https://tjee.tabrizu.ac.ir/ Tabriz Journal of Electrical Engineering en daily 1 Mon, 22 Dec 2025 00:00:00 +0330 Mon, 22 Dec 2025 00:00:00 +0330 https://tjee.tabrizu.ac.ir/article_20463.html A clamp-conductor system in power transmission lines serves two main functions: to withstand the conductor’s tensile forces and to ensure reliable, continuous electrical current transfer. This study investigates the mechanical and thermal performance of a typical dead-end tension clamp-Lynx conductor system. A finite element model was developed to simulate the system’s response to tensile and thermal stresses. The results indicate that the highest tensile stresses occur in the conductor’s steel core and the clamp’s steel anchor. Thermal performance was evaluated through both finite element simulation and laboratory current-injection testing. Findings show that the conductor’s surface temperature is consistently higher than that of the clamp, and both temperatures increase with rising current. The maximum difference between simulated and experimentally measured conductor surface temperatures was less than 2%, validating the accuracy of the simulation approach. https://tjee.tabrizu.ac.ir/article_19626.html The stock market’s significance in the global economy necessitates demands more accurate prediction methods. This paper introduces a novel hierarchical attention mechanism aimed at enhancing the performance of predicting stock price movements. Hierarchical attention networks assume that not all news segments hold equal relevance in forecasting stock market trends. Furthermore, we assert that not all daily news carries an equivalent significance in predicting market trends. To tackle this challenge, we suggest a hierarchical attention network plus BERT that emulates the news hierarchy and assigns weights to news items based on their significance, and also the most informative news articles in each trading day in stock market prediction. Our HAN+BERT method incorporates three levels of attention mechanisms, operating at the word, sentence, and news level. This allows the model to identify the most significant news stories of the day and select the most informative sentences and words within these articles. Using BERT as the word embedding approach has resulted in better performance for our stock trend prediction model. Empirical results on Persian financial news and three stock market indices reveal the effectiveness of our HAN+BERT model, with a peak accuracy of 65.49%, which is 3% higher than the best baseline model. https://tjee.tabrizu.ac.ir/article_19939.html With the continuous advancement of Software-Defined Networks (SDNs), the adoption of a distributed control plane architecture has become increasingly necessary. One of the primary challenges in these networks is the variable load on the controllers where high loads can lead to congestion. Such congestion can significantly degrade network efficiency. Although previous studies have attempted to address this issue, they have largely failed to effectively manage load exchange between the control plane and the data plane. This paper proposes a migration-based congestion control mechanism for multi-domain SDNs. In this approach, when a controller experiences high load and congestion, selected switches are migrated from the overloaded controller to one with a lower load. If the migration risks congesting the new controller, the mechanism swaps switches between controllers with minimal migrations, drawing inspiration from the Kadane algorithm to prevent congestion elsewhere. The proposed mechanism was evaluated using the D-ITG and IPerf tools with the RYU controller, demonstrating improved system performance. Simulation results show that the mechanism outperforms the baseline approach, increasing average network throughput by approximately 10%, while reducing average delay and jitter by about 30% and 25%, respectively. Furthermore, a comparison between the proposed method and the OptiGSM method reveals that the proposed method offers superior throughput and lower delay, although the OptiGSM method exhibits less jitter than the proposed method https://tjee.tabrizu.ac.ir/article_19443.html Cell free massive MIMO system is a promising candidate to overcome the shortcomings of the conventional cellular system and a suitable technique to achieve the goals of the sixth generation (6G) networks. Costs of implementing on large scale is one of the major disadvantages of this system. Recently, with the emergence of reconfigurable intelligent surfaces (RIS), we can hope for the reduction of these costs and the practical implementation of this system. In this regard, in this paper we have investigated the spectral performance of the scalable form of cell free with and without RISs. Lower bounds of uplink spectral efficiency involving imperfect channel estimation and pilot contamination effects for L-MMSE and LP-MMSE decoders is derived. The simulation results confirm the superior spectral efficiency performance of RIS-based proposed system (especially in centralized RIS- cell free mode) compared to the Scalable cell free system. https://tjee.tabrizu.ac.ir/article_19444.html This paper presents a robust sliding mode control approach based on integral backstepping for multi-input multi-output (MIMO) nonlinear systems. The proposed control scheme addresses input saturation, modeling uncertainties, and time-varying external disturbances. A novel auxiliary design system and Nussbaum gain functions are incorporated into the control scheme to tackle input saturation. The proposed control approach is applied to a nonlinear epidemic model. The model of flu epidemiology, which includes five non-negative state variables representing the susceptible, exposed, infected, asymptomatic, and recovered individuals, along with three control inputs for vaccination, antiviral treatment, and social distancing, is being studied. Simulation results demonstrate the effectiveness of the proposed control scheme in handling input saturation and achieving accurate trajectory tracking, highlighting its potential for uncertain nonlinear systems with input constraints. https://tjee.tabrizu.ac.ir/article_19442.html Integrating Distributed Energy Resources (DERs) with peer-to-peer (P2P) energy trading offers promising solutions for grid modernization by incentivizing prosumers to participate in mitigating peak demand. However, this integration also introduces operational uncertainties and computational challenges. This paper aims to address these challenges with a novel scalable and tractable distributionally robust joint chance-constrained (DRJCC) optimization framework that effectively facilitates P2P energy trading by enhancing flexibility provision from large-scale DER operations under uncertain supply and demand. Therefore, a practical framework is proposed to solve the core challenges of DRJCC by integrating three key components: (1) a Wasserstein ambiguity set that effectively quantifies uncertainty with sparse data, (2) a CVaR-based approximation of joint chance constraints to balance computational efficiency with risk control, and (3) a privacy-preserving ADMM algorithm that enables distributed implementation through decomposition. To discern patterns in the data that indicate collaboration potential and adjust ambiguity sets for improved efficiency, K-means clustering is applied to historical scenarios. Simulation results show that the proposed framework reduces peak demand by approximately 28% and total community costs by around 31%, underscoring its effectiveness in enhancing grid robustness, operational reliability, and economic optimization in renewable-based energy management. https://tjee.tabrizu.ac.ir/article_19943.html Lane detection is an important component in the development of autonomous vehicles, facilitating the real-time identification of driving paths and compliance with traffic regulations. Despite the promising performance of current models in controlled environments, they often encounter significant challenges in real-world scenarios, such as occluded lane visibility caused by snow, dust, traffic, or the absence of lane markings. This study presents a new approach to lane detection that leverages the spatiotemporal attributes of video frames by combining Long Short-Term Memory (LSTM) networks with Convolutional Neural Networks (CNNs) to enhance performance in the presence of occlusions. We employ a CNN to extract high-level spatial features from video frames, while an LSTM aggregates these features over time to model temporal dependencies and infer occluded segments when visual cues are absent. By representing lane marking detection as a sequential learning problem, the combined CNN-LSTM network effectively extracts spatiotemporal features. This dual architecture integrates both spatial and temporal information, thereby increasing robustness against occlusions and varying lighting conditions. The proposed model was evaluated under two conditions: low and high occlusion, using separate datasets, and was compared with the baseline architecture. The results confirm the effectiveness of the proposed approach. In low occlusion conditions, the model achieves an F1 score of about 96%, similar to the baseline method. In contrast, the baseline model suffers a performance drop in high occlusion scenarios, while the proposed model remains robust, also achieving an F1 score of about 96%. https://tjee.tabrizu.ac.ir/article_20397.html The increasing integration of renewable energy sources (RES) in hybrid microgrids has introduced new challenges in maintaining stability, reliability, and optimal performance. This paper proposes a novel control framework that combines deep reinforcement learning (DRL) with digital twin (DT) technology to address these challenges. The DRL agent is trained in a virtual DT environment, enabling rapid learning and optimization of control strategies under dynamic conditions without risking real-world operations. The proposed method is tested on a hybrid microgrid comprising photovoltaic (PV), wind, and battery storage systems. Simulation results demonstrate that the DRL-DT framework achieves a 28.5% improvement in energy efficiency compared to conventional model predictive control (MPC). Additionally, the proposed approach enhances system stability by reducing voltage fluctuations by 21.3% and achieves a 32.7% reduction in load shedding during peak demand scenarios. The training time for the DRL agent is reduced by 40% due to the efficient simulation capabilities of the DT. These results highlight the robustness and adaptability of the DRL-DT framework, making it a promising solution for next-generation hybrid microgrid management. This study provides a significant step forward in leveraging artificial intelligence and digital twins to optimize hybrid microgrid operations, ensuring sustainable and resilient energy systems. https://tjee.tabrizu.ac.ir/article_20543.html This paper introduces an algorithm for the Local Electricity Market Operator (LMO) in distribution networks to assess dynamic capacity withholding by Virtual Power Plants (VPPs). Its main contribution is providing ex-ante indices to evaluate this withholding. The paper also quantitatively analyzes how LMO's risk aversion impacts the market power exerted by these withholding groups. The day-ahead market problem is tackled in three stages: estimating VPPs' optimal withholding strategy, determining optimal system scheduling, and analyzing network configuration changes to reduce market power. The results indicate that the exercise of market power by capacity-withholding groups increases the LMO's cost by an average of 35%. Furthermore, the simulation of collusion in the proposed model reveals that the market power of capacity-withholding groups increases by an average of 5% under the risk-averse behavior of LMO. Moreover, the results demonstrate that proactively restructuring the network according to the proposed algorithm on the 123-bus IEEE test system can reduce the market power of capacity-withholding groups and lower the average index to 25.12%. https://tjee.tabrizu.ac.ir/article_19992.html Photomultiplier tubes (PMTs) require a stable, low-ripple high-voltage power supply with stringent load regulation and line regulation criteria. This paper introduces an innovative high-voltage power supply capable of converting a 220V AC input voltage into a DC output voltage that can be adjusted from -100V to -3kV. In the event of an arc or short circuit, the output voltage abruptly reduces to limit the output current to its rated value of 5mA. This system employs both linear and switching converters to yield a consistently smooth output voltage with voltage ripple as low as 400mV. The switching converter utilizes a push-pull converter while the linear converter meticulously regulates the output voltage. The paper outlines design considerations for the appropriate selection of components. Simulation results validate the performance of this proposed power supply under a variety of conditions, including input voltage and reference voltage variations, abrupt load changes, and short-circuit scenarios. Furthermore, experimental results from a fabricated prototype confirm the functionality of this high-voltage power supply for PMT applications. https://tjee.tabrizu.ac.ir/article_17757.html The Stack Overflow website is one of the most popular communities used by millions of programmers. If we consider a question and its corresponding answers as a knowledge unit on the Stack Overflow website, then there are different semantic relationships between two knowledge units, which include duplicate, direct, and indirect relationships with the proposed question. Recognizing different categories of semantic relationship between knowledge units in Stack Overflow can significantly improve the effectiveness and efficiency of information search. In this study, a hybrid approach based on deep learning methods and traditional similarity criteria is presented to detect the relationship between questions. In particular, two deep network architectures are presented, the first architecture consists of a long short-term memory network as well as a cosine similarity calculation layer. The second architecture is an extension of the first architecture by adding an attention mechanism. The proposed approach was evaluated on a dataset of Java programming language contining 40000 questions. The obtained results show that in terms of F1, Recall and Precision, the proposed model performs better than the existing models. Specifically, the model proposed in this article has a 17.3% improvement in terms of F1 measure compared to the best current model. Also, the results of the experiments show that using the pre-trained word embedding model significantly improves the performance of the presented models. https://tjee.tabrizu.ac.ir/article_17838.html We address the throughput maximization problem for downlink transmission in DF-relay-assisted cognitive radio networks (CRNs) based on simultaneous wireless information and power transfer (SWIPT) capability. In this envisioned network, multiple-input multiple-output (MIMO) relay and secondary user (SU) equipment are designed to handle both radio frequency (RF) signal energy harvesting and SWIPT functional tasks. Additionally, the cognitive base station (CBS) communicates with the SU only via the MIMO relay. Here, several combined constraints of the main problem complicate the solution. Therefore, we apply heuristic guidelines within the convex optimization framework to handle this complexity. First, consider the problem of maximizing throughput on both sides of the relay separately. Second, each side progresses to solve the complex problem optimally by adopting strategies for solving sub-problems. Finally, these optimal solutions are synthesized by proposing a heuristic iterative power allocation algorithm that satisfies the combinatorial constraints with short convergence times. The performance of the optimal proposed algorithm (OPA) is evaluated against benchmark algorithms via numerical results. The OPA is about 10000 times faster than the CVX solver. Additionally, the sum throughput especially at low thresholds increases by ~15-25% compared to the benchmark algorithms. Moreover, the constraints and objective function in OPA are 100% satisfied. https://tjee.tabrizu.ac.ir/article_18144.html Path planning for drones is one of the critical aspects of aerial operations, which holds significant importance in environments with complex entropy. Despite the differences in features and capabilities of drones, their interaction and coordination in various environments can enhance the performance of units. In this study, quadrilateral path planning for heterogeneous drones has been examined. By utilizing advanced adaptive clustering algorithms, drones are categorized into similar clusters and optimal paths are designed for each cluster. Simulation results indicate that the use of the algorithm introduced in this paper leads to a significant reduction in one of the most important components in drone routing, namely the deviation ratio, with the deviation ratio calculated in this paper being 1.1%, 7.52%, and 5.73% lower compared to other routing algorithms, such as HETRF, GA_HETRF, and APPA, respectively. The optimized results thereby increase routing accuracy compared to the mentioned methods and enhance the efficiency of the method presented in this paper. The results from the calculations and simulations of the paper show that efficient quadrilateral path planning can pave the way for optimal and efficient performance of heterogeneous drones in environments with complex entropy. https://tjee.tabrizu.ac.ir/article_18417.html This study investigates the challenges and methodologies in cross-corpus speech emotion recognition (CCSER), focusing on the generalization of speech features across diverse linguistic, speakers, and emotional contexts. We propose a novel SER system that leverages the transformer blocks of the HuBERT model combined with speaker embeddings and prosodic features to enhance feature extraction for emotion classification across different datasets. Our approach addresses dataset variability by utilizing transfer learning techniques, particularly through unsupervised methods that adapt feature distributions without requiring labeled data from target domains. Specifically, our transfer learning strategy employs a clustering method to select the most appropriate trained model for performing transfer learning from the source to target domains. We evaluate our proposed model using several datasets, including IEMOCAP as the source domain, and extend our validation to emotional datasets with different languages, demonstrating the adaptability of our system. The results indicate significant improvements in emotion recognition accuracy compared to traditional methods, highlighting the effectiveness of integrating advanced self-supervised learning models and transfer learning strategies in CCSER tasks. https://tjee.tabrizu.ac.ir/article_18533.html In this paper, three miniaturized filtering power dividers (FPDs) with arbitrary power-dividing ratio using the half-mode substrate integrated waveguide (HMSIW) based on the metamaterial concepts are proposed. The operational method of the presented structures is based on the theory of evanescent mode propagation. The meander technique has been employed to reduce the dimension of the conventional complementary split ring resonator (CSRR) unit-cell. In this technique, the meander slots are carved inside all of the interior space of the ring which is called modified complementary split-ring resonator (MCSRR). Accordingly, the length of the slot is increased which leads to an increase in the inductor and capacitor of the proposed structure without occupying the extra space. By means of the MCSRR unit cell, three miniaturized equal/unequal FPDs with power division ratios of 1:1, 1:4 and 1:8 are fabricated and tested. Additionally, to further size reduction of the proposed FPDs, the HMSIW platforms are used which, a nearly 50% reduction in the size can be achieved without changing the performance of the SIW structure. All of the proposed HMSIW FPDs are designed at 2.4 GHz which are suitable for WLAN applications. For demonstration of the applied procedures in the proposed HMSIW FPDs, the suggested equal/unequal HMSIW FPDs have been fabricated and measured. A reasonable agreement between simulated and measured results has been achieved. The entire size of the suggested equal/unequal HMSIW FPDs is about 0.07 λg × 0.07 λg. https://tjee.tabrizu.ac.ir/article_18680.html This paper presents a promising distribution post for the construction of a future active distribution network by designing a hybrid transformer with the ability to connect to DC microgrids. The topology of the proposed smart distribution substation consists of a direct power converter and a line frequency transformer. The combination of these two is introduced as a hybrid transformer. The main aim of designing a smart distribution substation with a DC link is to achieve power exchange between renewable energy sources (solar cells) and the distribution network, and as a result, adjust the voltage sag/swell and control the reactive power of the distribution network by means of the power electronics converter. The designed smart distribution substation uses a two-stage direct power converter with space vector modulation, which is connected to the distribution network through a series transformer. In the two-stage direct power converter, it is possible to connect to the DC microgrid. The steady state and dynamic performance of the proposed system have been confirmed by computer simulation and experimental results by prototype. These results prove the efficiency of the smart distribution substation in combining AC/DC networks and increasing the controllability and intelligence of the distribution network. It is also effective in correcting the power factor with the help of the DC link and as a result reducing the capacitor bank in the MV substation and also adjusting the voltage of the LV network. https://tjee.tabrizu.ac.ir/article_19441.html Alzheimer's disease is a neurodegenerative disorder characterized by a gradual decline in cognitive function, including memory loss and impaired thinking. Common symptoms include progressive memory decline, personality changes, difficulty with reasoning and decision-making, and decreased ability to perform daily tasks. Magnetic Resonance Imaging (MRI) can provide high-resolution images of brain structures, which can aid in identifying signs of brain disorders, including Alzheimer's. This study proposes a method for detecting Alzheimer's disease using MRI images from the Kaggle Alzheimer's disease database, which comprises four stages of the disease: healthy, very mild demented, mild demented, and moderate demented. The proposed method involves segmenting MRI images using Fuzzy C-means clustering and particle swarm optimization, followed by feature extraction using Histogram Oriented Gradients (HOG) and Local Binary Pattern (LBP). To address class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) is applied to generate a balanced dataset. Feature selection is then performed using the ReliefF algorithm. The evaluation results demonstrate that the proposed method achieves high accuracy (99.87%), sensitivity (99.74%), and specificity (99.91%) in classifying different stages of Alzheimer's disease, showing promise for early detection and diagnosis. https://tjee.tabrizu.ac.ir/article_19445.html AbstractThis article addresses the improvement of power system stability in the presence of SSSC with PMSG. A particular nonlinear control law based on a multi-input backstepping scheme is proposed. The main challenges in the power system are the instability caused by disturbances such as short circuits, large load disconnections, and changes in system topology. This control method effectively leads to a better performance in damping angular oscillations in comparison with conventional linear schemes. In this approach, a sliding mode observer is used to estimate time derivatives in the controller design formulas, and the Particle Swarm Optimization (PSO) algorithm is employed to determine the coefficients of the nonlinear controller. Additionally, the proposed method is robust against changes in system parameters and the location and timing of disturbances. This study demonstrates that the performance of PMSG is superior to that of DFIG in wind turbines at equal power levels and under identical circumstances in terms of damping oscillations. Implementation of PMSG leads to a more effective and rapid response with respect to applying DFIG. The new control law has been investigated on a 39-bus New England network using MATLAB coding. The simulations results indicate that the performance of the proposed method is superior to that of conventional linear techniques. https://tjee.tabrizu.ac.ir/article_19940.html With the transformation of traditional distribution networks (DNs) into active DNs, the possibility of congestion has increased. Also, the presence of uncertainty in factors such as electric car charging or solar panels, causes the predictability of congestion in the DN to decrease and the occurrence of congestion can be expressed as a probability. These cases have increased the concern of distribution system operators for the occurrence of congestion in DNs and also, for managing the amount of congestion with minimum costs. Therefore, providing solutions for congestion management in the shortest time and cost is a priority for operators. In this article, a solution for congestion management based on load shedding while considering the uncertainties of electric vehicles and solar panels is presented. At first, to estimate the amount of depth and time of congestion, modeling of congestion is done through probabilistic power flow. The evaluation of the proposed method in the Modified IEEE 33-bus home DN, while proving its effectiveness, shows that by using that, the operator performs the load shedding by minimizing their costs through improved multi-interval optimal power flow, and the amount of possible congestion is reduced depending on the level of risk-taking of the operator. https://tjee.tabrizu.ac.ir/article_19942.html During disturbances in the power system, the frequency deviates from its nominal value. The increased penetration of renewable ‎generations raises concerns about reducing system inertia and the system's capability to maintain frequency within an appropriate range. ‎In this paper, to enhance the participation of wind power plants based on variable speed generators in frequency control and in ‎coordination with thermal power plants, the parameters of the coordinated participation control system between wind turbines and ‎synchronous generator-based power plants are optimized. In tuning the control parameters, the NSGA-II algorithm is used to minimize ‎the frequency nadir and the rate of change of frequency (RoCoF) in the post-fault period. Simulation results of two test systems (the ‎single bus and Kundur two-area) show that the proposed scheme improves the frequency response characteristics of the system. ‎Specifically, it reduces the RoCoF, thereby preventing the unwanted operation of under-frequency-load-shedding relays. Additionally, ‎increasing the frequency nadir prevents the unwanted operation of frequency relays. Furthermore, the proposed method significantly ‎reduces the settling time of the system's frequency response.‎ https://tjee.tabrizu.ac.ir/article_19991.html Electrical nerve stimulation plays an important role in treatment of neural illnesses, as well as conducting the related research. Waveform is an important paratmeter of stimulation, for which, usually the rectgualar type is selected in commercial products and in research, for the sake of simplicity. However, in some references, the superiority of Gaussian waveform over rectangular, and the lower power consumption for its gerenration is poined out. In this paper, after a literature survey of neural stimulation, several Gaussian generator circuits from previous work are introduced, and finally, a new design for Gaussian signal generation is proposed. The presented circuit takes advantage of direct implementation of Gaussian signal, which has the general form of exp(-x2/2). In this circuit, a subtractor block is used to change the waveform peak point, a multiplier is used for square function generation, an exponential converter is used to generate exp(x) signal, a driver is used to amplify output current, and a H-bridge switching network is used for generation of bipolar waveform. The circuit is designed and simulated in 0.18µm CMOS technology, and has the capability of adjusting output current and stimulation pulsewidth. Static power consumption of the proposed circuit is 2.89W from 0.7V and 3.3V power supplies. Simulation results of the circuit is presented for various PVT corners, to show its acceptable performance. https://tjee.tabrizu.ac.ir/article_19993.html In more advanced technologies where the supply voltage is smaller, "half-select problem" becomes more pronounced in SRAM cells. To alleviate this problem without imposing extra power consumption and area, in this paper, a single-ended 10-T structures is proposed. The proposed structure suits low-power applications and because of the cross-point structure, the bit interleaving architecture is allowed to achieve soft error immunity. Since no output node floats during read and write operation, the proposed structure is suitable for working in low operating voltage and high precision circuits. Simulations of a 0.6-V 5-MHz 1-kb SRAM array, confirm that in a 65-nm CMOS technology, the energy required to write a “0” in the single-ended scheme is 53 aJ/kb. https://tjee.tabrizu.ac.ir/article_19995.html Employing convex and quadratic optimization problems, the optimal probability density function (PDF) for the frequency increment of a random frequency diverse array (RFDA) is extracted. Using a far-field approximation, the fundament harmonic of the radiation pattern is derived, and its power is equated to the peak sideband ratio (PSBR) for a pre-determined beamlength. That means that the power of undesired harmonics would be negligible within the main beam coverage range. Starting from the convex programming and utilizing a suitable variable transformation, an iterative quadratic programming is developed as an alternative technique. The objective is to make sure that the main beam would cover a specific range by a dot-shaped beam which should be as directive as possible. To perform this task, a PDF is generated as similar as possible to the uniform distribution. The radiation pattern for the generated optimal PDF is compared with those of the uniform, normal, and binomial PDFs. Subsequently, using a weighted random number generator, the radiation pattern of the RFDA is plotted. Furthermore, the generated discrete PDF, being in vector form, is modeled by a continuous PDF using the Kernel distribution function. To verify the proposed method, several simulations are performed which denote the accuracy of the proposed theory. https://tjee.tabrizu.ac.ir/article_20168.html After the 28th week of pregnancy, preterm infants respond to auditory stimuli. Examining the brain responses of infants to auditory stimuli provides new insights into their learning abilities. The electroencephalogram (EEG) responses of 15 healthy full-term infants (7 boys and 8 girls) with an average gestational age of 39.8 weeks (±1.2) and 14 preterm infants (11 boys and 3 girls) with an average gestational age of 30.1 weeks (±1.2) who listened to auditory stimuli (repetitive and alternating syllables) were analyzed. To determine whether preterm and full-term infants can differentiate between syllables (ba or ga), deep learning techniques, including Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) networks, were employed for both the repetitive and alternating protocols. The results indicate that both preterm and full-term infants are capable of distinguishing syllables in both repetitive and alternating auditory stimuli, with the CNN achieving higher accuracy than BiLSTM in both age groups. In the proposed method (combination of BiLSTM and CNN), the EEG responses of preterm and full-term infants to repetitive auditory stimuli reached accuracies of 0.855 and 0.892, respectively, while for alternating auditory stimuli, the accuracies were 0.826 and 0.859, respectively. deep learning methods demonstrated the ability to differentiate syllables without the need for feature extraction, even with a limited dataset, and the complexity of the alternating protocol did not significantly affect the responses compared to the repetitive protocol. https://tjee.tabrizu.ac.ir/article_20176.html The growth of renewable energy resources in distribution networks has been driven by investment constraints in large-scale generation and transmission systems, environmental factors, and supportive incentives. This expansion, alongside market deregulation, has fostered the development of local electricity markets. However, these emerging markets may enable the exercise of market power in transitional networks with few participants. This study examines equilibrium prices and their effects on social welfare and consumer surplus in settings where strategic entities compete in local distribution markets, conceptualized through Stackelberg game theory. This paper presents a new convex optimization methodology for determining Stackelberg equilibria in local electricity markets, employing variational inequalities to characterize the equilibrium conditions. By leveraging the principle of symmetry, uniqueness properties of the equilibrium solutions are established. According to the results, Cournot and Stackelberg competition between the strategic renewable energy resource and the distribution system operator, compared to a perfect competition market, leads to a 17% and 15.5% reduction in social welfare, respectively. Furthermore, the energy supply and demand at the local market level will experience reductions of 40% and 38%, respectively. Therefore, small-scale renewable resources have the opportunity to manipulate prices and gain unconventional profits by exercising market power within a limited, local, and transitional energy market, leading to a reduction in social welfare compared to a broad, perfectly competitive market. https://tjee.tabrizu.ac.ir/article_20177.html Wireless communication utilizing UAVs is recognized as a promising technology for future communication systems. In this paper, we investigate the energy-efficient communication between a UAV and a ground terminal by optimizing the UAV’s flight path, under the assumption that the UAV flies horizontally at a fixed altitude. This design approach jointly considers communication power and UAV energy consumption.To achieve this objective, we first introduce a theoretical model for the propulsion energy consumption of fixed-wing UAVs, which is a function of the UAV’s speed, direction, and flight acceleration. Building upon this derived model, and by disregarding signal processing energy consumption and radiated energy, we define the energy efficiency of UAV communication. This is quantified as the total number of transmitted information bits normalized by the propulsion energy consumed by the UAV over a defined time period. For an efficient design aimed at maximizing the energy efficiency of a UAV with general path constraints, including initial and final positions and speeds, as well as minimum or maximum speed and acceleration, we propose a Quadratic Transformation (QT) method. This method has demonstrated an increase in energy efficiency of 0.31 kbits/joule when compared to methods presented in previous research. https://tjee.tabrizu.ac.ir/article_20252.html A novel approach for enhancing short-term price prediction accuracy of emerging cryptocurrency tokens is presented. By addressing the challenges of limited historical data and high volatility through time series categorization, this method categorizes financial time series into distinct subseries based on shared behavioral patterns. For each category, Temporal Fusion Transformers (TFTs) are used to forecast the next step. A data augmentation technique is proposed to combat limited data, particularly when increasing the number of categories. This technique leverages time series data from multiple cryptocurrencies to enrich the training data, ensuring robust model training and improved predictive power. The methodology is tested on two emerging tokens, Notcoin and Dogs. Results demonstrate that the integration of time series categorization, TFT models, and data augmentation significantly improves short-term price forecasting accuracy. In simulated spot trading, the proposed method achieved a 2.45% higher return compared to the baseline TFT approach and a significant improvement compared to the baseline LSTM approach based on initial investment, without using leverage or futures contracts. These findings have valuable implications for traders and investors seeking to make informed decisions in the emerging token market. https://tjee.tabrizu.ac.ir/article_20398.html In recent years, medical imaging has made significant advancements with the use of deep learning techniques. Multiple sclerosis (MS) is a chronic disease caused by the demyelination of the central nervous system. This disease is characterized by plaques visible in MRI scans. Accurate detection of these plaques is crucial for prognosis and treatment monitoring. Manual segmentation of MS plaques by experts is a time-consuming process and prone to human error. This study presents a transformer-based approach for MS plaque segmentation. The proposed model consists of a swin transformer-based encoder for feature extraction and a specialized decoder for segmentation map reconstruction. The output is a labeled mask of the segmented plaques. The model was evaluated on ISBI2015 dataset, and its performance was compared with the classical U-Net model. The results show that the proposed model improved IoU and Dice scores to 0.71 and 0.83, respectively, outperforming U-Net model, which achieves IoU and Dice scores of 0.64 and 0.78. https://tjee.tabrizu.ac.ir/article_20399.html This paper proposes an analog neuron with the capability of operating under low-voltage supplies. The proposed neuron uses a current-mode low-pass filter with a variable gain to linearly integrate the input spikes onto the membrane capacitance. The spike generator circuit includes a positive feedback loop that not only increases the switching speed but also causes low energy consumption. The use of positive feedback eases the implementation of the frequency adaptation mechanism achieving a realistic time-constant and low power dissipation. An inverter-based voltage comparator is also employed to provide the threshold voltage of the spike generator circuit and re-adjust the membrane potential. Mathematical analyses in the subthreshold region show a first-order linear equation that confirms the structure’s simplicity and linearity of the proposed neuron. The simulation results in the 180-nm CMOS process under a 0.3-V supply voltage report an energy consumption of 176 fJ/spike. In addition, by adjusting the control parameters, the neuron shows different firing patterns such as slow and fast firing, chattering behavior, and frequency adaption mechanism. https://tjee.tabrizu.ac.ir/article_20400.html With the establishment of a competitive environment in the power industry, along with the smartening of power networks and the advancement of artificial intelligence algorithms, electricity price forecasting over different time horizons has become one of the most critical issues in power system planning and operation. The restructuring of the power industry and the integration of renewable energy sources and energy storage systems have fundamentally transformed electricity pricing mechanisms. In electricity markets, several factors—including energy supply and demand, production, transmission, and distribution costs, as well as governmental policies—play a significant role in determining electricity prices. With the increasing penetration of renewable energy and storage technologies, electricity prices are also increasingly influenced by these elements. In this study, short-term (24-hour) electricity price forecasting is performed using real market data and the Group Method of Data Handling (GMDH) neural network in the presence of renewable energy sources. Furthermore, the k-means clustering technique is employed to evaluate the impact of storage systems on market prices. The results obtained from this research demonstrate the high accuracy of the proposed model in short-term electricity price forecasting. https://tjee.tabrizu.ac.ir/article_20401.html In this paper, a two stage body driven double recycling folded cascode operational transconductance amplifier with enhanced slew rate and low power consumption is presented. The proposed OTA uses the body driven technique to reduce the power consumption. By combining the double recycling folded casode and body driven structures, this OTA can provide a good performance in the slew rate, CMRR, PSRR, gain, and power parameters. The first stage of the proposed OTA uses four paths to conduct a small signal current, which leads to an obvios increase in transconductance. C_C capacitors are used for frequency compensation. To demonstrate the capabilities and performance of the proposed OTA, post layout simulations with 180nm CMOS technology are performed. The proposed OTA provides a DC gain of 58.5 dB and a phase margin of 71.6 ° at a unity gain frequency of 1.9 KHz for a 200 pF capacitive load. The proposed OTA consumes very low power of 228.7 nW under a 0.9 V supply voltage. Also, the proposed OTA provides higher CMRR and PSRR of 104.5 dB and 78.7 dB, respectively. The proposed OTA is robust to PVT variations. Finally, the FOM performances are improved compared to the previous works. https://tjee.tabrizu.ac.ir/article_20402.html Since the content of dreams can mirror an individual's mental state, recognizing emotions during sleep can reveal valuable information about the state of mind. This research suggests a new method for the automated classification of emotions during the state of sleep known as the rapid eye movement (REM) state. In the suggested method, in the preprocessing step, electroencephalogram (EEG) signals are filtered with a Butterworth bandpass filter and are then broken into 20-second windows. Each filtered EEG window is then decomposed into five intrinsic mode functions (IMFs) with the empirical mode decomposition algorithm. In the next step, the sample entropy feature is extracted from each IMF. The ReliefF algorithm is utilized to select the best subset of features, which are applied as an input to random forest, support vector machine, and K-nearest neighbor machine learning algorithms. Emotional states are classified in a three-class setting (positive, negative, and neutral) as well as in two binary classification scenarios: (1) positive versus negative, and (2) neutral versus a combination of positive and negative emotions. EEG signals from the publicly available dataset known as the DEED dataset are utilized to see brain activity patterns which are present in different states of emotions in dreams and to analyze the performance of the suggested method. The outcome reveals that the suggested method provides an accuracy of 93.74% in three-class classification and an accuracy of 97.45% in binary classification (emotions versus neutral), which indicates good performance in the classification of different states of dreams' emotions during sleep. https://tjee.tabrizu.ac.ir/article_20403.html Superconducting cables have been considered due to their advantages in overcoming the limitations of conventional cables. However, the transient behavior of high-temperature superconducting cables under fault conditions in the presence of distributed generation sources poses challenges for design and protection. In this paper, the evaluation of the cable operating mode in steady and transient states under the influence of changing fault type, fault resistance, network impedance, and distributed generation location has been investigated as the main objective. The electrothermal performance of a three-phase coaxial superconducting cable, with a nominal voltage of 22.9 kV and an operating current of 1 kA at an operating temperature of 70 K, was evaluated in a distribution network including distributed wind turbine generation. The cable response to fault scenarios was investigated by varying parameters such as fault type and resistance, network impedance, and distributed generation location. The input current to the cable, resistance, and temperature of the superconducting layer were analyzed as performance components affected by the aforementioned parameters. Based on the input current, the operating mode limits the performance of the cable due to thermal and electromagnetic conditions. The results showed that changing the fault resistance from 0.01 to 10 Ω reduces the input current by 84%. As a result of these conditions, the operating mode of the cable changes to the flux-flow mode. In addition, increasing the line length reduces the input current and pushes the cable towards the quemch mode. https://tjee.tabrizu.ac.ir/article_20728.html In this paper, a design procedure is presented for design of tunable coding metasurface for beam controlling applications in terahertz frequencies. The designed unit cell is a graphene-based structure consisting of individual ribbons that can be independently biased. As a result, a two-bit structure with phase difference between adjacent states is obtained, which is used in the implementation of two controllable metasurfaces with beam steering and shaping abilities. The simple shape of the proposed unit cell is an important benefit for practical implementation of the metasurface. Moreover, the equivalent circuit model of the unit cell is presented to better understand the performance of graphene ribbons. In the reflection response of the cell and in the frequencies between 1.3 to 1.55 THz, the phase difference of with an error less than is kept between adjacent states while the normalized amplitude is between 0.74 to 0.87 in the desired frequency band. A theoretical method based on the generalized Snell's law has been used to verify the validity of the full-wave simulation results. The refraction angle of the beam in the first metasurface may be controlled by changing the phase difference along x and y directions of the metasurface. This subject requires the use of lattices with different dimensions. The reflected angle of beam varies in the range of 9 to 25 degrees with changing the size of lattice. In the second metasurface the cells are arranged in such a way that a vortex beam with l=+1 is obtained. https://tjee.tabrizu.ac.ir/article_20729.html Photovoltaic (PV) systems are increasingly integrated into microgrids, offering clean and sustainable energy solutions. However, partial shading conditions significantly impact PV system performance, leading to multiple local maxima in the power-voltage (P-V) curve. This necessitates robust Maximum Power Point Tracking (MPPT) algorithms to efficiently extract maximum power. This paper proposes a novel MPPT technique for PV systems connected to a microgrid under partial shading conditions, employing the Gravitational Search Algorithm (GSA). GSA, inspired by Newtonian gravity, effectively explores the solution space and converges towards the global maximum power point. The proposed method is evaluated through simulations, demonstrating superior performance compared to conventional MPPT techniques in terms of tracking accuracy, convergence speed, and robustness against varying shading patterns. The graph of the tracked power for the proposed method in this paper, under variable temperature and partial shading conditions, shows that it has a faster response (100ms) compared to similar methods, such as the Fuzzy Logic Controller (FLC) and the Gravitational Search Algorithm (GSA), for tracking the global maximum power point in grid-connected PV systems under partial shading conditions. The results highlight the effectiveness of the GSA-based MPPT in maximizing energy extraction from PV systems under challenging operating conditions, thereby enhancing the overall efficiency and reliability of microgrids. https://tjee.tabrizu.ac.ir/article_20730.html This paper presents a novel 3D spherical dome geometry to improve the accuracy of simultaneous direction-finding for the azimuth and the elevation angles for specific applications. The estimation of the arrival angle is implemented using the Multiple-Signal Classification (MUSIC) algorithm with considering the coupling effect between the elements. To evaluate the performance, the accuracy of direction-finding using the proposed geometry is compared with more common geometries such as cylindrical arrays and triangular prism with the same volume. The angular estimation accuracy is calculated based on the root mean square error using the Monte-Carlo simulations, and the simulation results indicate that the proposed geometry reduces the estimation error by 30 percent and 22 percenent compared with cylindrical and prism geometries respectively, and the spherical dome geometry has acceptable estimation accuracy compared to the Cramer-Rao lower bound criterion for this geometry. The simultaneous direction-finding accuracy for both azimuth and elevation angles in the considered scenario, using the 3D spherical dome geometry, has been calculated as 0.84 degrees (for 100 number of snapshots and signal-to-noise ratio of 7 dB, taking into account the coupling effects between the elements).The use of this geometry for specific applications (limited azimuth and elevation angles) provides a significant improvement in angle of arrival estimation error, especially by reducing the coupling effects between the ‌elements. The accurate estimation of the azimuth and the elevation angles of the signal and the interferences makes this geometry suitable for beamforming applications in new generations of mobile communications. https://tjee.tabrizu.ac.ir/article_20865.html The growing demand for electrical energy, coupled with the limited availability of resources, presents significant challenges in ensuring a reliable supply of electricity to end-users. Addressing these challenges necessitates the implementation of optimal power grid management strategies. Such management requires comprehensive oversight of information across both the generation and consumption sectors, relying heavily on the collection, analysis, and processing of large volumes of data. Consequently, the extensive and real-time exchange of data has become a fundamental characteristic of modern power networks. This paper investigates the critical role of information within the various sectors of contemporary power grids and analyzes the associated challenges pertaining to data security. Following an overview of the structural organization and functional significance of data exchange, the study examines potential security threats and the implications of cyberattacks on different components of the grid. Furthermore, it evaluates existing and emerging strategies aimed at mitigating these security risks. https://tjee.tabrizu.ac.ir/article_20866.html This work presents a novel and high-precision technique for capacitance measurement based on converting the capacitance charge to current. For capacitance detection, unlike the conventional methods that utilized transistors to control the charging and discharging process of capacitance, a dynamic supply voltage is employed. Eliminating the controlling transistors removes their parasitic capacitance effect, thereby improving the precision of the detection process and enabling operation at higher frequencies. Meanwhile, the consumed area will be diminished and making it a good candidate for the implementation of an array of sensors. Also, the current mirror stage incorporates a comparator to diminish channel-length modulation’s effects that provide high-accuracy mirroring. Post-layout simulation results using CADENCE for a 180 nm CMOS process are achieved. The designed structure shows a wide input dynamic range of 150 fF with zero power dissipation. https://tjee.tabrizu.ac.ir/article_20989.html The security of a tax system is directly linked to the development of a country. Traditional processes of tax calculation and collection have vulnerabilities that allow criminals to breach centralized financial systems, commit identity theft, and misuse personal information. These flaws also lead to tax evasion and fraud by enabling unauthorized alterations to financial records and documents, often in collusion with dishonest employees. Blockchain technology, as a distributed database, offers decentralization, transparency, immutability, and trustless operations, paving the way for advanced levels of transparency, security, and reporting. It significantly impacts how taxes are recorded and helps combat fraud and tax-related violations. This capability is achieved through smart contracts, which can automate processes such as payments, asset transfers, and data recording, thereby making tax calculations more efficient and reducing time and costs. In this study, for the first time, a locally designed and developed tax management system based on blockchain technology has been implemented to enable invoice management. This system minimizes tampering with audit records and prevents the issuance of fake invoices and fraudulent billing, leading to a substantial reduction in tax evasion. The proposed system digitizes invoices and automatically calculates taxes using Ethereum blockchain smart contracts. As a result, it reduces the risk of cyberattacks that may lead to data loss or forgery and enhances trust in tax payment execution by eliminating the influence of third parties. https://tjee.tabrizu.ac.ir/article_20990.html Extracting sentiment from textual data is crucial for understanding public opinion and guiding strategic decisions. We introduce a hybrid deep-learning pipeline that combines convolutional feature detectors, bidirectional recurrent units, and a custom attention mechanism. First, convolutional layers with pooling condense local n-gram patterns into compact feature maps. These maps are fed into a bidirectional LSTM network that captures sequence information in both forward and reverse directions. A specialized attention module then assigns relevance scores to individual tokens, sharpening the final classification. Evaluations on widely used sentiment benchmarks show that our method outperforms leading models in terms of accuracy as well as requires fewer computational resources, making it a practical solution for scalable emotion detection in text. https://tjee.tabrizu.ac.ir/article_21131.html In this paper, using an enhanced homogenization method, the behavior of a one-dimensional guided-mode resonant structure with zero-contrast configuration is investigated. The proposed method demonstrates significantly enhanced accuracy compared to conventional Effective Medium Theory (EMT)-based approaches. The modal behavior of the sample structure is analyzed using this method, revealing excellent agreement with rigorous numerical calculations performed via the Rigorous Coupled Wave Analysis (RCWA) method. The proposed enhanced homogenization method serves as an excellent candidate for rapid preliminary design of periodic optical devices and provides an efficient starting point for optimization algorithms. The technique bridges the gap between simplified EMT and time-consuming full-wave simulations, offering a practical tool for resonant photonic structure design. https://tjee.tabrizu.ac.ir/article_21132.html AbstractA novel power divider network for feeding arrays of arbitrary groundless spoof surface plasmon polaritons (SSPP) antennas with coplanar waveguide (CPW) input is proposed. The design approach is generalized for any SSPP antenna with CPW input by presenting a flowchart. The proposed feeding network is simulated and optimized via CST Studio software. We validate our designed network's effectiveness using a fabricated prototype eight-branch SSPP array antenna that shows a measured impedance bandwidth of 0.98 GHz (11.536-12.516 GHz) and a maximum gain increase of 7.48 dB compared with a single-branch SSPP antenna, with only a 1.55 dB deviation from the lossless theoretical increase value of 9.03 dB. The proposed power divider network can feed SSPP antennas with no attached ground, enabling performance optimization of the antenna by placing a detached ground. SSPP-based planar antennas are gaining popularity in aerospace due to their lightweight, low-profile design. They offer high efficiency, wideband operation, and resistance to electromagnetic interference, making them ideal for satellite communications and dynamic aerospace environments. Since single-branch SSPP antennas which use CPW inputs often lack sufficient gain, the power divider network proposed in this paper holds particular importance for combining multiple radiating elements, boosting gain and directivity without compromising compactness https://tjee.tabrizu.ac.ir/article_21198.html Long-code direct sequence spread spectrum (Long-DSSS) systems play an important role in advanced communications and next-generation cellular networks due to their high security and resistance to interference. However, traditional detection methods based on fixed thresholding face significant performance degradation under variable channel conditions, severe noise, and low signal-to-noise ratio (SNR). In this research, a novel approach based on the combination of the double folding (DF) technique and adaptive CFAR (Constant False Alarm Rate) thresholding is presented. DF accelerates synchronization by increasing the coherent integration time, while CFAR eliminates the limitations of fixed thresholding by dynamically and in real time adjusting the threshold according to the statistical characteristics of the signal and noise. To evaluate performance, various CFAR algorithms such as CA-CFAR and OS-CFAR were tested in homogeneous and heterogeneous environments in the presence of pulsed and random jammers. Simulation results show that the proposed method, in addition to reducing the Bit Error Rate (BER), provides more stable performance compared to conventional methods and can lead to improved reliability and security of communications in next-generation cellular networks. https://tjee.tabrizu.ac.ir/article_21199.html Brain-Computer Interfaces (BCIs) are transformative in neurorehabilitation, offering vital communication and control for individuals with severe motor impairments, such as those with ALS, spinal cord injuries, or stroke. By establishing direct links between brain activity and external devices, BCIs bypass damaged neural pathways, thereby restoring a degree of motor function and improving quality of life. Electroencephalography (EEG) is a leading modality for BCI development due to its accessibility and cost-effectiveness. However, a significant hurdle remains the inherent variability in cognitive and individual differences, which substantially impacts motor imagery (MI) task performance and BCI accuracy. This research introduces a novel approach for enhanced MI classification by specifically integrating the Common Spatial-Spectral Pattern (CSSP) filters with the Tunable-Q Wavelet Transform (TQWT). This synergistic combination, applied to the extensive CHO-2017 database (52 participants) which captures significant inter-individual cognitive variations, is designed to effectively address the challenges posed by individual differences in distinguishing between left and right-hand MI tasks. Critically, our method utilizes only the top 10 discriminative features extracted through this hybrid technique, significantly streamlining the process while maximizing classification efficacy. This tailored feature set demonstrated remarkable effectiveness across 99% of participants. When integrated with a K-Nearest Neighbors (KNN) classifier, this approach achieved an outstanding accuracy of 98.84%, surpassing current state-of-the-art methods. The findings of this research could pave the way for the development of more accurate BCI systems capable of extracting optimal commands for MI tasks. https://tjee.tabrizu.ac.ir/article_21247.html In this paper, we present a novel FPGA-based implementation of a multiplier-free reverse converter for the residue number system (RNS) using a new 4-moduli set {3^n, 3^n-2, 3^n+2, 3^n-1}. The proposed method leverages the unbalanced ternary logic system to achieve efficient residue-to-binary conversion with reduced computational complexity and hardware resource usage. The core architecture eliminates the need for hardware multipliers by exploiting mathematical simplifications and residue properties inherent to the chosen moduli set. Ternary digits are encoded in binary to enable implementation on conventional FPGA platforms. The design is suitable for high-speed applications. Since no comparable ternary RNS reverse converter exists in the current literature, a baseline method was implemented for evaluation. Simulation and synthesis results confirm the efficiency of the proposed design in terms of area, speed, and power consumption, making it a promising solution for low-power, high-performance signal processing and cryptographic systems. https://tjee.tabrizu.ac.ir/article_21248.html Data regarding faults that occur on transmission lines, which is recorded in distance relays, contain vital information that is essential for the settings of relays. The value of fault resistance is one of the critical information that is be effective in setting the operation zones of the distance relay. A variety of effective schemes have been suggested for determining the fault resistance value, and in many of these techniques, the initial step involves calculating the precise location of the fault. As a result, the accuracy of fault location estimation is needed for the calculation of the fault resistance. To address this issue, a off-line algorithm is proposed in this paper to estimate fault resistance, which utilizes the data recorded in distance relays at both ends of the transmission line. The proposed algorithm employs straightforward calculations that do not necessitate unavailable data, such as the equivalent Thevenin impedance at both ends of the transmission line and the calculations for fault location. The performance of the proposed method has been examined in a 9-bus network and T-connected lines, utilizing MATLAB/Simulink and EMTP software. Moreover, to further assess, actual data recorded in distance relays of a 150 kV transmission line has been utilized. https://tjee.tabrizu.ac.ir/article_21391.html This paper presents a dynamic load distribution scheme including a static synchronous compensator (STATCOM) with a proposed nonlinear controller, applied to a studied radial distribution network. Using this model, all the nonlinear equations governing the distribution network have been extracted, and the proposed dynamic power distribution, based on the phasor method, is used to investigate the network dynamics. In the simulation, it is assumed that each motor starts when the other has reached steady state. To provide the reactive power required by the motors, a static synchronous compensator is used, which is controlled by a combination of two controllers: a Passivity-Based Controller (PBC) and a lag phase compensator (LPC). So it could maintain the bus voltage at the reference value and prevent voltage drop, particularly during motor starting intervals. The simulation results, carried out in MATLAB, show that the proposed static synchronous compensator controller effectively improves the dynamic stability of the studied radial distribution network. https://tjee.tabrizu.ac.ir/article_21392.html Heart rate variability (HRV), obtained from RR intervals in ECG signals, reflects the activity of the autonomic nervous system, which is challenging to forecast due to its nonlinear and chaotic nature. In this paper, labeled ECG signals were first obtained from the Physionet database. Then, fourth-order Butterworth bandpass filter and phase space reconstruction with a sliding window approach were applied to these chaotic signals and trained with four machine learning methods. Since LSTM and CNN showed better performance than other methods due to their temporal and frequency capabilities, respectively, a combined LSTM+CNN method was proposed, which improved the RMSE and R² evaluation criteria and was shown to have the following capabilities: robustness against white and pink noise, the ability to distinguish unlabeled arrhythmia patient signals from healthy signals, and the ability to forecast both healthy and unhealthy HRV signals. https://tjee.tabrizu.ac.ir/article_21393.html Stress is one of the common psychological challenges of the 21st century that affects the function of the autonomic nervous system. In this study, the feasibility of non-invasive monitoring of muscle sympathetic nerve activity (MSNA) and skin sympathetic nerve activity (SSNA) was investigated using pre-extracted nonlinear features from multi-wavelength photoplethysmography (PPG) signals and by employing machine learning algorithms. PPG data from 32 healthy individuals (19 to 38 years old) at four wavelengths (red, infrared, blue, and green) were analyzed during three time phases (pre-stress, during stress, and post-stress induced by handgrip and cold pressor tests). Nonlinear features, including Higuchi, Katz, and Petrosian fractal dimensions, approximate entropy, and sample entropy, were used for classification. These features were applied in three classifiers—Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Recurrent Neural Network (RNN)—to classify stress states. The highest performance was observed using approximate entropy at the blue wavelength when distinguishing the pre-stress phase from the first two minutes of stress in the handgrip test (SVM: accuracy = 90.96%, AUC = 1.00). Independent t-tests and Wilcoxon tests revealed significant differences (p < 0.05) in the blue, green, and infrared wavelengths. These results confirm the role of nonlinear features and optimal wavelength selection in effective stress monitoring and highlight the potential application of PPG as a low-cost, non-invasive tool. https://tjee.tabrizu.ac.ir/article_21394.html Six-phase PMSMs are widely used in high-power applications due to their high torque density and reliability. This paper focuses on controlling a specific type of asymmetrical six-phase PMSM, where each stator phase consists of two co-directional windings. The motor's 12 windings are configured as four three-phase sets with star connections. Each three-phase set is powered by a six-switch three-phase inverter, and each inverter is independently controlled by a local microcontroller, resulting in a modular power and control structure for each set. Due to the absence of data exchange between local microcontrollers and the non-sinusoidal back-EMF of the motor phases, dynamic modeling in a two-axis reference frame is not feasible. Therefore, the motor is modeled and controlled in a stationary six-axis reference frame. The drive's behavior under various open-circuit fault conditions in one or more windings is analyzed, and a fault-tolerant control algorithm is proposed for each fault scenario to mitigate torque ripple. Theoretical analyses are validated through software simulations in Simulink. https://tjee.tabrizu.ac.ir/article_21395.html In this paper, a wideband low-noise amplifier (LNA) with a combined common-gate (CG) and common-source (CS) architecture is presented. The design featuring low power consumption and inductorless operation. Due to the large area and design complexity associated with inductors, they are eliminated and replaced with alternative techniques. Using gain-boosting approach, the proposed inductorless LNA employs a CG amplifier in a feedback structure as the input stage in order to enhance the input impedance matching condition, hence lowering the power consumption of the whole circuit. In adition, a current reuse approach is employed in the input stage to furtuer lower the power consumption. Besides, a new noise-canceling structure is used to improve overall noise performance. Also the design features variable gain functionality. These strategies result in promising performance metrics. The achieved results include a minimum noise figure (NF) of 2.84 dB, input return loss (S11) below –10 dB, and a maximum gain of 19.15 dB across the frequency range of 200 MHz to 1 GHz. The amplifier operates with a 1.8 V supply, consumes only 8 mW of power and occupies only 0.0202 mm2. The proposed LNA is implemented in a 0.18 µm TSMC CMOS technology and simulated using Cadence Virtusuo. https://tjee.tabrizu.ac.ir/article_21396.html Functional magnetic resonance imaging (fMRI) is a crucial tool for investigating brain activity. However, analyzing fMRI data presents significant challenges due to the complex temporal nature of the signals and uncertainties in the algorithms used. Classical Independent Component Analysis (ICA) algorithms, such as FastICA, often struggle with high false positive rates and unstable results because they rely on the strict assumption of complete statistical independence. This study aims to comprehensively compare three ICA-based algorithms: FastICA, Entropy Bound Minimization (ERBM), and Semi-Blind Spatial ICA (SBS-ICA). The objective is to assess how different statistical assumptions and prior information affect the quality of component separation in fMRI data and the accurate identification of brain activation regions. Evaluations were conducted using component numbers of 40, 50, 60, and 70. The results revealed that the SBS-ICA algorithm, which benefits from spatial prior information, demonstrated the best performance with an area under the ROC curve (AUC) of 0/999, a high correlation of 0/89, and the lowest number of spatial false positives. The ERBM algorithm, which models temporal correlations, outperformed FastICA, showing a lower mean squared error (MSE = 0/079) and more stable correlation values. In contrast, FastICA exhibited the weakest performance among the three algorithms. These findings highlight the advantages of ICA-based guided methods and emphasize the significance of incorporating task modeling for accurate analysis of fMRI data. https://tjee.tabrizu.ac.ir/article_21397.html Clustering analysis of functional Magnetic Resonance Imaging (fMRI) time series is a widely used method for non-invasively identifying and mapping active brain regions in response to specific tasks. However, fMRI clustering faces significant challenges due to the imbalance between active and inactive voxels, often leading to a high number of false positives and the creation of noisy, scattered activation maps. This study introduces a novel two-stage method designed to control the false positive rate and enhance the accuracy and quality of detecting active brain regions. In the first stage, the method establishes a statistical threshold for signal discrimination using randomization-based inference. The second stage removes noisy and isolated voxels by applying a spatial threshold based on three-dimensional neighborhood criteria. This control strategy was tested with Neural Gas (NG) and Growing Neural Gas (GNG) clustering algorithms during an experimental analysis of auditory fMRI data. The performance of the methods was evaluated using the Dice coefficient, spatial coherence and Calinski Harabasz Index metrics. The results showed that the proposed control method significantly enhanced performance, increasing the Dice coefficient for the GNG algorithm from 0/30 to 0/65 and improving compactness by reducing the spatial coherence value from 25/62 to 14/63. The Calinski Harabasz index also increased from 2/703 to 1/1192. These results indicate that combining the GNG algorithm with this control method leads to greater stability and reliability in identifying active neural regions, providing a more effective approach for analyzing fMRI data. https://tjee.tabrizu.ac.ir/article_21398.html The rapid spread of fake news on digital platforms has become a major societal challenge, influencing public opinion, political stability, and social trust. Despite extensive research, existing detection models often suffer from limited generalization, weak robustness to noisy or short texts, and reliance on manually engineered features. Moreover, the evolving linguistic and contextual nature of fake news reduces the long-term effectiveness of conventional approaches. To address these limitations, this study proposes a novel hybrid fake news detection framework that, for the first time, integrates Kernel Fuzzy Rough Set (KFRS)–based feature selection with a two-stage ensemble of deep learning and traditional machine learning classifiers. Unlike prior studies that rely solely on either neural or statistical models, the proposed approach combines LSTM and Bi-LSTM networks with Logistic Regression, Support Vector Machine, and XGBoost, leveraging their complementary strengths through soft voting and stacking. The KFRS component plays a critical role in refining feature representations by handling uncertainty and reducing noise, which is particularly beneficial for short and noisy textual data. Skip-Gram word embeddings are employed to capture semantic relationships between words and enhance contextual understanding. The proposed framework is evaluated on three benchmark datasets—LIAR, FakeNewsNet, and FakeEdited—demonstrating consistently strong performance across diverse domains. Ablation experiments further confirm that incorporating KFRS leads to substantial improvements in Recall and F1-score. Overall, the proposed method offers a robust, scalable, and practically deployable solution for real-world fake news detection. https://tjee.tabrizu.ac.ir/article_21399.html This article presents the design and simulation of a 45-degree polarizer on a broadband monopulse printed circuit antenna in the KU frequency band. Considering the need for a low-profile and lightweight planar structure, an optimized polarizer element was selected. To create 45-degree polarization on the broadband monopulse antenna, parallel lines were implemented on the parallel plates with an angular offset of 11.25 degrees between them. First, the existing monopulse antenna is introduced with its simulation and construction, and then the effects of the polarizer plates on the monopulse simulation characteristics are examined and analyzed. With the polarizer placed on the antenna, no fundamental changes were observed in the monopulse characteristics. The pattern shape of the monopulse array antenna with the polarizer was maintained in the broadband range. The simulation of the final monopulse array structure with 45-degree polarization was carried out using CST software, achieving a 30% bandwidth with an average gain of 23 decibels. https://tjee.tabrizu.ac.ir/article_21400.html In DC-AC voltage conversion applications where an output voltage higher than the input voltage is required, utilization of a voltage source inverter encounters limitations. This structure usually requires an extra boosting stage, which increases the complexity of the system. Therefore, single-stage converters with boosting capability have been proposed, one of the prominent types of which is the split source inverter. In this paper, a new structure of this type of inverter based on an impedance network is introduced with the purpose of increasing the output voltage gain. The proposed structure has a quadratic boost factor and is controlled by two parameters, duty cycle and turns ratio. Considering the specific switching requirements of the inverter, the modified space vector modulation method is used, and the method of generating pulses and the relevant equations are explained. Also, the principles of inverter operation in different operating modes and its mathematical analysis in steady state conditions are presented. To prove the system's proper operation and verify theoretical analysis, simulations are performed, and an experimental prototype of the system is built. The results indicate the proper performance of the converter and well compliance with theory. https://tjee.tabrizu.ac.ir/article_21401.html In this paper, a robust adaptive fuzzy backstepping controller is designed for multi-input multi-output (MIMO) stochastic switched systems with time delays, which encompass nonlinear uncertainties. The nonlinear backstepping control technique is employed as the main framework for the controller design. A state observer is also designed to estimate the unmeasurable states. To simplify the design process, eliminate the need for a supervisory controller, prevent parameter drift in the adaptive laws, and ensure robust performance in the presence of unstructured uncertainties (such as external disturbances and modeling errors), the adaptive laws are fortified using the sigma modification method. The boundedness of all closed-loop signals and the convergence of both the tracking error and observer error to a small neighborhood of the origin are mathematically proven based on Lyapunov-Krasovskii stability analysis and expressed in the form of Theorem 1. Finally, simulation results demonstrate the desirable performance and effectiveness of the proposed robust adaptive fuzzy controller in stabilizing the aforementioned systems. https://tjee.tabrizu.ac.ir/article_21402.html Timely detection of acute myocardial infarction (AMI) is essential for reducing mortality rates and improving treatment outcomes. However, traditional wearable devices face limitations in capturing multi-lead ECG signals, especially from chest leads, which makes accurate AMI detection challenging. This study aimed to evaluate the performance of a wrist-based ECG (wECG) device equipped with three electrodes, comparing it to the reference ECG signals (12 leads). The proposed framework effectively distinguishes AMI patients from healthy individuals and those with other cardiovascular diseases (CVD). Initially, raw ECG and wECG signals were preprocessed to extract higher-order statistical features, Hjorth descriptors, and amplitude and phase components derived from the Fast Fourier Transform. Key features and leads were selected using mutual information (MI) and the F-test. These selected features were then used in various machine learning algorithms (SVM, DT, KNN, XGBoost) and deep learning models (CNN, ResNet, DenseNet, LSTM) to develop an accurate and interpretable model for AMI detection. The results indicated that the 12-lead ECG achieved an impressive average accuracy of 100% in distinguishing healthy subjects from patients (both AMI and CVD) when using XGBoost and CNN. However, the accuracy for differentiating AMI from CVD decreased to 99.4% (using ResNet) and 96.2% (using SVM). For wECG data, the V5-LA lead demonstrated the best performance, achieving an average accuracy of 98.1% in differentiating control subjects from CVD patients with XGBoost. Nevertheless, the accuracy for distinguishing AMI from CVD was limited to 93.2% (using CNN).These findings suggest that wECG has potential for the early detection of AMI. https://tjee.tabrizu.ac.ir/article_21403.html This paper presents an innovative online approach for univariate fault detection and alarm generation using the power of spectral analysis. The new approach utilizes a Spectral Stability Index (SSI) for simultaneous and sensitive monitoring of small and incipient changes in both the mean and variance of a process signal, which are common precursors to industrial alarm systems. The SSI framework accomplishes this by quantifying the deviation of the real-time power spectral density of the process signal from a reference model of normal operation. The method inherently carries an important advantage: it can detect low-level, developing faults much earlier than methods based on time-domain analysis. Its superior effectiveness is demonstrated through numerical and industrial case studies, where it significantly outperforms conventional methods, such as CUSUM, deadbands and delay timers, and more sophisticated advanced methods that have been recently introduced, such as serial and cascaded alarm systems. The results confirm the method's potential for substantially improving the responsiveness and accuracy of industrial alarm systems. https://tjee.tabrizu.ac.ir/article_21536.html The increasing use of Internet of Things (IoT) devices has introduced new security challenges, making timely detection of anomalies and abnormal nodes essential. Limitations in hardware resources, protocol diversity, and network traffic complexity render traditional methods ineffective in identifying threats. This paper presents a hybrid approach based on graph theory and artificial neural networks to classify network nodes into three categories: healthy, suspicious, and unhealthy. Graph features such as centrality measures are extracted, and anomaly detection accuracy is improved to 99.03% using LOWESS curve fitting, compared to 98.22% for a neural network without curve fitting and 98.87% for linear and polynomial models. This method outperforms other approaches and serves as an effective tool for improving security in IoT networks. https://tjee.tabrizu.ac.ir/article_21537.html User stance detection means determining the user's attitude (agree, disagree, or neutral) towards a particular issue. Academic and industry research is highly interested in the automatic analysis of opinions in social networks. One common method for training efficient models is knowledge distillation, in which knowledge is transferred from a large, complex model (the teacher) to a smaller, lighter model (the student). Calibration refers to the degree to which the model’s confidence aligns with its actual accuracy. In previous methods, using a single calibrated teacher in a multi-generational framework causes error propagation due to its dependence on the previous generation, resulting in unstable training. To address this challenge, ensemble calibrated knowledge distillation is proposed. In the proposed framework, a dynamic ensemble of the best calibrated models from all previous generations is used, which makes the training process more robust and diverse. In addition, a self-paced calibration annealing strategy is introduced, which, by applying a simple calibration objective in the early stages, helps in stable feature learning and focuses on more precise error optimization in the later stages. The results evaluated based on F-micro and F-macro metrics show an improvement of at least 3% compared to the baseline calibrated model on different real-world datasets (COVID-19, P-Stance, and AM). https://tjee.tabrizu.ac.ir/article_21538.html Cyber-dependent control capability of the power grid has been recently provided, introducing a wide range of the new cyber-physical risks. This paper is devoted to the vulnerability assessment of this interdependent system under different types of coordinated cyber (denial of service to disrupt control centers) and physical (intentional failure of physical equipment) attacks. Herein, an aggregated vulnerability index (VI) is proposed, taking into account both deterministic (direct consequences) and probable (due to the degradation) effects of the attacks. The suggested index consists of three key factors of inherent, corrective (based on a different variety of the operation corrective actions) and expected load shedding. Until now, this expected VI has not been yet investigated well in the literature. This work is novely focused on this probable impact due to the degradation of the cyber network. This study aim to handle the problem complexity by providing a simplified fast yet sufficiently accurate framework. The method’s correctness and effectiveness are demonstrated based on the simulation results of the IEEE 39-bus test systems, with comparisons to cascading failure process. The numerical results also highlight the computational efficiency of the proposed method rooted in its straightforward calculations. https://tjee.tabrizu.ac.ir/article_21539.html Intrusion detection in the Internet of Medical Things (IoMT) is a safety critical task, as healthcare infrastructures depend on highly sensitive data and real time medical devices. While Long Short Term Memory (LSTM) networks offer strong capability in detecting complex network attacks, their performance is highly influenced by appropriate hyperparameter selection. To address this challenge, this study introduces a Modified Harris Hawks Optimization (MHHO) algorithm that enhances the standard HHO by integrating dynamic inertia weighting and Lévy flight based exploration. These mechanisms improve the balance between exploration and exploitation, effectively reducing premature convergence and strengthening global search behavior.The proposed MHHO is employed to adaptively optimize key hyperparameters of an LSTM based intrusion detection model, eliminating the need for manual tuning or conventional methods such as grid search. Furthermore, a robust, recall focused fitness evaluation strategy is used to obtain hyperparameter configurations that improve model reliability and decrease sensitivity to class imbalance.Experimental results on the CICIoMT2024 dataset across five independent runs show that both HHO LSTM and MHHO LSTM outperform the base LSTM in terms of average F1 score, with the MHHO LSTM model demonstrating significantly lower performance variability. Since the optimization is performed only once during offline training, the final deployed system maintains the standard LSTM inference complexity. These findings highlight that the proposed MHHO LSTM framework delivers enhanced stability, dependable performance, and practical suitability for safety critical IoMT environments. https://tjee.tabrizu.ac.ir/article_21540.html Proportional–Integral–Derivative (PID) controllers have been widely used in industrial systems due to their simple structure and ease of implementation. However, because the parameters of classical PID controllers are fixed, uncertainties, nonlinearities, and variations in system dynamics may degrade their control performance. To address these limitations, fuzzy logic systems (FLSs) have been incorporated into PID controllers to enhance adaptability and robustness without requiring an accurate mathematical model of the controlled plant. This paper presents a comprehensive survey of model-free fuzzy PID and PI control approaches. In particular, two widely used architectures, fuzzy gain scheduling PID (FGPID) and fuzzy PID (FPID) controllers, are reviewed and analyzed. Different implementations of these controllers based on type-1 fuzzy logic systems (T1FLS), interval type-2 fuzzy logic systems (IT2FLS), and general type-2 fuzzy logic systems (GT2FLS) are discussed. Furthermore, various tuning strategies for fuzzy parameters, including analytical methods, gradient-based learning, and evolutionary optimization algorithms, are examined. The challenges associated with parameter tuning and closed-loop stability analysis are also highlighted. The survey provides a structured overview of recent developments and identifies open research directions for improving the design, stability, and practical implementation of fuzzy PID controllers in complex nonlinear systems. https://tjee.tabrizu.ac.ir/article_21652.html This paper presents a hybrid intrusion detection system for Industrial Control Systems that leverages real-world data to detect known cyberattacks with very high accuracy. The research begins by designing a realistic laboratory testbed comprising a Siemens S7-1200 PLC, an HMI, and WinCC software. Three types of cyberattacks DDoS, Start/Stop, and Port Scan are simulated using Metasploit, and the corresponding network traffic is captured via Wireshark with Port Mirroring enabled. During preprocessing, the data are filtered, labeled, and behavioral features—such as packet inter-arrival time, control command frequency, and destination port diversity are extracted. A hybrid LSTM-DT model is developed and trained in three configurations; the parallel configuration (Config 3), which fuses the outputs of both LSTM and Decision Tree components, demonstrates superior performance. Evaluation using 10-fold cross-validation on labeled testbed data shows that the proposed model achieves 99.45% accuracy, 99.23% precision, and 99.49% recall. By focusing on flow-based behavioral patterns rather than payload inspection, the system ensures low computational overhead, making it well-suited for resource-constrained industrial environments. https://tjee.tabrizu.ac.ir/article_21653.html In this paper, a novel approach for shaping the divergent Gaussian beam of a mm-wave source into a flat-top beam using a single curvature is presented. According to the proposed method, a single curved profile is achieved by combining the collimator profile with the beam shaper profile. To validate the approach, a one-sided lens (plano-convex lens) profile is extracted using this method, then located at a distance of 52 millimeters from a 100 GHz conical horn antenna, and the Gaussian radiated beam of the horn with a focal radius of 11.9 millimeters is shaped into a flat-top beam of radius 400 millimeters at a distance of 3 meters. The lens profile is optimized based on a full-wave simulation, and its flat-top beam is compared to the result of a typical double-sided (concave-convex) lens. According to the simulation results, the flat-top beam of the one-sided lens has the desired ripple and roll-off in the field of view. Finally, some studies on the sensitivity of the optimized lens profile to frequency deviation have been performed. The designed lens is suitable for generating a flat-top intensity in the active mm-wave stand-off imaging systems. https://tjee.tabrizu.ac.ir/article_21654.html A millimeter-wave rotary joint based on printed microstrip gap waveguide technology is presented. To this goal, a unit cell structure which its bandstop contains the desired frequency band was first designed. The rotary joint structure is proposed as a rectangular cavity surrounded by the designed periodic mushrooms realized on microstrip board and its top layer is a metallic disk placed with an air gap distance from microstrip board. The rectangular cavity is fed by two coaxial ports connected at its both sides. Simulation results show a narrow band insertion loss for designed rotary joint because the cavity is excited at its first mode. To widen its bandwidth, shorting patch and window are introduced in the cavity. The dimensions of the proposed rotary joint were optimized by time domain solver of CST software to obtain the best insertion loss behavior over the desired frequency band. Simulation results show the return loss is better than 10 dB and the insertion loss is less than 0.5 dB over about 21% relative bandwidth. Comparsion of the simulation results with those obtained by freqency domain solver of CST software and HFSS simulator confirms the accuracy of the design and simulations. https://tjee.tabrizu.ac.ir/article_21655.html Temporal lobe epilepsy (TLE) is increasingly recognized as a network disorder involving widespread disruptions of functional brain connectivity. Resting-state functional MRI (rs-fMRI) captures these dynamics, but conventional methods often assume static connectivity or ignore inter-regional interactions. We propose an Adaptive Masked Spatio-Temporal Graph Convolutional Network (AdaMST-GCN), which learns sparse, data-driven adjacency masks from partial-correlation graphs and integrates them with temporal convolutions to model dynamic network patterns. Evaluated on a TLE rs-fMRI dataset using 5-fold cross validation with sliding windows (50, 100, 150, 200 TRs), AdaMST-GCN achieved a mean held-out test F1 score of 79.0%, outperforming the original ST-GCN (75.7%) and LSTM baseline (58.5%). At 50-TR windows, it peaked at 82.3% F1. The learned masks consistently identified high-centrality regions, including the precuneus, temporal pole, and orbitofrontal cortex, corresponding to known TLE pathology. These results demonstrate that adaptive graph learning improves both predictive accuracy and interpretability, providing clinically relevant biomarkers.