Data transfer infrastructures composed of Data Transfer Nodes (DTN) are critical to supporting distributed computing and storage capabilities for clouds, data repositories, and complexes of supercomputers and instruments. The infrastructure’s throughput profile, estimated as a function of the connection round trip time using Machine Learning (ML) methods, is an indicator of its operational state, and has been utilized for monitoring, diagnosis and optimization purposes. We show that the inherent statistical variations and precision of the throughput profiles estimated by ML methods can be exploited for unauthorized use of DTNs’ computing and network capacity. We present a game theoretic formulation that captures the cost-benefit trade-offs between an attacker that attempts to hide under the profile’s statistical variations and a provider that counters by using additional measurements at an added cost. The Nash equilibrium conditions of this game provide qualitative insights and bounds for the success probabilities of the attacker and provider, based on the generalization equation of the ML-estimate. We present experimental results that illustrate scenarios wherein a significant portion of DTN computing capacity is compromised without being detected by an attacker that exploits the ML estimate properties.

Motivated by the large number of wearables offering geolocation, human mobility mining has emerged as an novel research field  within AI. The study of mobility creates increasingly predictable models in which it is easy to find patterns of behaviour. However, this data is not publicly available and access to it is restricted to large telecommunications operators. In this context, this paper aims to solve one of the main problems of human mobility databases, i.e. the scarcity of data for the generation of human mobility models. For this purpose, Generative adversarial network (GANs) have been proposed to generate synthetic time-series mobility data. Moreover, several neural network models are proposed to assess the impact of synthetic data generation on the prediction of human mobility. Our results show that the use of synthetic data improves predictions of human mobility compared to models based on available measured data.

This work evaluates the most popular digital services: Facebook, Instagram,TikTok, YouTube, Netflix, Amazon, Zoom, Skype, and Spotify. To date, such a complete analysis of the damage to the environment caused by  these applications has not been carried out and it is hoped that this paper will contribute mainly in three ways: raise awareness  of our digital carbon footprint; disseminate knowledge on how to estimate pollution from our online behavior; and pointing to the  causes of the issue to find optimal solutions.

In this paper, we propose a conjecture that endogenous security without any prior knowledge is similar to perfect secrecy without any prior knowledge. To prove the conjecture, we first establish a cryptography model of instinct function security to transform the security problem in the network domain into an encryption problem in the cryptographic domain. Then, we inherit and apply the established ideas and means of the Perfect Secrecy, and propose the concept, definition and corollaries of the perfect instinct function security (PIFS) corresponding to the Perfect Secrecy. Furthermore, we take the DHR system as a concrete implementation of PIFS and propose the DHR Perfect Security Theorem corresponding to Shannon’s Perfect Secrecy Theorem. Finally, we prove that the DHR satisfying the “One-Time Reconstruction” constraint is the sufficient and necessary condition to achieve perfect security. This means that the existence of PIFS is also proven. The analysis shows that any reconfigurable system can be encrypted by its construct and that the PIFS converts the one-way transparent superiority of the attacker into a double-blind problem for both the attacker and the defender, which leads to that the attacker is impossible to obtain useful construction information from the attacks and unable to find a better way than blind trial-and-error or brute-force attacks. Since the attackers are required to have the new powerful ability to crack the structure cryptogram, the threshold of cyber security is raised to at least the same level as cryptogram deciphering, thereafter the ubiquitous cyber threats are destined to be significantly reduced.

The determinization of fuzzy automata is a well-studied problem in theoretical computer science celebrated for its practical applications. Indeed, in the fields of fuzzy discrete event systems, fault diagnosis, clinical monitoring, decision-making systems, and model checking, when a suitable model of a fuzzy automaton is employed, it is desirable to find its language-equivalent deterministic version because of its computational efficiency. Although many methods have been developed to convert a fuzzy automaton to its language equivalent fuzzy deterministic finite automaton (FDfA), they can be applied only for fuzzy automata defined over specific underlying sets of truth values. For example, recently developed determinization methods employ the concept of maximal factorization, which can be defined only on non-locally finite lattices or the Boolean lattice. In addition, not all such determinization methods result in a minimal FDfA. On the other hand, even such determinization methods have been developed for fuzzy automata over specific underlying structures, these methods cannot be generalized for fuzzy automata over locally finite lattices. This article focuses on filling this gap and develops a novel method for computing a minimal FDfA for a fuzzy automaton defined over a locally finite and divisible residuated lattice. Our method uses the concept of a reduction graph that helps us achieve a minimal FDfA.

This is the final part of the two- part series paper, where the simulation results namely leg’s trajectory, path tracing, obsacles detected, path constructed, movements of three motors and test like two-element chain three- element chain, are presented. I have left r&d, so will not correct update or complete papers. The earlier version of two parts, which was able to get saved from computer crash is uploaded.

The main goal is to produce an output to convert the global real image into supported impressions (Animated Image). The concept of the paper is based on one-of-a-kind photos that are turned into an art shape akin to oil. This project would use an OpenCV package in Python to create the layout and a Deep Convolutional Generative Adversarial Network (DCGAN) to generate realistic-looking photos. The system is made up of three components: (a) a generative adversarial network that has been trained to generate comic characters, (b) a cartoon model that will generate edge animated images using OpenCV, and (c) an edge detects that takes input from portraits and then generates comic characters based on the resulting edge images.

We report on a systematic procedure for the port-wise decomposition of electrical networks, which we refer to as the template-based approach. Our method exploits the SPQR-tree decomposition as an algorithmic procedure for decomposing a given circuit into a maximal set of one- and multiports. In particular, we provide a solution for representing subcircuits by suitable template graphs, such that they are encapsulated not decomposed/processed by the SPQR-algorithm. Contrary to current methods, our replacement graphs are not sophisticated but rather simple, which, in general, enhances the performance of the algorithm, when it comes to decomposing large electrical networks with multiports. Some electrical devices, such as transistors or op-amps, are commonly described in terms of their terminals rather than their ports. Thus, we cover the application of the template-based approach to three-terminal devices. Lastly, we discuss the advantages of the template-based SPQR-tree decomposition in the context of wave digital emulation.

Spin-Vertical Cavity Surface Emitting Lasers (spin-VCSELs) are undergoing increasing research effort for new paradigms in high-speed optical communications and photon-enabled computing. To date research in spin-VCSELs has mostly focused on Quantum-Well (QW) devices. However, novel Quantum-Dot (QD) spin-VCSELs, offer enhanced parameter controls permitting the effective, dynamical and ultrafast manipulation of their light emission’s polarization. In the present contribution we investigate in detail the operation of QD spin-VCSELs subject to polarization modulation for their use as ultrafast light sources in optical communication systems. We reveal that QD spin-VCSELs outperform their QW counterparts in terms of modulation efficiency, yielding a nearly two- fold improvement. We also analyse the impact of key device parameters in QD spin-VCSELs (e.g. photon decay rate and intra-dot relaxation rate) on the large signal modulation performance with regard to optical modulation amplitude and eye-diagram opening penalty. We show that in addition to exhibiting enhanced polarization modulation performance for data rates up to $250Gb/s$, QD spin-VCSELs enable operation in dual (ground and excited state) emission thus allowing future exciting routes for multiplexing of information in optical communication links.

In many natural and engineered systems, the circle and the spiral are the appropriate primitives. As example, rotating motion of fluid around a common centerline constitutes a vortex. In many of these situations, the associated fractal dimension is of analytical interest. The box counting approach to determining this dimension is not appropriate for such cases, because unlike squares and boxes that span 2- and 3-dimensional spaces, circles and spheres don’t. Since self-similarity or scale invariance is not an exact property, we propose the computation of the dimension is such cases based on the number of circles or spheres in the mapping process. This is illustrated by a few examples. How this may apply to spiral patterns that are repeated across scale is indicated.

This study shows the application of AI image generators with fine-tuning to the Kansei engineering design process.  Kansei engineering is a method for surveying and analyzing users’ latent feelings and preferences to product and service development (Nagamachi, 1991, 2011).. There are several common difficulties in applying Kansei engineering, including a lack of variety in evaluation samples, the need for non-designers to participate in the design process, and the fact that designers are often too busy to fully engage in the design process. In order to address these challenges, we have applied an AI image generator, StableDiffusion  (Rombach et al. 2021),to the Kansei engineering design process. The “Hypernetworks” framework to fine-tune StableDiffusion in order to make better representation of the shape and appearance of milk cartons. Based on previous Kansei engineering study on milk cartons (Ishihara et al., 1996) , several attempts are shown to use AI-generated designs to create innovative and visually appealing milk carton designs, including designs incorporating red and blue colors and abstract shapes, and a design inspired by the abstract paintings of Jackson Pollock. The authors conclude that the use of AI technology can be a valuable tool in boosting innovative design through Kansei engineering, and that further research is needed to explore the potential for interaction between humans and AI in the design process.

We have observed a boom in video streaming over the Internet, especially during the Covid-19 pandemic, that could exceed the network resource availability. In addition to upgrading the network infrastructure, finding a way to smartly adapt the streaming system to the network and users’ conditions to satisfy clients’ perceptions is exceptionally critical, too. This paper proposes a new QoE-aware adaptive streaming scheme over HTTP - ABRA - to make flexible adaptations based on the network and the client’s current status. Besides, we propose a technique that can keep the buffer at an average high for more than 10s. We were limiting the phenomena of rebuffering due to unexpected and unpredictable bandwidth changes. The algorithm keeps the quality of subsequent versions at a constant level even when the average bitrate decreases, increasing the QoE. Experimental results show that our method can improve QoE from 7.86% to 20.41% compared to state-of-the-art methods. ABRA can provide better performance in terms of QoE score in all buffer conditions compared to the existing solutions while maintaining a minimum secured buffer level for the worst case.

This paper presents a methodology of adapting complete open-source digital tooling, ISA, IPs and manufacture-able PDKs to tapeout a minimalist RISC-V based SoC.

This paper studies analytically and quantitatively the influence of the power source on the dynamic performance of the altitude control system of a quadrotor powered by an electrochemical battery. This paper also proposes the formulation of the altitude as a constrained optimization problem in which the drone, actuators, control law, and electrochemical battery models are considered, to define a trade-off between the power consumption rate and the closed-loop dynamic performance loss. An analytical representation of the effect of the battery discharge over the altitude dynamics is obtained through a linear approximation, enabling an analysis of the system poles. The problem of designing an altitude controller is then posed as a constrained optimization problem that can include the battery as a factor. A comparison of the error dynamics between the cases of the battery-unaware controller design and the battery-aware controller design is performed in simulations and experimental flight tests. The results lead to the following conclusions: i. the analytical demonstration agrees with the worse performance observed in the in-flight dynamics as the battery discharges and ii. through a battery-aware controller design approach this effect can be diminished, at the cost of a trade-off in the battery discharge rate.

This is a preprint only and has not been peer-reviewed or published anywhere so far. Abstract : Blockchain technology (BCT) is a game changer for many industries due to its distinct advantages like conducting peer-to-peer transactions in trustless environment, enhanced data security, greater transparency, and the inclusion of smart contracts. Its application in the energy industry in Blockchain Enabled Interconnected Smart Microgrids (BSMGs) is on the rise as it can automate local energy markets; execute energy trading; and implement market operations and management. However, they are limited by their scalability and low transactions rate. Also, with the increase in adoption of BSMGs, different types of BCT platforms will emerge, creating heterogeneity in the system. This shortcoming may lead to a monopoly of certain platforms over the rest. These drawbacks can be overcome by establishing interoperability between heterogeneous blockchains. Interoperability can also enable inter-microgrid transactions which will be hindered if there is no inter-chain communication and transaction. Cosmos is a network of blockchains which allows blockchains, applications and services to be interconnected through Inter Blockchain Communication (IBC) protocol. Ignite CLI is an open-source, command line interface which easily creates modular and customizable blockchains which are inherently connected to Cosmos through IBC. In this paper, Ignite CLI is used to create blockchains for BSMGs which exchange data and calls for inter-microgrid transactions via IBC. The procedure to establish inter-chain communication is defined. Interoperability between BSMGs is explored and demonstrated through different examples, for the first time in the energy domain.

Chronic obstructive pulmonary disease (COPD) is a major public health concern across the world. Since it is an incurable disease, early detection and accurate diagnosis are very crucial for preventing the progression of the disease. Lung sounds provide reliable and accurate prognoses for identifying respiratory diseases. Recently, Altan et al. recorded 12-channel real-time lung sound dataset, namely RespiratoryDatabase@TR, for five different severity levels of COPD at Antakya State Hospital Turkey, and proposed deep learning frameworks for two-class COPD classification and five-class classification using a deep belief network (DBN) classifier and extreme learning machine (ELM) classifier respectively. A classification accuracy of 95.84% and 94.31% were achieved for two-class and five-class respectively. In this paper, we have proposed a melspectrogram snippet representation learning framework for both two-class and five-class COPD classification. The proposed framework consists of the following stages: preprocessing, melspectrogram snippet representation generation from lung sound and fine tuning of a pretrained YAMNet. Experimental analysis on the RespiratoryDatabase@TR dataset demonstrates that the proposed framework achieves accuracies of 99.25% and 96.14% for binary and multi-class COPD severity classification respectively, which is superior to the only existing methods proposed by Altan et al. for severity analysis of COPD using lung sounds.

This paper presents a method for spatial hands-on detection, which not only allows the recognition of the hands-on the steering wheel, but also at which location and strength the driver is gripping the wheel allowing a classification if the grip complies with safety regulations. This paper discusses the selected measurement principle called time domain reflectometry and how the corresponding measurement setup can be implemented inside a steering wheel. Based on a series of tests different machine learning approaches are reviewed which obtain the location as well as the grip strength based on the output of the time domain reflectometry.

In previous papers various True Random Number Generators (TRNG) and Physically Unclonable Function (PUF) are discussed individually but only a few of them include unified TRNG and PUF. TRNG’s are used in cryptography and simulations for conditions in background noise that impact the system. for a PUF its very unlikely to make another device alike due to various intrinsic properties during manufacturing process. PUF’s have various applications which include low-cost authentication, secure key generation. In this work a new architecture of unified true random number generator (TRNG) and physically unclonable function (PUF) is introduced based on ring oscillators with high Uniqueness, reliability, throughput, entropy. The main working of circuit is dependent on two reference values Nij and Nref . The PUF response is based on various intrinsic variables whereas TRNG’s randomness is based on LSB of the Nij . With the current method, results and analysis indicates increase in uniqueness, reliability and other parameters compared to previous papers. The whole circuit is designed using 45nm CMOS technology.