Network slices for delay-constrained applications in 5G systems require computing facilities at the edge of the network to guarantee ultra-low latency in processing data flows generated by connected devices, which is challenging with larger volumes of data, and larger distances to the edge of the network. To address this challenge, we propose to extend 5G network slices with Unmanned Aerial Vehicles (UAV) equipped with multi-access edge computing (MEC) facilities. However, onboard computing elements (CE) consume UAV’s battery power thus impacting its flight duration. We propose a framework where a System Controller (SC) can turn on and off UAV's CEs, with the possibility of offloading jobs to other UAVs, to maximize an objective function defined in terms of power consumption, job loss, and incurred delay. Management of this framework is achieved by reinforcement learning. A Markov model of the system is introduced to enable reinforcement learning and provide guidelines for the selection of system parameters. A use case is considered to demonstrate the gain achieved by the proposed framework and discuss numerical results.

In the last decade, video surveillance systems have become more and more popular. Thanks to a decrease in price of video camera devices and the diffusion of cheap small unmanned aerial vehicles (UAVs), video monitoring is today adopted in a wide range of application cases, from road traffic control to precision agriculture. This leads to capture a great amount of visual material to be monitored and screened for event detection. However, information that is gathered from a platform of video monitoring UAVs may produce high-volume data, whose processing is unfeasible to be done locally by the same UAVs that perform monitoring. Moreover, because of the limited bandwidth of wireless links connecting UAVs to computing infrastructures that are installed on ground, offloading these data to edge clouds renders these platforms infeasible for video analysis applications with low-latency requirements. The target of this paper is to extend a 5G network slice for video monitoring with a Flying Ad-hoc NETwork (FANET) constituted by UAVs with multi-access edge computing (MEC) facilities (MEC UAVs), flying very close to the layer of UAVs monitoring the area of interest. A policy for mutual help among MEC UAVS is defined in order to increase the performance of the whole aerial MEC platform, so further reducing end-to-end latency between sources and actuators, and increasing system reliability. A use case is considered for a numerical analysis of the proposed platform.

The next generation of wireless communications networks, namely 6G, will be aimed at realizing a fully connected world, and at providing ubiquitous connectivity to people and objects even in remote areas that are very far from the structured Internet core network. These goals include the definition and the design of intelligent communications environments mainly characterized by pervasive artificial intelligence and large-scale automation. The target of this paper is the design of a management framework for edge networks realized with Flying Ad-Hoc Networks (FANET) consisting of a set of Unmanned Aerial Vehicles (UAVs) to provide a remote geographic area with computing and networking facilities for delay-sensitive applications. To this purpose, each UAV is equipped with a Computing Element (CE) to process jobs received through vertical offloading from ground devices. In addition, horizontal offload among UAVs of the FANET is introduced for load balancing purposes, to guarantee that the FANET computation delay for each received job is minimized and is almost independent of the activity state of the area covered by the UAV receiving that job. The proposed FANET management framework is based on Deep Reinforcement Learning (DRL) to allow zero-touch adaptation to the timevariant activity state of the area covered by each UAV. Numerical results demonstrate the power of the proposed framework and the enhancements achieved with respect to the current literature.

Deployment of technologies for Intelligent Transportation Systems (ITS) involves the installation of Road Side Units (RSU) on the roadway and On-Board Units (OBU) inside vehicles. In this direction, 5G technologies will make a great impulse providing low latency communication and computation at the edge of the network. First, this paper defines VMEC-in-a-Box, a smart RSU combined with a MEC station, aimed at providing edge computing for vehicular applications by enabling job offloading from vehicles. VMEC-in-a-Box is equipped with a microeolic power generator to be autonomous and self-consistent even in presence of low levels of wind. The behavior of VMEC-in-a-Box is controlled by artificial intelligence to vary its computing capacity dynamically to pursue the best tradeoff between performance and power consumption, and to cooperate by offloading jobs to each other (horizontal offload) to improve performance and reliability of the system. Hence, the paper defines a Markov model to support decisions to optimize by means of Reinforcement Learning the system behavior according to two reward functions defined at the MEC and at the Vehicular Domains. To the best of our knowledge, this is the first work proposing an integrated framework to maximize reliability and performance at both Vehicular and MEC Domains.

Some representative 5G application scenarios regard geographic areas very far from the structured core network, but are characterized by the need for processing huge amount of data that cannot be transmitted to multi-access edge (MEC) facilities installed at the edge of that network. To this purpose, this paper proposes to extend a 5G network slice with a fleet of UAVs, each providing computing facilities, and for this reason referred to as MEC UAVs. The paper proposes a cooperation between MEC UAVs belonging to the same fleet based on job offloading, aiming at minimizing power consumption due to active computer elements providing MEC, job loss probability and queueing delay. A Reinforcement Learning (RL) approach is used to support the System Controller in its decisions. A numerical analysis is presented to evaluate achieved performance.