In this paper, we propose a novel method for abnormality detection in Unmanned Surface Vehicles (USVs) based on a Multi-Modal Bayesian generative model to enhance safety and monitoring.During the training phase, we use a Null Force Filter and an unsupervised clustering algorithm on multimodal data collected from Global Positioning System (GPS) and motor current sensors.In the testing phase, we use a Coupled Modified Markov Jump Particle Filter (CM-MJPF) to infer the GPS position and motor current of the USV, as well as to detect abnormalities in both modalities.Due to the coupled methodology, the system is able to learn the statistical similarity between the evolving GPS and motor current data.As a result, the causality of defects is inherently captured within the dynamical inference, making the proposed approach explainable.

We investigate the secrecy performance of integrated satellite and terrestrial networks (ISTNs) with the support of a drone (aka UAV). An optimisation problem is formulated to maximise the secrecy rate while guaranteeing the quality of service, including the minimum secrecy rate of the legitimate user, the minimum data rate of normal users, and power consumption. A nested-loop algorithm including outer and inner loops is proposed to convert the initial non-convex problem into multiple convex problems, which are solved by the Dinkelbach algorithm. Simulation results prove the efficiency of our methods in terms of secrecy rate compared to traditional benchmarks.

This paper exploits a quantum-empowered machine learning algorithm to enhance computation learning speed. Under stochastic behaviours and quantum uncertainty, we examine the offloading problem to maximize the computational task processing efficiency, considering the computation latency, energy consumption, and quantum network adaptability. From the Markov decision process, the paper proposes a novel quantumempowered deep reinforcement learning (Qe-DRL) approach, combining quantum computing theory and machine learning to achieve exploration and exploitation trade-off via quantum parallelism significantly. Furthermore, we develop a modified Grover’s algorithm with exponential convergence speed to provide a searching strategy for transition quantum states probabilities. Simulation results establish the effectiveness of the proposed QeDRL algorithm and its superior computational learning speed.

This paper investigates the unmanned aerial vehicle (UAV)-assisted communication network with multiple downlink users (DLUs) and uplink users (ULUs) in the presence of multiple malicious jammers. To guarantee fairness among the users and their uplink and downlink communication throughput, we aim to maximize the minimum average throughput by jointly optimizing the scheduling of ULUs/DLUs, three dimensional (3D) trajectory and the UAV transmission power. Although the optimization problem is computationally intractable due to its non-convexity, we develop an iterative algorithm based on the block coordinate descend approach and the successive convex approximation technique to solve the problem efficiently. Numerical outcomes show that our proposed algorithm can improve throughput significantly over several benchmark schemes.

This article proposes an optimization framework for power and time resource allocation during time sharing non-orthogonal multiple access (TS-NOMA) transmissions performed by an unmanned aerial vehicle (UAV) in the context of a large-scale scenario. The objective of the proposed UAV-TS-NOMA system and optimization framework is to jointly maximize the energy efficiency (EE) and the downlink throughput fairness among users within the UAV communication range. The idea behind is to propose a communication system that: i) merges the advantages of UAV communications with the ones offered by the TS-NOMA paradigm and ii) maximizes the EE and the downlink fairness among users. The resulting model finds applicability in performing energy efficient and throughput fair transmissions into power-constrained communication scenarios. Performance investigations regarding the proposed framework in finding the optimal set of resources which maximizes jointly the above mentioned network metrics, have shown the advantage of the proposed two-step optimization framework in finding the optimal configuration of both power and time resources, respecting both the power constraints at the transmitter and the quality-of-service requirement of the users. In addition, it is shown how under particular conditions the proposed framework jointly optimizes the aforementioned network metrics in only one step.

LTE is rapidly gaining momentum for building future 4G cellular systems, and real operational networks are under deployment worldwide. To achieve high throughput performance, in addition to an advanced physical layer design LTE exploits a combination of sophisticated mechanisms at the radio resource management layer. Clearly, this makes difficult to develop analytical tools to accurately assess and optimise the user perceived throughput under realistic channel assumptions. Thus, most existing studies focus only on link-layer throughput or consider individual mechanisms in isolation. The main contribution of this paper is a unified modelling framework of the MAC-level downlink throughput of a sigle LTE cell, which caters for wideband CQI feedback schemes, AMC and HARQ protocols as defined in the LTE standard. We have validated the accuracy of the proposed model through detailed LTE simulations carried out with the ns-3 simulator extended with the LENA module for LTE.

Offloading is gaining momentum as a technique to overcome the cellular capacity crunch due to the surge of mobile data traffic demand. Multiple offloading techniques are currently under investigation, from modifications inside the cellular network architecture, to integration of multiple wireless broadband infrastructures, to exploiting direct communications between mobile devices. In this paper we focus on the latter type of offloading, and specifically on offloading through opportunistic networks. As opposed to most of the literature looking at this type of offloading, in this paper we consider the case where requests for content are non-synchronised, i.e. users request content at random points in time. We support this scenario through a very simple offloading scheme, whereby no epidemic dissemination occurs in the opportunistic network. Thus our scheme is minimally invasive for users' mobile devices, as it uses only minimally their resources. Then, we provide an analysis on the efficiency of our offloading mechanism (in terms of percentage of offloaded traffic) in representative vehicular settings, where content needs to be delivered to (subsets of the) users in specific geographical areas. Depending on various parameters, we show that a simple and resource-savvy offloading scheme can nevertheless offload a very large fraction of the traffic (up to more than 90%, and always more than 20%). We also highlight configurations where such a technique is less effective, and therefore a more aggressive use of mobile nodes resources would be needed.

Adaptive Modulation and Coding (AMC) in LTE networks is commonly employed to improve system throughput by ensuring more reliable transmissions. Most of existing AMC methods select the modulation and coding scheme (MCS) using pre-computed mappings between MCS indexes and channel quality indicator (CQI) feedbacks that are periodically sent by the receivers. However, the effectiveness of this approach heavily depends on the assumed channel model. In addition CQI feedback delays may cause throughput losses. In this paper we design a new AMC scheme that exploits a reinforcement learning algorithm to adjust at run-time the MCS selection rules based on the knowledge of the effect of previous AMC decisions. The salient features of our proposed solution are: i) the low-dimensional space that the learner has to explore, and ii) the use of direct link throughput measurements to guide the decision process. Simulation results obtained using ns3 demonstrate the robustness of our AMC scheme that is capable of discovering the best MCS even if the CQI feedback provides a poor prediction of the channel performance.