Abstract The usage of swarms of drones is expected to continue growing in the next years, particularly in dangerous scenarios, such as monitoring and rescue missions in hostile and disaster areas. Small-sized Unmanned Aerial Vehicles (UAVs) are highly suitable for use in such scenarios due to their agility and maneuverability. On the other hand, their limited battery capacity poses significant challenges, especially during missions requiring full coverage of large areas in a short time and extreme weather conditions. This work proposed an energy efficiency approach, which makes use of mobile ground-based battery-swapping stations (BSSes), to speed up the UAV’s battery replacement and reduce energy waste in the round trip to the charging station. Specifically, a Context-Aware Coverage Path Planning (CACPP) problem has been formulated to determine the complete coverage path of a large area by a swarm of UAVs, minimizing the path overlapping and UAV battery swapping. The model takes into account the need to continue re-planning the mission, depending on the weather conditions (i.e., temperature and wind), the presence of obstacles, and the residual energy levels of the drones, as well as the relative positions of the drones and mobile BSSes. To solve the CACPP problem, an iterative approach leveraging two synchronized optimization models for planning UAV paths and BSS routes has been presented. As the CACPP problem is NP-hard, a heuristic procedure for solving it has also been evaluated. Experimental results show that it can be appropriate for large instances of the problem.

The continuous emergence of new and sophisticated malware specifically targeting Android-based Internet of Things devices is causing significant security hazards and is consequently fostering the need for effective detection models and strategies able to work with these hardware-constrained devices. In addition, since such models are often trained on confidential application data, many involved subjects are reluctant to share their data for this purpose. Accordingly, several Federated Learning-based solutions are emerging, which rely on the capabilities of Machine Learning models in malware detection/classification without sharing user data. However, Federated Learning methods are often adversely affected by non-independent and identically distributed data in terms of both the required training time and classification results. Therefore, a promising solution could be to overcome the Federated Learning-related issues by preserving the privacy of end-user data. In this direction, the capabilities of Markov chains and associative rules are extended within a federated environment to face malware classification tasks in the IoT scenario. The presented approach, evaluated on several malware families, has achieved an average accuracy of 99% in the presence of centralized and decentralized unbalanced training/testing data by overcoming the most common state-of-the-art approaches. Also, its runtime performance is comparable with centralized ones by considering several non-independent and identically distributed dataset partitions, splitting criteria, and clients, respectively.

The controller placement problem (CPP) is one of the main issues that need to be addressed in the context of Software Defined Networking (SDN), especially when different aspects are being considered, such as latency, capacity, reliability, and load balancing. Most of the solutions in the literature address these aspects by considering a fixed load for each controller and attempting to equally distribute the traffic demand of the switches among the controllers, which also have a fixed common capacity. On the contrary, in this work, the CPP is studied by considering load, controller capacity, and the failure probability of controllers and links as varying over time. The CPP is formulated in terms of a robust optimization problem, which, by introducing the concept of scenario, takes into account changes in the network status due to failures, load variations, and changes in switches’ demand and controllers’ capacity. The provided solution is robust, that is, neither controllers’ re-placement nor switches’ re-assignment is required as network conditions change. Besides, a co-evolutionary algorithm is provided to solve the aforementioned optimization problem. Two populations coevolve based on the concept of complementary evolution of allied species in nature. Experimental results on a set of real-world network topologies and comparisons with the state-of-the-art have proven the superiority of the proposal, in terms of better latency, load balancing and resilience, in solving the CPP under different network status changes that might occur over time.

With the emergence of COVID-19, mobile health applications have increasingly become crucial in contact tracing, information dissemination, and pandemic control in general. Apps warn users if they have been close to an infected person for sufficient time, and therefore potentially at risk. The distance measurement accuracy heavily affects the probability estimation of being infected. Most of these applications make use of the electromagnetic field produced by Bluetooth Low Energy technology to estimate the distance. Nevertheless, radio interference derived from numerous factors, such as crowding, obstacles, and user activity can lead to wrong distance estimation, and, in turn, to wrong decisions. Besides, most of the social distance-keeping criteria recognized worldwide plan to keep a different distance based on the activity of the person and on the surrounding environment. In this study, in order to enhance the performance of the COVID-19 tracking apps, a human activity classifier based on Convolutional Deep Neural Network is provided. In particular, the raw data coming from the accelerometer sensor of a smartphone are arranged to form an image including several channels (HAR-Image), which is used as fingerprints of the in-progress activity that can be used as an additional input by tracking applications. Experimental results, obtained by analyzing real data, have shown that the HAR-Images are effective features for human activity recognition. Indeed, the results on the k-fold cross-validation and obtained by using a real dataset achieved an accuracy very close to 100%.

Abstract Resource Planning Optimization (RPO) is a common task that many companies need to face to get several benefits, like budget improvements and run-time analyses. However, even if it is often solved by using several software products and tools, the great success and validity of the Artificial Intelligence-based approaches, in many research fields, represent a huge opportunity to explore alternative solutions for solving optimization problems. To this purpose, the following paper aims to investigate the use of multiple Artificial Neural Networks (ANNs) for solving a RPO problem related to the scheduling of different Combined Heat & Power (CHP) generators. The experimental results, carried out by using data extracted by considering a real Microgrid system, have confirmed the effectiveness of the proposed approach.

Abstract Ductal adenocarcinoma of the pancreas is a cancer with a high mortality rate. Among the main reasons for this baleful prognosis is that, in most patients, this neoplasm is diagnosed at a too advanced stage. Clinical oncology research is now particularly focused on decoding the cancer molecular onset by understanding the complex biological architecture of tumor cell proliferation. In this direction, machine learning has proved to be a valid solution in many sectors of the biomedical field, thanks to its ability to mine useful knowledge by biological and genetic data. Since the major risk factor is represented by genetic predisposition, the aim of this study is to find a mathematical model describing the complex relationship existing between genetic mutations of the involved genes and the onset of the disease. To this end, an approach based on evolutionary algorithms is proposed. In particular, genetic programming is used, which allows solving a symbolic regression problem through the use of genetic algorithms. The identification of these correlations is a typical objective of the diagnostic approach and is one of the most critical and complex activities in the presence of large amounts of data that are difficult to correlate through traditional statistical techniques. The mathematical model obtained highlights the importance of the complex relationship existing between the different gene’s mutations present in the tumor tissue of the group of patients considered.

Summary At present, 96% of the resources available into the World‐Wide‐Web belongs to the Deep Web, which is composed of contents that are not indexed by search engines. The Dark Web is a subset of the Deep Web, which is currently the favorite place for hiding illegal markets and contents. The most important tool that can be used to access the Dark Web is the Tor Browser. In this article, we propose a bottom‐up formal investigation methodology for the Tor Browser's memory forensics. Based on a bottom‐up logical approach, our methodology enables us to obtain information according to a level of abstraction that is gradually higher, to characterize semantically relevant actions carried out by the Tor browser. Again, we show how the proposed three‐layer methodology can be realized through open‐source tools. Also, we show how the extracted information can be used as input to a novel Artificial Intelligence‐based architecture for mining effective signatures capable of representing malicious activities in the Tor network. Finally, to assess the effectiveness of the proposed methodology, we defined three test cases that simulate widespread real‐life scenarios and discuss the obtained results. To the best of our knowledge, this is the first work that deals with the forensic analysis of the Tor Browser in a live system, in a formal and structured way.

Due to their open nature and popularity, Android-based devices have attracted several end-users around the World and are one of the main targets for attackers. Because of the reasons given above, it is necessary to build tools that can reliably detect zero-day malware on these devices. At the moment, many of the frameworks that have been proposed to detect malware applications leverage Machine Learning (ML) techniques. However, an essential requirement to build these frameworks consists of using very large and sophisticated datasets for model construction and training purposes. Their success, indeed, strongly depends on the choice of the right features used for building a classification model providing adequate generalisation capability. Furthermore, the creation of a training dataset that well represents the malware properties and behaviour is one of the most critical challenges in malware analysis. Therefore, the main aim of this paper is proposing a new dataset called Unisa Malware Dataset (UMD) available on http://antlab.di.unisa.it/malware/, which is based on the extraction of static and dynamic features characterising the malware activities. Additionally, we will show some experiments concerning common ML tools to demonstrate how it is possible to build efficient ML-based malware classification frameworks using the proposed dataset.

Modern interconnected power grids are a critical target of many kinds of cyber-attacks, potentially affecting public safety and introducing significant economic damages. In such a scenario, more effective detection and early alerting tools are needed. This study introduces a novel anomaly detection architecture, empowered by modern machine learning techniques and specifically targeted for power control systems. It is based on stacked deep neural networks, which have proven to be capable to timely identify and classify attacks, by autonomously eliciting knowledge about them. The proposed architecture leverages automatically extracted spatial and temporal dependency relations to mine meaningful insights from data coming from the target power systems, that can be used as new features for classifying attacks. It has proven to achieve very high performance when applied to real scenarios by outperforming state-of-the-art available approaches.

Nowadays, modern vehicles are becoming even more connected, intelligent, and smart. A modern vehicle encloses several cyber-physical systems, such as actuators and sensors, which are controlled by electronic control units (ECUs). Such ECUs are connected through in-vehicle networks, and, in turn, such networks are connected to the Internet of Vehicles (IoV) to provide advanced and smart features. However, the increase in vehicle connectivity and computerization, although it brings clear advantages, it introduces serious safety problems that can also endanger the life of the driver and passengers of the vehicle, as well as that of pedestrians. Such problems are mainly caused by the security weaknesses affecting the controller area network (CAN) bus, used to exchange data between ECUs. In this article, we provide two algorithms that implement a data-driven anomaly detection system. The first algorithm (cluster-based learning algorithm), is used to learn the behavior of messages passing on the CAN bus, for base-lining purposes, while the second one (data-driven anomaly detection algorithm) is used to perform real-time classification of such messages (licit or illicit) for early alerting in the presence of malicious usages. The experimental results, obtained by using data coming from a real vehicle, have shown that our approach is capable of performing better than other anomaly detection-based approaches.

SARS-CoV-2 is a novel severe acute respiratory syndrome-like coronavirus (SARS-CoV), which is responsible of the ongoing world pandemic of COVID-19 disease. Although many approaches are being investigated to address this issue, nowaday there are no vaccines available and there is little evidence supporting the efficiency of potential therapeutic agents. Moreover, the high mutation rate of this virus heavily affects the understanding of its evolution and diffusion mechanisms, and, in turn, the development of effective solutions. In this study, two novel algorithms are provided for finding out recurrent patterns of nucleotide subsequences of different SARS-CoV-2 genomes as a unique signature capable of identifying the most peculiar features of the pathogen. In particular, we provide several subsequence patterns related to the Spike glycoprotein, which is believed to be the main target for developing effective drugs and vaccines against the COVID-19 disease because of its role in the entrance of coronaviruses into host cells. The experimental results, obtained by analyzing 5000 genomes of SARS-CoV-2, have shown that the extracted patterns are able to recognize the Spyke protein in the 99.35% of the considered genomes. In addition, such patterns have proven to be highly discriminating with respect to other pathogenic genomes, such as SARS, Middle East respiratory syndrome, Nipah, and the streptococcus bacteria. We hope that the findings presented in this study can help specialists in speeding up the design of more accurate drugs or vaccines against SARS-CoV-2.

BACKGROUND: The huge quantity of data produced in Biomedical research needs sophisticated algorithmic methodologies for its storage, analysis, and processing. High Performance Computing (HPC) appears as a magic bullet in this challenge. However, several hard to solve parallelization and load balancing problems arise in this context. Here we discuss the HPC-oriented implementation of a general purpose learning algorithm, originally conceived for DNA analysis and recently extended to treat uncertainty on data (U-BRAIN). The U-BRAIN algorithm is a learning algorithm that finds a Boolean formula in disjunctive normal form (DNF), of approximately minimum complexity, that is consistent with a set of data (instances) which may have missing bits. The conjunctive terms of the formula are computed in an iterative way by identifying, from the given data, a family of sets of conditions that must be satisfied by all the positive instances and violated by all the negative ones; such conditions allow the computation of a set of coefficients (relevances) for each attribute (literal), that form a probability distribution, allowing the selection of the term literals. The great versatility that characterizes it, makes U-BRAIN applicable in many of the fields in which there are data to be analyzed. However the memory and the execution time required by the running are of O(n(3)) and of O(n(5)) order, respectively, and so, the algorithm is unaffordable for huge data sets. RESULTS: We find mathematical and programming solutions able to lead us towards the implementation of the algorithm U-BRAIN on parallel computers. First we give a Dynamic Programming model of the U-BRAIN algorithm, then we minimize the representation of the relevances. When the data are of great size we are forced to use the mass memory, and depending on where the data are actually stored, the access times can be quite different. According to the evaluation of algorithmic efficiency based on the Disk Model, in order to reduce the costs of the communications between different memories (RAM, Cache, Mass, Virtual) and to achieve efficient I/O performance, we design a mass storage structure able to access its data with a high degree of temporal and spatial locality. Then we develop a parallel implementation of the algorithm. We model it as a SPMD system together to a Message-Passing Programming Paradigm. Here, we adopt the high-level message-passing systems MPI (Message Passing Interface) in the version for the Java programming language, MPJ. The parallel processing is organized into four stages: partitioning, communication, agglomeration and mapping. The decomposition of the U-BRAIN algorithm determines the necessity of a communication protocol design among the processors involved. Efficient synchronization design is also discussed. CONCLUSIONS: In the context of a collaboration between public and private institutions, the parallel model of U-BRAIN has been implemented and tested on the INTEL XEON E7xxx and E5xxx family of the CRESCO structure of Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA), developed in the framework of the European Grid Infrastructure (EGI), a series of efforts to provide access to high-throughput computing resources across Europe using grid computing techniques. The implementation is able to minimize both the memory space and the execution time. The test data used in this study are IPDATA (Irvine Primate splice- junction DATA set), a subset of HS3D (Homo Sapiens Splice Sites Dataset) and a subset of COSMIC (the Catalogue of Somatic Mutations in Cancer). The execution time and the speed-up on IPDATA reach the best values within about 90 processors. Then the parallelization advantage is balanced by the greater cost of non-local communications between the processors. A similar behaviour is evident on HS3D, but at a greater number of processors, so evidencing the direct relationship between data size and parallelization gain. This behaviour is confirmed on COSMIC. Overall, the results obtained show that the parallel version is up to 30 times faster than the serial one.