Master in Electronics and Communications

Sep 1, 2014 - Jan 1, 2017

I have a master degree in Electronics and communications. I published many papers about deep learning, optimization, and neural networks.

Deep Learning-Based Intrusion Detection in Wireless Sensor Networks using Optimized Convolutional Neural Network with IGOA Algorithm

Jul 5, 2025

Journal Scientific Research Journal of Engineering and Computer Sciences

publisher https://www.rame.org.in/ijceae/

DOI https://doi.org/10.47310/srjecs.2025.v05i02.001

Issue 2

Volume 5

The open nature and resource limitations of Wireless Sensor Networks (WSNs) make them more exposed to numerous cyber threats. IDSs function effectively as key components for securing such networks. The research introduces a new deep learning intrusion detection system that integrates IGOA with CNN to enhance WSN intrusion detection capabilities. IGOA automatically optimizes CNN classification performance through an improvement of both accuracy and generalization capabilities. The proposed method demonstrates superior performance in WSN-DS dataset experiments since it reaches 99.94% accuracy that exceeds several state-of-the-art methods. Multiple assessments of precision, recall, F1-score and confusion matrix analysis and ROC-AUC metrics prove that the model can deliver reliable performance in real-world applications.

Read Publication

A Hybrid Deep Learning and Bagging Method for Automatic Modulation Recognition Utilizing Time-Frequency Data

May 21, 2025

Journal CENTRAL ASIAN JOURNAL OF MATHEMATICAL THEORY AND COMPUTER SCIENCES

publisher https://cajmtcs.centralasianstudies.org/index.php/CAJMTCS

Issue 3

Volume 6

In satellite communications systems, submarine communications, and military communications, determining out the type of modulation is a crucial issue. A newly developed digital modulation classification model is introduced in this study to identify many types of modulated signals. At the first step, the density of specter for the frequencies companied with the modulation signals at the scalogram image is visually represented using continuous wavelet transform (CWT). Then, a deep convolutional neural network (CNN) is utilized to extract features from the scalogram pictures. The MRMR method is then used to get the best features. By decreasing the size of the features, the MRMR method improves classification speed and model interpretation. Using the group learning technique, the modulations are categorized in the fourth stage. Modulated signals with various levels of noise and SNRs ranging from 0 to 25 dB are taken into consideration in the simulations. The simulations' result shows that the suggested model outperforms other earlier research and functions effectively related to various noise levels.

Read Publication

Study and Evaluation a New Predictive Control Method for Speed and Stator Current Control of Induction Motor

Apr 1, 2024

Journal BIO Web of Conferences

publisher BIO Web of Conferences

DOI https://doi.org/10.1051/bioconf/20249700109

Issue 00109

Volume 97

It is clear that modern industries rely heavily on electric motors, especially induction motors. These motors convert electrical energy into mechanical energy. The distinction in the performance of the induction motor lies in that it is powerful when exposed to various operational and environmental variables, and it is also inexpensive. However, there are many traditional disadvantages that appear during the operation of the induction motor in its non-linear mechanical properties in addition to the difficulty in regulating the speed of the motor. In this paper , we present a new control method for controlling the stator current and speed of induction motor based on current control with speed control technique. The present model is based on conventional predictive controller development with a structure which is similar to rotor control and the direct torque control .It has double loops and both loops will use the prediction power. The inner loop controls the stator current based on Finite Control Set - Model Predictive Control (FCS-MPC) and the outer loop controls the speed to maximize the dynamic response of the loop. The MATLAB software has been used to implement the controller circuit. The obtained simulation results indicates that the presented control method has comparable performance to conventional controllers with some reduction in the overshoot and fast response interval.

Read Publication

Calculations of the Effect of Photovoltaic Renewable Resources on the Power Generation Grid Continuity

Feb 2, 2023

Journal International Journal of Computational and Electronic Aspects in Engineering

publisher https://www.rame.org.in/ijceae/

DOI https://doi.org/10.26706/ijceae.4.1.202311751

Issue 1

Volume 4

In order to identify these challenges and investigate the potential for solutions, it was it is vital to research and evaluate how integrating renewable energy sources into the electrical power network would affect the system. This was done to be able to address the increasing need for electrical energy as well as the ongoing integration utilize electric power systems and renewable energy sources (RES). Additionally, as the need for affordable, clean energy, such as renewable sources, has grown globally, so too has its availability. This has had an influence on the efficiency, reliability, and performance of electrical distribution networks. This study evaluated the impact of PV sources on by observing and documenting the electric power grid energy losses from acting or responding sources, as well as the grid networks' voltage variations, which resulted from The use of renewable energy sources is being integrated. Several technological aspects linked A review of the voltage output's effectiveness and quality was done. MATLAB Simulation software was used to implement the simulation model in this study. The outcomes were contrasted with the IEEE 9-bus grid standard.

Read Publication

Adaptive Disassembly Using Deep Reinforcement Learning Using Path Planning Communication Approach

Sep 1, 2022

Journal International Journal of Advances in Engineering and Emerging Technology

publisher Emerging Library

DOI https://doi.org/10.26706/ijceae.4.1.202311751

Issue 2

Volume 13

For free-flying robots, Disassembly-Based Motion Planning (DBMP) is a fresh and affordable single-query sampling-based motion planning method. Disassembly-based motion generation calculates the space volume of a potential response path using space data, and then uses this information to block out large portions of the configuration area from investigation. Because of its benefits to the economy, ecology, and society, remanufacturing is receiving more and more attention. Disassembly is the first and most difficult step in the remanufacturing of a product. Disassembly sequence planning is required in order to improve efficiency and cut costs, with the goal of identifying the ideal disassembly sequence. Numerous studies have been done in this field, and numerous sequence disassembly planning strategies have been developed. In order to enable speedy repair using virtual reality technology, disassembly planning of faulty components must be done on site and quickly. This is necessary due to the significant discomfort caused by the development of a random defect during maintenance operations. In order to address dynamic and stochastic challenges, this study uses deep reinforcement learning to tackle adaptive disassembly sequence planning. On the basis of user inputs, sequences may be dynamically constructed in the virtual reality sustaining training system. Disassembly The disassembly process is described and simulated using Petri networks and the planning of the disassembly sequence is thus referred to as a Markov decision process.

Read Publication

Enhancement in PI Parameter Prediction Using Segmented Mutation based Genetic Algorithm

Dec 1, 2020

Journal Webology

publisher -

DOI 10.14704/WEB/V17I2/WEB17025

Issue 2

Volume 17

This paper present the improvement in integration of electrical grid and induction generator with self-excited mode. This induction generators are mostly used in wind turbine. For this purpose static synchronous compensator is adapted with direct adaptive strategy with some adjustable controller parameters and an adjusting mechanism to adjust them using model reference adaptive control (MRAC). Model reference adaptive control is used to balance reactive power flow of this integration. Voltage-source inverter sinusoidal pulse width modulation handles the different operational condition in wind energy system. According to that ability of staying connected with grid in Brownout and blackout also increases. Because of occurrence of some faults at coupling of grid and generator, some abnormal operational condition generates. In proposed model, Genetic Algorithm with modified mutation function is used to tune proportional-integral controller. Instead of conventional random selection function for mutation, a modified algorithm is applied. For evaluation purpose Zafarana wind system integrated with Egyptian 220 networks is used. Results shows an improvement in performance of proposed model reference adaptive control using genetic algorithm which balances the current, voltage and speed of wind generator. It is also observed ability of staying connected with system under abnormal situation and during Brownout and blackout also increases with segmented mutation based genetic algorithm.

Read Publication

Security framework Communications of Asynchronous Transfer Mode

Jun 1, 2020

Journal International Journal of Psychosocial Rehabilitation

publisher -

Issue 4

Volume 24

There is growing enthusiasm for improving broadband communication administrations and networks for business use in both local area-wide networks. The main reasons are that demands for increased bandwidth for the interconnection of remote sites and “high speed data transmission” of mass data including photographs etc. must have to meet.The features of network traffic have also changed significantly.The form of the burst traffic is increasingly characterized by a constant requirement for multi-megabytes of bandwidth.The necessity for evolving and unclear data bandwidth transfer requirements has increased with the evolution of another era of networking technologies. The ATM system is used mainly for broadband networks communication under cellular ISDN banners. ATM provides high-speed interconnection in Mbit / s or Gbit / s across wide areas, which removes the bottleneck from networks to end frameworks.For addition, on demand, the client can access data transfer capabilities and the client is actually charged for the transmission power. As more information is transferred through ATM networks (audio, images and data), security issues get more critical. The rapidly growing usage of the Internet to transfer confidential and sensitive information only enhances the importance of security services.One could even argue that the success of ATM is not determined by its efficiency but by the degree of confidence in its performance, safety and availability. The purpose of this thesis is to address questions relating to the design of ATM network security services.