In the past two years, technology has undergone significant changes that have had a major impact on healthcare systems. Artificial intelligence (AI) is a key component of this change, and it can assist doctors with various healthcare systems and intelligent health systems. AI is crucial in diagnosing common diseases, developing new medications, and analyzing patient information from electronic health records. However, one of the main issues with adopting AI in healthcare is the lack of transparency, as doctors must interpret the output of the AI. Explainable AI (XAI) is extremely important for the healthcare sector and comes into play in this regard. With XAI, doctors, patients, and other stakeholders can more easily examine a decision s reliability by knowing its reasoning due to XAI s interpretable explanations. Deep learning is used in this study to discuss explainable artificial intelligence (XAI) in medical image analysis. The primary goal of this paper is to provide a generic six-category XAI architecture for classifying DL-based medical image analysis and interpretability methods.The interpretability method/XAI approach for medical image analysis is often categorized based on the explanation and technical method. In XAI approaches, the explanation method is further sub-categorized into three types: text-based, visualbased, and examples-based. In interpretability technical method, it was divided into nine categories. Finally, the paper discusses the advantages, disadvantages, and limitations of each neural network-based interpretability method for medical imaging analysis.

Explainable AI (XAI) techniques are used for understanding the internals of the AI algorithms and how they produce a particular result. Several software packages are available implementing XAI techniques however, their use requires a deep knowledge of the AI algorithms and their output is not intuitive for non-experts. In this paper we present a framework, (XAI4PublicPolicy), that provides customizable and reusable dashboards for XAI ready to be used both for data scientists and general users with no code. The models, and data sets are selected dragging and dropping from repositories While dashboards are generated selecting the type of charts. The framework can work with structured data and images in different formats. This XAI framework was developed and is being used in the context of the AI4PublicPolicy European project for explaining the decisions made by machine learning models applied to the implementation of public policies.

This work proposes an interpretable Deep Learning framework utilizing Vision Transformers (ViT) for the classification of remote sensing images into land use and land cover (LULC) classes. It uses the Shapley Additive Explanations (SHAP) values to achieve two-stage explanations: 1) bandwise feature importance per class, showing which band assists the prediction of each class and 2) spatial-wise feature understanding, explaining which embedded patches per band affected the network’s performance. Experimental results on the EuroSAT dataset demonstrate the ViT’s accurate classification with an overall accuracy 96.86 \%, offering improved results when compared to popular CNN models. Heatmaps in each one of the dataset’s existing classes highlight the effectiveness of the proposed framework in the band explanation and the feature importance.

Understanding the temperature dependence of acoustic and photoacoustic (PA) properties is important for the characterization of materials and measurements in various applications. Ultrasound methods have been developed to estimate these properties, but they require careful consideration of multiple variables and steps to obtain reliable results. This study aimed to develop an automated system for simultaneous characterization of acoustic and PA properties of materials. The system was designed to minimize operator errors, ensuring robust temperature control and reproducibility for acoustic measurements. This was made possible through the integration of a commercially available PA imaging system with a custom-built platform specifically tailored for ultrasound-based acoustic characterization. This platform consisted of both hardware and software modules. The system was evaluated with NaCl solutions at different concentrations and a gelatin/agar cubic phantom prepared with uniformly distributed magnetic nanoparticles serving as optical absorbers. Results obtained from the NaCl solution samples exhibited a high Lin s concordance coefficient (above 0.9) with previously reported studies. In the ultrasound/PA experiment, temperature dependences of the speed of sound and PA intensity revealed a strong Pearson s correlation coefficient (0.99), with both measurements exhibiting a monotonic increase as anticipated for water-based materials. These findings demonstrate the accuracy and stability of the developed system for acoustic property measurements.

In this work, we investigated the design of low loss and wideband shear horizontal surface acoustic wave (SH-SAW) acoustic delay lines (ADLs) on a sapphire-based thin-film lithium niobate on insulator (LNOI) platform. The SH-SAW propagates in a Y-cut LN/SiO2 double-layer thin film atop the sapphire substrate, where the significant acoustic impedance mismatch between the thin film and the substrate confines the acoustic energy at the surface, thus minimizing the propagation loss. The single-phase unidirectional transducers (SPUDT) used in this work is implemented with gold (Au) to maximize the electromechanical coupling as well as the directionality. The proposed ADLs based on YX-LN/SiO2/Sapphire centered at 830 MHz showed a minimum insertion loss (IL) of 3 dB, a wide fractional bandwidth (FBW) of 4.19\%, and a low propagation loss (PL) of 2.51 dB/mm, which yields an effective quality factor (QPL) exceeds 2,700. These results demonstrate the competitive performance of the proposed devices compared to state-of-the-art thin film LN ADLs, offering extremely low propagation loss for RF signal processing.

This paper presents the design of a MEMS resonator with capacitive transduction as an acoustic sensor, intended for cantilever-enhanced photoacoustic spectroscopy. The sensor employs area-variable capacitive detection by surrounding the silicon resonator with dense comb teeth. To reduce gas damping effects on the resonator motion, the anchor height is increased to 260 µm. This approach successfully resolves the capacitance detection sensitivity and motion damping trade-off commonly seen in acoustic detection. Experimental results exhibit a maximum sensitivity of 3749 mV/Pa at the resonant frequency of 1870 Hz with a 15 V bias voltage. The equivalent noise has a peak value of 7.9 µPa/Hz1/2 and the noise sources are analyzed.

This work presents a modified AlN/Sapphire layered SAW structure localized partial removal of AlN thin film and sapphire, respectively. The SAW propagation and resonance characteristics of the proposed structure with periodic grooves and voids are analyzed using finite element method (FEM). Compared with conventional AlN-based SAW, the proposed structure with optimization configuration and parameters effectively improves the K2 while maintaining a high V, meanwhile eliminates spurious modes. It is demonstrated that the Sezawa mode on the proposed SAW resonator structure offers operating frequencies above 5GHz, K2 values above 6.5\%, and an excellent impedance ratio of 98dB, which makes it a potential candidate for advanced 5G applications.

In this work, the shear horizontal surface acoustic wave (SH-SAW) resonators were demonstrated on 15° YXLiNbO3/SiO2/sapphire (LiNbO3-on-sapphire, LNOS) substrate. Compared to the reported gigahertz SAW resonators based on piezoelectric heterogeneous substrates, the fabricated resonator in this work exhibits a state-of-the-art electromechanical coupling coefficient (k2) of 42.2\%, a maximum Bode-Q (Qmax) of 1457 and an excellent figure of merit (k2×Qmax) of 615. Besides, several methods for suppressing transverse modes were implemented and compared. Tilted interdigital-transducers combined with the apodization technique can suppress the transverse modes more thoroughly while maintaining decent Q values. Overall, SAW devices based on the LNOS substrate have great potential for RF filters with low insertion loss, steep skirts, and wide bandwidth.

This paper investigates acoustic cross-coupling and remote excitation in an array of PMUTs (piezoelectric micromachined ultrasound transducers). Though undesired cross-talk can impact on PMUT array performance, the same can be also employed for remote excitation. The device array under study comprises of 7 PMUTs with constant pitch which is designed and characterized at the fundamental and higher order modes. The insights are employed to demonstrate a remote frequency filter and dual-channel excitation employing acoustic coupling.

The availability of Piezoelectric-On-Insulator (POI) substrates, made of a thin single crystal LiTaO3 film atop a silicon substrate, has promoted the development of innovative Surface and Bulk Acoustic Wave (SAW and BAW) devices. However, these substrates are so far only commercially available in 100 and 150 mm diameter. In this work, we successfully demonstrate acoustic devices based on 200 mm POI substrates. First, we fabricate SAW resonators displaying an electromechanical coupling coefficient of 8.8\% at a resonance frequency of 1.6 GHz. Then, we implement Film Bulk Acoustic Resonators (FBAR), integrating buried electrodes and an acoustic isolation structure, which exhibits a single resonance at 2.8 GHz, with an electromechanical coupling coefficient of 8.8\% and a quality factor close to 190. Eventually, we show a Solidly Mounted Resonator (SMR) based on a dielectric (AlN/SiO2) Bragg mirror, which exhibits performances close to AlN-based resonators, i.e. a coupling coefficient of 6.1\% and a quality factor of 405 at 4 GHz. For the later, a Temperature Coefficient of Frequency (TCF) of -14 and -22 ppm/°C at resonance and antiresonance are obtained respectively. Such TCF values are among the lowest ever reported for LiNbO3 and LiTaO3 BAW resonators. These results offer promising perspectives towards the development of 200 mm SAW and BAW filters based on POI substrates.

In this paper, a 30° YX-Lithium Niobate (LN) 0-th shear horizontal (SH0) plate acoustic wave (PAW) resonator is proposed. The SH0 mode characteristics the superiority of interdigital transducer (IDT) in the frequency definition over most other plate modes. Using finite element analysis method, the rotation angle of LN and the thickness of each layer were optimized for large effective coupling coefficient (k2eff) and high acoustic velocity. The rotation angle and the thickness of LN membrane are optimized as 30° and 0.2, respectively. To improve the temperature stability of proposed PAW resonators, a SiO2 film are added and the thickness is designed as 0.2. The measurement results derived a k2eff of 25.1\%, a Bode-Qmax of 604, and a Figure of merit (FoM) of 151, which is higher than the reported similar-type PAW resonators. The measured first-order temperature coefficients of frequency at resonant frequency (TCFfs) and anti-resonant frequency (TCFfp) are -38ppm/°C and -26ppm/°C, suggesting the temperature stability improvement in comparison with only LN membrane-based resonators.

This paper presents a new method to suppress spurious modes in lithium niobate thin-film acoustic devices by twisting the piezoelectric coupling properties of the spurious modes. The excellent piezoelectric properties of lithium niobate (LiNbO3) advance performance but lead to significant spurious modes accompanied by the targeted main mode. To harvest the benefits and avoid the spurious modes, this work investigates solidly mounted LiNbO3 thin films with different substrates to twist the zero-coupling orientations of spurious modes to be close to the maximum-coupling orientation of the targeted main mode. The fabricated devices, based on the solidly mounted LiNbO3sapphire structure and surface guided acoustic wave, show an operating frequency of 2.4 GHz with a large electromechanical coupling of 22\% and a spurious-free response in the wide frequency range. This work could overcome a significant bottleneck in commercializing LiNbO3 thin-film acoustic devices.

This work proposes a novel one-port 3D acoustic resonator based on the lithium niobate thin film on conductive silicon carbide substrate (LiNbO3-on-SiC, LNCSiC). The fabricated resonator shows coupled frequency responses of the shear-horizontal surface acoustic wave (SH-SAW), the longitudinal leaky SAW (LL-SAW), and the high-overtone bulk acoustic waves (HBAWs). The HBAWs propagating in the thickness direction of LNCSiC show a wide frequency response span exceeding 4 GHz and an excellent maximum quality factor ( ) of 7980. The GHz SH-SAW propagating in the surface of LNCSiC show a large electromechanical coupling coefficient ( ) of 25.95\%, while the LL-SAW shows an extremely high velocity of \textasciitilde6900 m/s. Such hybrid resonators could potentially open up new applications in radio frequency communications, 3D imaging, and sensing.

The practical Internet of Things at the current stage still persists in handling an energy minimized network. For a proper network communication an energy consumption of 80\% is indulged only on the communication setup. 6LoWSD (6LoWPAN Software Defined) is an SDN based IoT network protocol developed to minimized the IoT constraints. The SDN’s feature of decoupling the controller plane from the data plane enhances the network efficiency. These target conducts towards data rate, traffic, throughput and duty cycling management. Besides these it also provides a sense of flexibility towards program-ability for the current IoT networks. Efficient power system is a highly Important domain which needed for handling the stability for the whole SDN-IoT system. An effort towards enveloping state transition schedulers for energy optimization has been experimented in this paper.

The Routing Protocol for Low power and Lossy networks (RPL) has been developed by the Internet Engineering Task Force (IETF) standardization body to serve as a part of the 6LoWPAN (IPv6 over Low-Power Wireless Personal Area Networks) standard, a core communication technology for the Internet of Things (IoT) networks. RPL organizes its network in the form of a tree-like structure where a node is configured as the root of the tree while others integrate themselves into that structure based on their relative distance. A value called the Rank is used to define each node’s relative position and it is used by other nodes to take their routing decisions. A malicious node can illegitimately claim a closer position to the root by advertising a lower rank value trapping other nodes to forward their traffic through that malicious node. In this study, we show how this behavior can have a detrimental side effect on the network via extensive simulations and propose a new secure objective function to prevent such an attack.

IoT will be capable to openly provide entry to selected data groups to enable the building of diverse digitized programs while also clearly and fluidly integrating a large range of different and unsuitable end devices. It is a highly challenging task to develop a common design for IoT due to the large variety of devices, connection layer technologies, and applications that could be incorporated in such a system. Urban Iot applications, while still a sizable segment, are the focus of this investigation. The target application domain of these algorithms sets them apart. Urban IoTs are actually created to support the idea of the "Urban Development," which aims to use the most modern networking technology to allow additional offerings for both the municipal government and the citizens. Thus, this article provides a full survey of technology options, rules and regulations, and building design for simply an urban IoT. This Padova initiative, that serves as a convincing example of an IoT offshore rollout conducted out in cooperation with the municipal administration inside the Italian province of Padova, will be covered in detail along with the methodological techniques and finest standards employed there.

Autonomous and Supported Lifestyle (AAL) has been highlighted as a requirement in today s environment in a number of theories, techniques, and different uses for the Internet of Things. (IoT). Technologies standardization initiatives like Wireless V4.x (Wireless smart), for example, have sparked a meteoric rise in creative relatively brief wireless devices that can provide a variety of services to AAL. Additionally, new potential for major carrier is created by enabling equipment (Sq.m) connectivity between all of these technologies. To support M2M exchanges, telecommunications companies, especially telecom companies, might have to build new infrastructure and rethink their corporate objectives. Simple Square meters or IoT products often need another suitable tool, like a telephone, to serve as a doorway to the World wide web in order to function to their fullest capacity. The unique Concept of Iot examined in this study enables any nearby Innertubes device to serve as an M2M entry point for Internet of things. As a result, the user of a Sensor node no longer has to own a smartphone or other Innertubes equipment in order to access capabilities like internet - based. In this research, an unique IoT architectural prototype system for short-range signal repeaters is described. The test bed s installation, benefits and drawbacks, and sampling analysis using data acquired from a real-world event are discussed, and the findings are positive.

In this research, a power consumption analysis of wireless devices for Internet of Things applications is described. The research analyzes and contrasts a variety of tiny wireless communication techniques and their modules, including ZigBee, Energy Saver Wi-Fi, Six-Low-PAN, and LPWA, all of which aim to conserve energy and lengthen the lifespan of the devices that make up an IoT network. This focuses on the significance of employing small wireless techniques and components in IoT applications. The study s methodology is defined by the individual module used to implement the protocol. According to the degree of communication between sensor nodes, the proposed protocols are categorized. ZigBee, 6LoWPAN, and low power Wi-Fi are the candidate protocols for connectivity over short distances. The LoRaWAN protocol is a possibility for long-distance connectivity. Given the wide variation in power consumption between modules and protocols, the results of this study demonstrate how carefully selecting units for every protocol can greatly affect the duration of its use. Accordingly, protocols are compared with one another in various ways based on the module in question.

Proposed system, pollution monitoring, the automobile industry, and sports are just a few of the application areas that have grown as a result of ubiquitous sensing and the distinctive features (Sensor systems). As the underlying significantly expanded the number of linked things with realtime communication and data computation, WSNs have grown in importance in recent years. However, owing to the scale and accessibility of IoT, building a complex challenge, and past methodologies established for Iot technologies cannot be implemented directly. In this paper, pairwise clusters models for Iot networks in the Iot paradigm are proposed: I a resource grouping model and (ii) a business clusters model where responsibilities are allocated to individual sensor nodes depending on how well they provide services. The end-to-end latency, and communication bandwidth balancing.

The resource-constrained IPV6-based low power and lossy network (6LowPAN) is connected through the routing protocol for low power and lossy networks (RPL). This protocol is subject to a routing protocol attack called a rank attack (RA). This paper presents a performance evaluation where leveraging model-free reinforcement-learning (RL) algorithms helps the software-defined network (SDN) controller achieve a cost-efficient solution to prevent the harmful effects of RA. Experimental results demonstrate that the state action reward state action (SARSA) algorithm is more effective than the Q-learning (QL) algorithm, facilitating the implementation of intrusion prevention systems (IPSs) in software-defined 6LowPANs.

Scientific and technological advancements, particularly in IoT, have greatly enhanced the quality of life in society. Nevertheless, resource constrained IoT devices are now connected to the Internet through IPv6 and 6LoWPAN networks, which are often unreliable and untrusted. Securing these devices with robust security measures poses a significant challenge. Despite implementing encryption and authentication, these devices remain vulnerable to wireless attacks from within the 6LoWPAN network and from the Internet. Researchers have developed various methods to prevent attacks on the RPL protocol within the 6LoWPAN network. However, each method can only detect a limited number of attack types, and there are still several drawbacks that require improvement. This study aims to implement several attack prevention methods, such as Lightweight Heartbeat Protocol, SVELTE, and Contiki IDS. The study will provide an overview of these methods theories and simulate them on Contiki OS using Cooja software to assess their performance. The study s results demonstrate a correlation between the simulated data and the proposed theories. Furthermore, the study identifies and evaluates the strengths and weaknesses of these methods, highlighting areas that can be improved upon.

IoT technology establishes a platform for automating services by connecting diverse objects through the Internet backbone. However, the integration of IoT networks also introduces security challenges, rendering IoT infrastructure susceptible to cyber-attacks. Notably, Distributed Denial of Service (DDoS) attacks breach the authorization conditions and these attacks have the potential to disrupt the physical functioning of the IoT infrastructure, leading to significant financial losses and even endangering human lives. Yet, maintaining availability even when networking elements malfunction has not received much attention. This research paper introduces a novel Twin eye Architecture, which includes dual gateway connecting every IoT access network to provide reliability even with the failure or inaccessibility of connected gateway. It includes the module called DDoS Manager that is molded into the gateway to recognize the dangling of the gateway. The effectiveness of the proposed model is evaluated using dataset simulated in NS3 environment. The results highlight the outstanding performance of the proposed model, achieving high accuracy rates. These findings demonstrate the proposed network architecture continues to provide critical authentication services even upon the failure of assigned gateway.

The growing Internet of Things (IoT) has led to an increasing number of interconnected devices across diverse locations. To enable efficient data transmission in resourceconstrained IoT networks, selecting the right communication protocols is crucial. This study compares the performance of 6LoWPAN-CoAP and RPL-CoAP in LoRaWAN networks under limited settings, focusing on Packet Delivery Ratio (PDR) and latency. Tests with simulated LoRaWAN settings were conducted at various scales to evaluate both protocols’ scalability and dependability. The findings demonstrate that RPL-CoAP outperforms 6LoWPAN-CoAP in constrained LoRaWAN scenarios, consistently showing higher PDR and reduced latency. The RPL routing algorithm’s inherent characteristics contribute to this improved performance, effectively constructing routes while considering energy usage and link quality. Additionally, the study highlights LoRaWAN networks’ inherent PDR benefits over conventional networks, making the RPL-CoAP and LoRaWAN combination a powerful option for IoT applications in limited settings. These insights can guide the design of reliable and effective IoT applications in resource-limited environments, maximizing the IoT ecosystem’s potential.

Wave filtering is one of the mandatory features of the state estimators in a dynamic position system. The optimization of statistical parameters of these state estimators can be done by covariance matching algorithms and appropriate objective (cost) functions. The proposed cost function has predictive behavior, based on some tuning parameters, which control the quality of wave filtering. These parameters assure convergence of the solution and consistent results in different adaptive algorithms based on the Kalman filter framework as AKF, AEKF, and AUKF.

An adaptive command filter control of nonlinear system with friction input is formulated in this paper. First, based on the obtained state space model, a command filter control method is proposed, which can address the “explosion of complexity” problem existed in traditional backstepping design and ensure the asymptotic convergence of the tracking errors. Moreover, to cope with the problem of filter error between filter output and virtual control signal, dynamic error compensation system is designed. Next, a HONN system is employed to simplify the calculation and approximate the uncertainties in the system. At last, in order to clarify the effectiveness of the above theory, simulation results are given.