDs within one cluster may share the information and deter malicious devices from network access by modifying and accessing the controls. The security of SDs can be provided through various security frameworks and mechanisms with and without blockchain. However, the SDs are not fully secured by spreading fake information and text alterations in the network. In this paper, we have proposed a secure and trusted communication mechanism using blockchain-based adaptive and comprehensive trust computation of each DS. The proposed mechanism ensures significant security and trust even with various malicious involvements. The simulation results are further demonstrated against various mechanisms such as probability attack, accuracy and communication delay in comparison of existing approach.

This chapter describes the promising role of mobile edge computing or multi-access edge computing (MEC) in enhancing the operations of various applications that fall under the umbrella of the Internet of Unmanned Things (IoUT). The value that MEC systems bring to 5G and beyond networks is unprecedented, and as such, it is extensively studied and recognized by both industry and academia, due to the opportunities to dynamically create means for deploying services in close proximity from the mobile end users, through the optimal deployment of edge servers that offer both computing and communication capabilities with enhanced quality of service (QoS) and quality of experience (QoE). Although MEC enables a significant reduction of end-to-end latency, and optimal utilization of bandwidth compared to traditional cloud computing deployments, maintaining service continuity for mobile users still requires complex management and orchestration systems to coordinate the deployment and operation of services on different MEC servers that are placed on the fixed locations. Thus, blending the concepts of MEC with unmanned aerial vehicles (UAVs) is expected to play an essential role in mitigating the aforementioned challenge, thereby achieving ubiquitous connectivity for mobile users. Leveraging on their flexibility and mobility, UAVs can be efficiently spawned at critical locations to boost wireless communication for various applications. The MEC-enhanced IoUT application domains covered in this chapter aim to enable scalability for boosting the operations in smart cities, through studying the advanced architectures, benefits, and challenges, for various use cases such as disaster management, intelligent healthcare, intelligent traffic, smart education, among the others.

Mobile edge computing (MEC) is an emerging technology which is becoming an important component in the networking infrastructure supporting Unmanned Aerial Vehicles (UAV). MEC paradigm contributes significantly to the reduction of communication latency and by allowing the presence of cloud-like services such as computation and storage at the edge of the network. UAVs can be viewed as flying Internet of Things (IoT) devices characterized by high mobility and energy scarcity. Therefore, MEC functionalities play an important role in addressing UAV application requirements in terms of resource allocation, task offloading, and energy efficiency optimization. In this chapter, we provide an architectural and functional overview of edge computing for UAV applications deployed in the context of IoT. We review the elements of MEC-assisted IoUT systems, the functionalities offered by MEC to UAV applications, the integration of distributed artificial intelligence (AI) methods in edge computation and the role of cooperative machine learning (ML) in the efficiency of MEC-based UAV applications.

Road markings are an essential and integral part of safe driving where main landmarks are used to guide drivers. Developing a robust road-marking interpretation system is challenging because of several aspects such as changes in light conditions, varying weather conditions, shadows, and faded signs and text. This chapter investigates the use of deep learning methods such as convolutional neural networks (CNNs) to classify symbolic road markings. Previous work in the literature has reported techniques which are predominantly based on feature extraction and template matching which restricts the use of such methods in real time. For autonomous vehicles, road markings need to be interpreted in real time to make timely decisions. This book chapter investigates and presents CNN-based image preprocessing methods to detect road markings for autonomous vehicles. Several CNN architectures were investigated with multiple convolutional, max pooling, and fully connected layers. This chapter will contribute by developing a model with low computational requirements which is essential for autonomous vehicles. It will further explore state-of-the-art image preprocessing methods such as grayscaling, top-hat, and Otsu’s method. The performance of the proposed road-marking detector will be benchmarked using a public dataset with labeled road-marking images.

The technology necessary to create drones has been around for a considerable amount of time. The technology was first employed by the military, but it has since made its way into the consumer market and can be put to use in a variety of contexts. From aerial photography to the monitoring of construction sites, drone technology possesses a number of appealing characteristics that make it particularly useful in a number of contexts and are also becoming increasingly affordable. There is already a rising demand among the community of telecom operators to employ these unmanned flying aircraft for particular objectives such as site monitoring. One example of this type of application is site inspection. This chapter aims to detail how drone technology can be further extended to improve serviceability, particularly whenever a network failure occurs that may affect the service quality, if not completely disrupt the whole connectivity service. In particular, the chapter will focus on improving serviceability whenever a network failure occurs.

High mobility, heterogeneous communication, and low latency are some of the challenges that Unmanned Aerial Vehicles (UAVs)-based networks must face to ensure reliable and long-distance connections. Unfortunately, the TCP/IP model partially fails to fulfill these prerequisites, as it was not designed for such types of networks. To overcome these limitations, the research community has explored alternatives to support these requirements. Several Future Internet Architectures (FIAs) were proposed. One of the most promising FIA candidates is Information-Centric Networking (ICN), which attracted a large research community since its creation. ICN adopts a content-driven networking paradigm rather than the conventional location-driven networking paradigm adopted by TCP/IP. The communication process relies on name-based routing, which means that content retrieval is independent of its source. This feature enables fast content dissemination and in-network caching. It also works as a built-in multicast mechanism. Another fundamental feature that ICN provides has to do with data security. It focuses on securing the content itself rather than communication channels. These properties position the ICN architecture as a reliable model for mobile networks in general and inter-UAV communications in particular, compared to the traditional TCP/IP architecture. In this chapter, we will provide an overview of ICN-based communication for UAVs.

An aerial Internet of Unmanned Things (IoUT) is exposed to many practical issues, such as signal propagation in unknown 3D environments, simultaneous heterogeneous network traffic types, and the need to coordinate with aerial vehicles, ground vehicles, and humans. Typically, several wireless channels co-exist to serve aerial control communication that requires low-latency and time guarantees and primarily video transmission that calls for high data rates. Environmental context information is often utilized and exchanged as well, above all location context, which is important for navigation and coordination of unmanned aerial vehicles. While in principle both communication and positioning technologies are available, practical inaccuracies and disturbances are challenging for an aerial IoUT. Thus, the validation of solutions for aerial networked systems strongly requires an experimental approach to discover deficiencies and to ensure practicality. In this chapter, we review the requirements for aerial networking and communications and discuss the capabilities and limitations of major candidate wireless technologies: Wi-Fi, 4G/5G, and LoRaWAN. We present a survey of current testbeds and achieved performance of single and multi-hop links, which is intended to serve as a guide for the setup of an aerial IoUT testbed.