Alzheimer's disease (AD) is the most prevalent type of dementia, resulting in gradual memory and cognitive impairment. Radiomic characteristics acquired from brain MRI have shown significant promise as non-invasive indicators for this illness. However, their use for particular brain areas has not yet been investigated. To study Alzheimer's disease (AD), conventional machine learning approaches have evolved from image decomposition methods like principal component analysis to more complicated, non-linear algorithms. Now that the deep learning paradigm has arrived, it's feasible to extract high-level abstract features directly from MRI scans that characterize how data is distributed in low-dimensional manifolds internally. In this paper, we proposed a new Convolutional Neural Network based architecture for classifying Alzheimer's disease. The model is also evaluated using performance measures like precision, recall, f1 score, confusion matrix, etc. The proposed model was evaluated and achieved a classification accuracy of 97% for the classification of 4 different classes, which surpassed almost all cutting-edge technologies.
Today, designing a reliable and efficient step-up converter with high voltage gain for Photovoltaic applications has been a major concern for designers. This work proposes a fully soft switched high step-up converter that combines synchronous Cuk and Boost converters utilizing the shared component approach. The advantages of this converter compared with traditional Cuk and Boost converters are higher voltage gain and zero switching loss. A PSpice simulation was done at 225w power, 24 V input voltage, 150 V output voltage, and 100k Hz switching frequency to verify the proposed converter's functioning and the efficiency 96.56% has been achieved at mentioned operating power. Additionally, a bond graph model is designed to examine the dynamic behavior of the converter as it is presented, taking into account parasitic parameters and the specifications of realistic components. The bond graph model's simulation results have been explored. In the final section of this study, a PID controller employing the state space averaging approach is designed using SISOTOOL in MATLAB. Additionally, a Simulink model of the closed loop proposed converter is designed, and simulation results are reported.
This paper discusses interference cancellation in an Uplink Multi-User (MU) Multiple Input Multiple Output (MIMO) system, in which a Full Duplex (FD) Decode and Forward (DF) relay is used. In order to decode the symbols at the relay, Block Diagonalization (BD) based processors are applied for MU interference cancellation. Moreover, the self-interference signal at the relay is suppressed by using the equivalent relay concept. Minimum Mean Squared Error (MMSE) technique is then adopted at the relay for information extraction. For MU interference cancellation and information extraction at the Base Station (BS), a similar procedure is applied. Simulation results show the superiority of the proposed system to the Amplify and Forward (AF) relaying system. In addition, we provide the appropriate number of antennas and transmit powers for each setup in order to achieve a relatively high sum rate.
Next word prediction for technological devices has become a contentious issue in predictive analytics. Word pre-diction requires a huge amount of data from a vast number of different users, which introduces issues like heterogeneous dataset management and privacy concerns, thus raising the importance of Federated Learning (FL) as it can learn from a large amount of sparsely distributed data while ensuring privacy. Further investigations on related studies have led us to contribute in this area based on the Bangla language due to the minimal appearance of this language in the domain of word prediction based on FL. In consideration of implementing our work, some of the prime issues have been noted where avoiding data poisoning attacks, prioritizing user preferences during the global update, and focusing on fast convergence to reach a high-quality model in a few training rounds have become unavoidable matters of contention. Besides this, generating initial models which can be easily adapted by the current or new users is also an important research problem. Based on these, we have proposed a structure that combines semantic analysis and personalized FL using Model-Agnostic Meta-Learning (MAML) to obtain the desired goals. Extensive evaluations have been done on Non-IID data for both FL and personalized FL with MAML on the next word prediction model, which justifiably achieved 83.05% and 86.23% test accuracy respectively on the server model after ten communication rounds across ten clients. Apart from that, the effective results of semantic approaches show an impressive outcome as intended. Thus, the limitations and the challenges of FL on next word prediction conceptually allowed us to use a variety of algorithms to turn them into advanced ones for further usage.
This study presents a design for a 5G NR (New Radio) Microstrip Patch antenna (MSP A) operating in the n257 band, which is more commonly known as the 28 GHz band. The simulations were used to determine its free space properties. The initial findings were optimistic. The antenna produced a 4.26 GHz wide bandwidth with a substantial amount of gain above 6.8 dBi and 74% efficiency. To analyze its on-body performance, a virtual human torso phantom was created, and simulations were done again by placing the antenna over the torso with a 4 mm gap. The antenna sustained its wide bandwidth with a high gain of over 8 dBi. Even in the on-body scenarios, the antenna's efficiency remained close to 70%. The antenna is designed for the 5G-enabled Body Centric Network (BCN), which may be used in a wide range of smart medical applications.
Accurate disease categorization using endoscopic images is a significant problem in Gastroenterology. This paper describes a technique for assisting medical diagnosis procedures and identifying gastrointestinal tract disorders based on the categorization of characteristics taken from endoscopic pictures using a vision transformer and transfer learning model. Vision transformer has shown very promising results on difficult image classification tasks. In this paper, we have suggested a vision transformer based approach to detect gastrointestianl diseases from wireless capsule endoscopy (WCE) curated images of colon with an accuracy of 95.63%. We have compared this transformer based approach with pretrained convolutional neural network (CNN) model DenseNet201 and demonstrated that vision transformer surpassed DenseNet201 in various quantitative performance evaluation metrics.
Internet of Things (IoT) empowers us to connect physical objects ranging from smart buildings to portable intelligent devices, such as IoT wearable devices. The principal advantages of IoT wearable devices are allowing us remotely to collect data and also giving us the capability to control and monitor objects. These features can be utilized in various ways, such as establishing a connection, locating devices, authenticating users, protect users' privacy, sharing data and which is central to this proposed paper. Data sharing is receiving a widespread application in our daily lives since it facilitates cooperation between two ends, e.g., User-to-User, User-to-Devices, Devices-to-Devices, etc, and provides services to both ends or one of them. Data sharing can be granted using different factors, one of which is something in a user's/an IoT device's environment which is in this paper broadcast signals. Using broadcast signals to measure Received Signal Strength Indicator (RSSI) values and Machine Learning (ML) models, this paper implements an IoT data sharing scheme based on something that is in a user's/an IoT device's environment. The proposed scheme is experimentally tested using different ML models and shows 97.78% as its highest accuracy.
On the basis of ensuring the communication performances of different kinds of communication services in the emergency communication systems, the access fairness of various types of communication services should be considered when we design an emergency communication system. Based on the perceptual classification of emergency communication traffics, this paper proposes a dynamically resource allocation method for emergency communication systems according to the weighted priorities of communication traffics. In the method, the scheduling priority index of different services are calculated according to their priorities and time urgency, then calculates the bandwidth resource allocation weights of the traffic are reckoned, and their upper bandwidths allocated are obtained. Finally, the bandwidths are allocated according to their needs and upper bandwidths. Simulation results show that the proposed dynamic resource allocation method has the same performance in bandwidth efficiency but higher access fairness compared with the traditional absolute priority resource allocation method.
The control of the electromagnetic spectrum is of great importance to guarantee military operations, communication of the civil population, strategic government communication, among others. The main difficulty to have a control and guarantee a use according to the norms of each country of the spectrum, is in the high cost of the control equipment and the need for permanent operators to process the information of these sensors. In the present work, a low-cost automatic carrier detection system is proposed, which allows deploying a greater number of sensors and with automatic detection automatically emit alerts in case of use of restricted frequencies. This facilitates the control of the spectrum, especially of critical bands for military operations, being the first step to maintain the sovereignty of the spectrum.
This paper presents an autonomous robot for industrialapplication, controlled via voice commands given by its user. Speech recognition is seen as one of the most promising interfaces between humans and robots, since it is probably the most natural and intuitive way of communication between human. Despite this fact, voice is rarely utilized in industrial robots. Therefore, this paper presents an application where an industrial robot is taught and instructed through voice commands rather than learning from demonstration. In addition to voice, the manipulator was equipped with acquiring vision information from the surrounding, in order to identify objects that are to be manipulated. An experiment is conducted to evaluate the behavior of the integrated voice and vision systems in a simulated environment. Findings of the experiment are presented. Identified capabilities of the proposed system and possible improvements are discussed.
In this paper, a single input single output Non-Orthogonal multiple access (NOMA) communication networks over a Rayleigh fading channel in the presence of an adversarial jammer is considered. The impact of different jamming techniques (Barrage Jamming, Partial-Band Jamming, and two schemes of Pulse Jamming) on the outage probability of the NOMA system is analytically derived. The throughput of the NOMA system is also analytically derived in the presence of the mentioned jamming techniques. In the end, a comparison among the different jamming techniques is held to propose the optimum jamming technique against the NOMA communication system according to the appropriate application of the adversarial jammer. All the derived analytical results are validated by extensive Monte-Carlo simulation results.
Product ciphers composed of multi-byte permutations and substitutions are generally considered to be more secure than the substitution ciphers. However, this is not the case. In fact, product ciphers using the same boundaries can be isomorphically reduced to substitution ciphers. These product block ciphers are characterized by block definitions having the same defined block sizes. This also means that such product ciphers can be attacked using the same methodology as can be applied to all substitution (S) ciphers. We show that Permutation-Substitution-Permutation (PSP) ciphers are insignificantly more secure than a substitution cipher and propose a chosen plaintext attack for solving these product ciphers.
The direction of arrival (DoA) angles estimation of several narrowband RF sources in multipath environments is crucial in real-time applications. In this paper, a new DoA algorithm is presented to reduce the computational complexity of estimating the DoA's of highly correlated signals in pairs while maintaining estimation accuracy. The method proposed employs Cholesky decomposition in conjunction with a novel forward/backward operation (FBO) applied to the signal subspace matrix to de-correlate the incident coherent sources while most of the existing methods employ conventional pre-processing techniques to the covariance matrix. Employing FBO to the signal space matrix along with Cholesky decomposition results in significant reduction in the computation time and the complexity of hardware implementation. These advantages make the method proposed well-suited for applications such as in massive-MIMO systems as well as in enabling next generation wireless communication networks (beyond 5G and 6G). Simulation results using Root Mean Square Error (RMSE) and results from real-time implementation on a Software Defined Radio (SDR) testbed are used to validate the efficacy of the proposed method. Results show that the performance of the method proposed compares favorably with the Root-MUSIC method.
This paper examines how a user's preference of device while browsing social media may affect his/her risk of falling prey to misinformation and disinformation attacks. An experimental survey $(\mathrm{N}=327)$ was deployed across the United States, where participants were tested on their ability to discern facts against misleading news content. The findings strongly suggest $(\mathrm{p}=.000)$ that device preference for social media use is a strong determinant of misinformation and disinformation vulnerability and that users who preferred desktop computers such as PCs and laptops performed poorly compared to their counterparts who preferred handheld devices such as smartphones and tablets. This paper hopes to reach technology policy experts, digital forensic officers, communication academics, and policymakers that may find this area of cyber deception to their interest.
Multi-processing System-on-Chip (MPSoC) field-programmable gate arrays (FPGAs) are increasingly being employed in cloud computing platforms to accelerate many domain-specific applications. In cloud computing, FPGAs are constantly shared among multiple tenants dynamically by leveraging the partial reconfiguration property of FPGAs. Such devices are embedded with hundreds of large and configurable memory blocks, which offer the most flexible and high-performance memory resources for processor cores and IPs. The bare-metal access model of the unprotected memory system of MPSoC FPGAs has raised various security and privacy concerns. In this paper, we present the design and implementation of our novel run-time memory protection scheme based on hardware isolation for cloud MPSoC F-PGAs. Experimental results on a Xilinx Zynq Ultrascale+ development board demonstrate our protection design successfully limits each tenant to only their allocated portions of the memory space and protects memory against programming errors and malicious attacks.
Routing is the basis for the transmission of mobile ad-hoc networks (MANET). Global routing can achieve theoretically optimal transmission efficiency, and it is crucial to use network information to filter out the optimal routing. This paper proposes a multipath routing performance prediction model based on a graph attention network (GAT). The ability of GATs to model the structure and function of data makes it possible to determine the characteristics of links, paths, nodes, etc., and the complex relationships between them, to predict network throughput and stability. Finally, using a software-defined network (SDN), routing can be optimized by predicting the network performance of alternative routings. The simulation results showed that this method improves throughput by 5-25%, and network stability by 10-30%, under dynamic topology conditions compared with other SDN-controlled MANET routing protocols.
Due to his charisma. exceptional insight and logical capabilities Nobel Laureate Richard Feynman has had a decisive role in how the quantum computing industry has evolved. He equated nature's apparent probabilistic dynamics with the need for a different kind of probabilistic computer, beyond the classical digital computer, that could sufficiently imitate this, and worked out the initial mathematics of such a computer. This has been the genesis of the hardware and software development of today's entire quantum computing industry. Yet, by piercing behind Nature's probabilistic veil based on a light-based quantum computational model, it becomes possible to enhance Feynman's conceptualization and resulting mathematics to do with simulating time, space, and nature's unpredictability itself, to create a foundationally different quantum computer. This would result in reinterpreting foundational quantum equations such as Heisenberg's Uncertainty Principle, Schrodinger's Wave Equation, and Euler's Identity, to thereby conceptualize a fundamentally different type of quantum computation leading to an entirely different genre of emergent-type quantum computational devices. The very foundation of the entire quantum computing industry would be changed as a result.
Non-orthogonal multiple access (NOMA) scheme en-ables serving users with the same resource block i.e. frequency or time by multiplexing the signal of the users'. Intelligent reflecting surfaces (IRS) or reconfigurable intelligent surfaces (RIS) is a potential approach for increasing transmission efficiency by mod-ifying signal propagation by tweaking typically passive reflective components. IRS reconfigures the wireless network in this manner to improve system performance. The research considered an IRS-assisted downlink NOMA system. This work modified the reference model by adopting an IRS-specific frequency-distance- dependent path loss model for IRS-NOMA downlink received power and signal-to-interference plus noise ratio (SINR) measurement instead of a typical or conventional distance-dependant path loss model. The incorporation of an IRS-specific path loss model provides a more convenient and better measurement compared to the conventional path loss model. Furthermore, this work figures out an improvement scope in the reference model.
In this paper, three different technologies were used to measure lithium battery charging behaviors. Graupner ultra duo 60 plus charger, EVM BQ27546, and LTC4054 chargers were used to compare the functionality of Li-Ion, LiFePO4, and Li-Po batteries. Charging at different C-rates was analyzed. Logged data from the chargers give information about the batteries charging characteristics, e.g., voltage, current, capacity, internal resistance, and efficiencies. The results showed that Graupner and EVM chargers were following the CC-CV algorithm, while the manufacturer of the LTC4054 battery charger says that it is using the CC-CV method, but our results showed that the current starts decreasing from the very beginning until the battery is fully charged. From the data, it can be concluded that LiFePO4 is the best choice at recommended 1C-rate as the charging efficiency of the battery is 100% as compared to 96% and 90% for Li-Po and Li-Ion, respectively. From the research, it is considered that a charger is equally essential as they must be compatible depending upon the battery technology.
