مجله مهندسی برق و الکترونیک ایران
سال نوزدهم شماره 3 (پاییز 1401)

In this paper, a novel waveguide array antenna composed of non-inclined slots in the narrow wall of the waveguide, which are excited with rectangular Iris is introduced. This new method in addition to improving impedance and radiation characteristics has an easy production process. Also, because of using non-inclined slots, the antenna has very low cross polarization level. In this paper, in the first step, suitable radiating elements for designing a linear slot array antenna were considered, and then a linear array suitable surveillance radar with 136 elements with Taylor distribution is designed and simulated with HFSS at the frequency of (2.7 - 3.1) GHz. Finally, the antenna has been fabricated and tested. The results of the test and simulation are in good agreement.

To utilize Ka-band frequency spectrum, has extensive applications in satellite communication and various radars implementation. Also, a portion of the frequency spectra required for implementation of fifth generation mobile telecommunication networks (5G) is laid into this frequency band. On the other hand, many mentioned applications, requires the achievement to radiation beam with precise steering capability in a narrow region of solid angles. This feature is achievable by utilizing the narrow beam-width antennas. Array antennas, besides providing the demanded gain, can fulfill such requirement. Also, the beam-width can be tunable in an acceptable range of different directions. Such a tunability can be realized using the phased array antennas. The capability of change in maximum direction (main lobe) of these antennas is provided using the active phase shifting components as feeding elements of the phased arrays, such as PIN diodes and ferrite devices. However, employing the passive Butler matrix components is considered as simpler and cheaper tool to realize the approach. To use the Butler matrix as feed of the antennas, with more input-output ports, leads to narrower beam-width. In this paper, after a brief review on Butler matrix structures principles, a simple design of 32×32 Butler matrix for Ka-band frequency spectrum is proposed, and the simulation results of its performance are presented. After applying the appropriate waves to two specific input ports and connecting the Butler matrix structure to the microstrip array, the beam-width of 3.5 degrees is obtained.

Massive MIMO system has been introduced as a promising technology in achieving the benefits of the fifth generation of wireless communications, including high spectral and energy efficiencies. Having an accurate channel estimation is the main challenge for the utilization of the benefits of this system. Pilot contamination causes errors in user channel estimation and is a serious obstacle to access the benefits of the Massive MIMO system. In order to eliminate the destructive effect of the pilot contamination and to track the temporal variation of the users’ channels (due to displacement) in the Gauss-Markov channel model, an adaptive smart pilot allocation method in uplink TDD (Time Division Duplex) mode is proposed. Moreover, by using the virtual channel modeling technique, an alternative solution for calculating virtual assignment cost equivalent to the actual allocation cost in each block is presented in an adaptive manner which has low computational complexity. Finally, the performance of the proposed adaptive pilot sequence allocation algorithm is evaluated using computer simulation. Our results demonstrate the good performance and fast convergence of the proposed algorithm under different users' speeds.

In this article, three types of interference ,  and  were applied to the SAR radar signal. Then, the suppression of the interference effect was investigated by various adaptive filters. After implementing different types of adaptive filters with the same conditions to suppression interference, the performance of adaptive filters were compared with each other. According to the simulation results, the output of Kalman filter has a higher quality than other existing filters. Then, a proposed algorithm to improve the interference reduction was presented. The accuracy of suppression the interference in the proposed algorithm was improved by 3dB compared to the Kalman filter. This improvement was achieved in return for increased processing time. The proposed algorithm is 55 times faster than the Kalman filter. so, in case of high precision and robust hardware for complex calculations, the use of the proposed algorithm will improve interference suppression.

In this paper, a method is presented for design and realizing ultra-wideband phase shifter in frequencies beyond 10 GHz with about a hundred percent fractional bandwidth. Its structure is composed of microstrip transmission lines and short circuited stubs. Its combination has advantages, from fabrication and realization viewpoint, over conventional structures that utilize coupled transmission lines and multi-layer structures. With this design and optimization method, it is possible to manufacture phase shifters with arbitrary phase shifts as well as arbitrary input and output impedances. The values of circuit dimensions is obtained using the optimization of least mean square combination of error function. To design phase shifters, a computer program is written which utilizes circuit models of microstrip transmission lines and microwave network relations. Its results are then improved using full-wave electromagnetic softwares to consider parasitic effects of T-junctions. 45 degree phase shifter with 50 ohms input-output impedances is designed, simulated and fabricated. Also, the 90 degree phase shifter was designed and simulated in the case that the input and output impedances be unequal, assuming to be 50 and 75 ohms. By investigation of the results of transmission and reflection coefficients and phase shifts obtained from full-wave simulation and practical realization, it is clear that the design goals are highly satisfied.

Millimeter wave communication is able to support wireless communications at data rates of several gigabits per second due to its abundant spectral resources. However, severe path losses and high directionality have made millimeter-wave systems vulnerable to obstruction. One solution to overcome this phenomenon is to use relays. Since the relays require power consumption and, in addition, some millimeter-wave equipment imposes high costs on systems, Intelligent Reflective Surfaces are being introduced as a new low-cost technology to overcome the above issues. In this paper, we consider an intelligent reflective surface-based millimeter wave system in which a base station, i.e., Alice sends its confidential messages to a user, i.e., Bob in the presence of a passive eavesdropper. With the aim of increasing the ergodic secrecy rate, the base station uses the idea of ​​combining information signal with artificial noise. In this regard, optimizing the transmitted signal power with artificial noise and the use of appropriate radiation in Alice and also the design of the optimal phase shift in the Intelligent Reflective Surfaces are presented. In addition, in this paper, we calculate a closed form expression for the cumulative distribution function of the legal and illegal channels to evaluate the ergodic secrecy rate. Numerical results show that, firstly, the proposed design improves the ergodic secrecy rate metric and, secondly, the proposed design offers better performance against the phenomenon of obstruction in millimeter-wave communications.

A reconfigurable compact small size multiband meander line power divider with low insertion loss using two varactor diodes is presented in this paper. Design methods along with analysis equations are provided. The reconfigurable responses are realized by tuning the controlling voltages of the two varactor diodes thereby resulting compact size and less loss. One diode is used to tune the return loss of the input port and the other diode is utilized to tune the return loss of the two output ports and the isolation between the two ports. The tuning frequency range is from 0.8 GHz to 1.2 GHz. By tuning the bias voltages of the diodes, power divider can operate in 0.8 GHz to 1.2 GHz frequency bands.  A reconfigurable power divider is designed, fabricated and measured to validate the proposed methods. The power divider has the merits of compactness, low insertion loss, good isolation between two output ports, good return loss, and only two varactor diodes.

In this paper, a low power 2×3 matrix distributed amplifier (DA) with tapper transmission lines is introduced in 180 nm CMOS technology. The matrix structure is used to provide the mechanisms of multiplication and additive of the current for increasing the gain and reducing the power consumption. In the input stage, a controllable cascade gain cell is used to expand the bandwidth and remove to need the additional capacitors in the input gate and central transmission lines. Moreover, the terminating resistor of the input gate transmission line is replaced with an RL network. The proposed distributed amplifier is designed and simulated using TSMC 0.18 µm CMOS technology in Cadence Spectre-RF over the frequency of 1-30 GHz. Operated at 1 V, the proposed DA consumes 25.16 mW. Simulation results show that the proposed DA achieves a direct power gain (S21) of 12±1 dB with an average NF of 5.75 dB and average IIP3 of -6.11 over the 1–24 GHz band of interest. The input and output return losses are also more than 10 dB.

In this paper, the design and fabrication of a hybrid microwave integrated circuit (HMIC) power amplifier (PA) with more than 8 Watts of output power and 50 dB of large signal gain in the frequency range of 10.95-11.20 GHz for satellite communications is presented. In order to have a high-efficiency PA, the last two stages of the PA are custom designed and fabricated using GaN transistors as separate modules. To evaluate the repeatability of the designed PA, 16 samples of the custom-designed PA are fabricated and measured. All of the measured PAs provided 8 Watts of output power with more than 55% of power added efficiency (PAE). To have a gain of more than 50 dB, commercial integrated circuits (ICs) are used as gain stages. Finally, the performance of the fabricated PA is evaluated using a 10 MSym/s QAM signal and 6 Watts of average output power with 33% of efficiency is achieved at 11 GHz.

In this paper, a new digital pre-distortion model is presented for the joint compensation of power amplifier nonlinearity, gain and phase mismatch of quadrature modulator, and frequency-dependent impairments, in indirect conversion transmitters. This pre-distorter is designed based on the diagonal Volterra kernels, which is a simplified model of the Volterra series, as its parameters are linear. So, we can use the indirect learning structure to estimate the pre-distortion parameters. Due to the numerical instability that may occur in the nonlinear modelling of the power amplifier and its linearization, in this structure, a nonlinear model with basic functions based on the Jacobi polynomial is proposed, which significantly reduces the conditional number of the matrix of pre-distortion coefficients to less than 50 dB. Also, the power of the signal in the adjacent channel decreases by more than 17dB for the 16QAM-LTE and 64QAM-LTE input signals and 14 dB for the OFDM input signal. Simulation results in MATLAB software confirm the efficiency and performance of this model. NMSE and EVM parameters are also introduced to analyze and evaluate the proposed pre-distortion structure, which is simulated in MATLAB for the 16QAM-LTE input signal. The results showed an NMSE value of -45 dB and EVM value of -37 dB for Jacobi DPD with a DPD coefficients number of 19.  The values of these parameters have been calculated and plotted for the various Jacobi polynomial orders, and the results have been significantly improved compared to the conventional memory polynomial model.

In this study, a plasmonic infrared refractive index sensor (RI) based on the Metal-Insulator-Metal (MIM) structure by two gold nanorods in the top layer to measure the analytic refractive index has been proposed. Sensor characteristics such as Sensitivity (S), Full Width at Half Maximum (FWHM) and Figure of Merit (FOM) were calcuted 736.842 nm / RIU, 40 nm and 18.421, respectively.Since the two-dimensional nature of collective provocations in graphene makes plasmon imprisonment in graphen is more powerful than metals and in addition, the  metals does not ideally absorb biological molecules, A graphene layer has been added to increase the sensitivity of the sensor. Finally, the gold nanorod sensor using graphene layer between the nanorods and dielectric was investigated and the characteristics of the sensor with thickness of graphene layer, 2 nm and chemical potential of 0.7 eV have obtained as S = 778.571 nm / RIU, FWHM = 46 and FOM = 16.045.

In this paper, capacitive tilt micro sensors based on the MEMS technology are designed, analyzed and simulated. These micro sensors are cylindrical and interdigital capacitors (IDC) of the structure from copper, are located on one of their bases. Electrodes of the IDCs are located on the same plate and connected to the ground or terminal voltage. Half of the cylinder is filled with the silicone oil while there is air in the other half. By rotating the cylinder around the x-axis, as the oil level inside the cylinder remains horizontal due to the gravity, position of the electrode’s surfaces which are located inside the oil changes, so tilt around this axis is measured. Two structures are proposed to measure tilt around the x-axis. In this paper, using detailed analysis, the effect of the angular changes on the structural capacitors and outputs is investigated. Moreover, the proposed structures have been simulated by COMSOL Multiphysics software. The results of simulation and analysis are in good agreement. The measuring range of these structures is -90 ° to 90 ° or 0 ° to 180 °  with the sensitivity of 44 fF/deg and 88 fF/deg for one and two-capacitor structure, respectively.

The happiness, fear, sadness, anxiety, and etc. feelings appear in the face for a short time after controlling and suppressing. This unmanageable expression is micro-expression. It can show trickery and lie. Therefore, its recognition has a wide of applications in tribunal and therapy. However, it is necessary to identify the rapid and weak movements of the facial muscles. To this end, in this paper, we magnify motions and we have considered six planes along the time dimension to identify changes well. We also propose the multi-color uniform local binary pattern from six intersection planes and the histogram of gradient direction from these planes, which produce good results in micro-expression recognition. These methods can be used in a real environment. Because they work better in illumination and light changes. The result of the experiments shows 74.12%, 86.19%, and 66.23% accuracy using the proposed method on CASME I, CASME II, and SMIC-NIR databases, respectively. Thus, our method has promising performance in comparison with the methods of previous researches.

Deep Semantic segmentation of images as an integrated solution for image analysis is based on the classification of individual image pixels, especially in applications such as oil spill detection in the marine areas which have no clear boundaries. To curb pollution and environmental hazards caused by oil spills, it is important to provide more accurate algorithms. Synthetic aperture radar images are widely used in the oil spill detection field. In these images, there are challenges such as speckle noise as well as distinguishing between oil spills and lookalike areas. The application of new machine learning methods can help reduce the involvement of human taste in decision-making. In this paper, the feature channel shuffling method on CNN networks, atrous block, and decoder parts are used and the computational complexity is drastically reduced and also provides much better oil spill segmentation results than other methods. The proposed network architecture is based on the Vgg16 architecture. The overall accuracy, accuracy, intersection over :union:, weighted IoU, and BF score is used as evaluation parameters. In the proposed method, the accuracy of detecting the oil spills and look-alikes was improved by 7.8% and 7.3%, respectively, compared to the previous simulated methods.

Imposing a constraint on the gradient descend algorithm for the purpose of neural network training with limited weights has several applications such as network transparency, reducing network volume in terms of storage, increasing the level of generalizability. In addition, it speeds up convergence and finds a more accurate answer. In this paper, using the kernel trick as a method for imposing various constraints on the training algorithm, 21 different constraints have been compared those 16 constraints of which are novel and inspired by the existing uncertainty in biological neural networks. There have been no data augmentation or regularization techniques are used to clearly show the effect of each constraint functions. To solve the classification problem of MNIST, CIFAR-10 and CIFAR-100 datasets for each constraint function, 63 experiments have been done. The results show constraint functions have different impact on solving each dataset and specially our biologically inspired constraints may lead to train a more accurate neural network than the other constraints.

One of the platforms for implementing quantum technologies is optical systems where the cavity is a major component. Furthermore, state estimation has a significant role in understanding the behavior and control of quantum systems. This paper concerns derivation of quantum-filtering equations for a one-sided cavity driven by the field in the vacuum state. To this end, the number operator has been selected as an appropriate observable for the cavity. Then, filtering equations, namely stochastic master equation, have been derived considering observations which are measured by monitoring the output field either by Homodyne or photon detector. In addition, the effect of changing cavity parameters has been investigated on its behavior. Finally, the accuracy of the result has been validated by simulating the derived stochastic master equations and comparing with the master equation.

​One of the most important issues in the field of the mobile swarm robot is the problem of path planning.  The goal of path planning is determining the optimal trajectory of the path for multi-robot in a defined search space that directs the robot from a start point to a destination point. This optimal trajectory should be such that the robot does not encounter obstacles and other robots. It should be mentioned that the optimal trajectory should also meets the requirements of being safe and optimal; moreover the optimal trajectory should be the lowest cost path. So far, numerous literature have been devoted to swarm robotic path planning in two dimension environments. But in the real environment, the path is not quite flat, and the search space has lots of roughness. In this paper, the problem of path planning for a swarm of robots will be solved in a three-dimension (3D) feasible search space using the improved particle swarm optimization method. In the experiments, the proposed method is examined in term of number of robots and different environments. Simulation results demonstrate the effectiveness of the proposed approach in minimizing path length and travel time.

In this paper is proposed a new approach to controlling a flexible-joint robot manipulator based on a predictive controller and voltage control strategy in the presence of external disturbances. In this approach is used a 2DSFJ[1] Robot for modeling, design, analysis, and simulation.  The controller design is done once with feedback from all system states (main controller) and again only with feedback of the motor position (reduced controller), which in this case, some hardware equipment will be eliminated. The designed controller will have the advantages of predictive control method and voltage control strategy at the same time. So, in addition to simplicity, the control law has an independent joint structure. Also this approach resolves the limitations of the torque control strategy, such as the high computational volume, ignoring of the actuators electrical dynamics, and implementation problems. The prominent features of the proposed approach as follow: online optimization of the voltage control signal, design of predictive controller based on voltage strategy and controller order reduction. Other advantages of this approach include: simplicity in design and proof of stability, flexibility in tuning the sampling time and predictive control horizon, consideration of system input and output limitations in the voltage control law, good disturbance rejection, and improved performance of closed-loop system. In order to analyze the performance of the proposed approach, stability analysis of control system for both main and reduced controllers was performed by Lyapunov-based method, and computer simulation in problem of regulation and tracking of the reference signal along with increasing and decreasing its frequency for evaluating the response speed, and adding external disturbance by using the main and reduced controller will be done. Also, the advantages and limitations of the presented approach in terms of hardware and controller performance are analyzed by comparing the results of the simulations with other papers, and thus the stability of the closed-loop system and its proper performance is guaranteed.

The process of empowering muscles in order to make them to a normal and common value is an expensive and prolonged work. There are some commercial exercise machines used for  this purpose called rehabilitation systems. However, these machines have limited use because of some reasons. In this paper, an algorithm and an improved rule are presented for controlling a rehabilitation system of lower limbs which is implemented on a 3-Degree Of Freedom  planar robot. Estimation and optimization of control parameters will be done by artificial neural networks and genetic algorithms, respectively (intelligent strategy). MATLAB Simulink is used for simulations. It shows that proposed algorithm in comparation with other similar methods is a low-cost method and needs to less energy and force with high accuracy.

In this paper, the 2-D modified Fornasini-Marchesini (MFM) state-space model of iterative learning control for uncertain state-delayed systems is introduced. Based on the 2-D model, the problem of designing a robust iterative learning controller is transformed to robust H_inf  stabilization of a 2-D state-delayed system. Dynamic output feedback controller is applied to stabilize the uncertain 2-D model of the process. Linear matrix inequality (LMI) approach is used to find the coefficients of the controller. Finally, two numerical examples are simulated by MATLAB software to compare the proposed method with the classic ILC method and to show the efficiency of the proposed ILC method.

In this paper, the performance of the internal model controller on the behavior of suspended load quadcopter is investigated, where due to the inherent instabilities of the quadcopter and the hazards and instabilities that may be caused by suspended mass oscillation, the use of a method to stabilize and ensure the safe performance of the system is important and necessary. The proposed controller in this paper is responsible for controlling the attitude and position of the quadcopter and maintaining the overall stability of the system as well as the robustness against internal disturbances and suspended load swing. Internal model control is a robust model-based controller that directly uses the nominal system model and its approximate inverse in its synthesis. When the inverse of nominal model is improper, a filter is also used in the design. The compensator, used to control the system, employs only one control parameter. This is the reason of the controller synthesis simplicity as well. The required nominal model, used in internal model controller structure, is obtained via linear parameter varying modeling method. Furthermore, in order to reduce the suspended load oscillation, the input shaping technique is used, where the simulation results show the appropriate performance of the system to maintain attitude and position of the quadcopter and suspended load swing reduction during the path tracking.

This paper studies the Platooning of heavy vehicles based on the Internet of Things. In this context, the term “Platoon” means that the cars move in groups together in a certain order and direction, led by one of their teammates. Platooning of heavy vehicles are very effective in reducing the pollution, the traffics and its consequent accidents, while increasing the road capacity. In the Internet of Things, vehicles can connect together over internet and share their position and other data that are necessary to control vehicles in group that is called V2V (Vehicle-to-Vehicle) communication. When vehicles controlled over internet of things, drivers can receive information such as road traffic, and traffic sings which is called V2I (Vehicle-to-Infrastructure) communication. In this paper, a novel resilient controller is designed to prevent cyber-attacks while it is robust against static/moving obstacles inside the group. At the end, the proposed theory is implemented on two real non-holonomic robots with three wheels. Moreover, a number of virtual robots are designed and implemented inside software-in-the-loop system to emulate the Platoon. The results confirmed the superiority of the proposed method.

Permanent magnet synchronous motors are widely used motors in various industries. Speed ​​control of these motors is of special importance that can be done using a closed loop control system. Due to the nonlinear dynamics of this type of motor, in recent years, nonlinear controllers have been used to control permanent magnet synchronous motors. One solution for controlling uncertain nonlinear systems is to use an observer-based nonlinear controller. In this method, uncertainty estimation is provided to the controller using an extended state observer. Another challenge in controlling a permanent magnet synchronous motor is the two input- two output of the dynamic system of this type of motor. Therefore using multivariate control techniques will be desirable for the design of the control input vector. In this paper, a combined vector observer-based nonlinear method is used to control the two inputs - two outputs system of a permanent magnet synchronous motor. For this purpose, using an extended observer, the load torque, which is assumed to be uncertain, is estimated, and then the smooth multi-input-multi-output sliding mode method is used to design the control input vector. The proposed hybrid method has the advantages of nonlinear, robustness to uncertainty, finite time convergence, and vector based controller design. These features are demonstrated analytically as well as using computer simulations.