per Iranian Research Institute for Electrical Engineeringفصلنامه مهندسی برق و مهندسی کامپيوتر ايران16823745168237452024-12223151168articleAtousa Daghayeghiatousa.daghayeghi@stu.qom.ac.ir1Mohsen Nickraynickraymohsen@gmail.com2Department of computer engineering and information technology, university of Qom, Qom, Iran<p style="text-align: left;">It is hardly possible to support latency-sensitive and computational-intensive applications for mobile devices with limited battery capacity and low computing resources. The development of mobile edge computing and wireless power transfer technologies enable mobile devices to offload computing tasks to edge servers and harvest energy to extend their battery lifetime. However, computation offloading faces challenges such as the limited computing resources of the edge server, the quality of the available communication channel, and the limited time for energy harvesting. In this paper, we study the joint problem of decentralized computation offloading and resource allocation in the dynamic environment of mobile edge computing. To this end, we propose a multi-agent deep reinforcement learning-based offloading scheme that considers the cooperation between mobile devices to adjust their strategies. To be specific, we propose an improved version of the multi-agent deep deterministic policy gradient algorithm by employing the features of clipped double Q-learning, delayed policy update, target policy smoothing, and prioritized experience replay. The simulation results reveal that the proposed offloading scheme has better convergence performance than other baseline methods and also reduces the average energy consumption, average processing delay and task failure rate.</p>http://ijece.org/Article/44275Computation offloading resource allocation mobile edge computing multi-agent deep reinforcement learning energy harvestingper Iranian Research Institute for Electrical Engineeringفصلنامه مهندسی برق و مهندسی کامپيوتر ايران16823745168237452024-12223169182articlePayam Porkar Rezaeiyepayam.porkar@srbiau.ac.ir1Hamid Shokrzadehshokrzadeh@gmail.com2Dehghan Mehdidehghan@aut.ac.ir3Amir Masoud Rahmanirahmani74@srbiau.ac.ir4Department of Computer Engineering, Science and Research Branch, Islamic Azad University, Tehran, IranDepartment of Computer Engineering, Amirkabir University, Tehran, IranDepartment of Computer Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran<p class="MsoNormal" style="text-align: justify;">Nowadays integrated wireless networks have become very important. Among the important technologies in this field is the combined technology of visible light and radio frequency communications, an important example of which is the combination of Wi-Fi and Li-Fi local networks. This combination covers the weaknesses and strengthens the strengths of the local wireless network.</p> <p class="MsoNormal" style="text-align: justify;">Also, an issue that can increase productivity in the network is load balancing, especially when the presence of access points from both networks will lead to more choices. In fact, in the proposed access point selection algorithm in this research, it has been done in such a way that when being at an access point, the decision to choose the location is based on the balance between the factors in the Markov game based on the strategic behavior of objects. In this way, network delay will be reduced and load balance will be increased.</p> <p class="MsoNormal" style="text-align: justify;">Therefore, a dynamic method has been proposed, which can be used to make decisions according to the conditions at any time, especially when the topology changes in the network. The proposed method has advantages such as dynamic selection of access points according to network conditions, direct feedback on the efficiency of the network and shared channel, intelligence and learning towards changes to select points, interaction with similar agents in nodes, and reducing the probability of congestion at each access point. Also, with the increase in user traffic, which leads to congested conditions and the possibility of congestion in nodes and access points, this method helps more in terms of load balancing and reducing the level of congestion. So that its difference with compared methods that use more stable techniques such as fuzzy method increases significantly.</p> <p class="MsoNormal" style="text-align: justify;">According to the obtained results, this method has been able to improve the efficiency of the local network by more than 10% compared to the previous methods such as the fuzzy method and more than 30% compared to the SSS selection policy in high traffic load conditions.</p>http://ijece.org/Article/39801Integrated local networks Li-Fi network Wi-Fi network access points and load balancingper Iranian Research Institute for Electrical Engineeringفصلنامه مهندسی برق و مهندسی کامپيوتر ايران16823745168237452024-12223183196articleseyed ebrahim dashtisayed.dashty@gmail.com1fatemeh moayedifmoayyedi@gmail.com2adel salemiadelsalemi.sadaf@gmail.com3Larestan universityIslamic Azad University<p style="direction: ltr;">With the increasing use of Internet of Things in daily life and especially in industry, improving efficiency and delay time with the help of data offloading is one of the goals of these issues. Controlling these factors will improve energy consumption and longer use of things batteries. In this article, the method is introduced to improve sensor data processing and edge and cloud computing in industrial Internet of Things systems. The architecture is considered in accordance with the real world; in this architecture, edge servers with computing capabilities at the edge of the network, especially Used in base stations. Delay-sensitive requests can be forwarded to nearby edge servers through wireless channels, thereby reducing traffic in the core network and user data transmission latency, especially for data-intensive industrial applications. In the Industrial Internet of Things aims to manage network resources, transfer calculations and minimize energy consumption in Internet of Things devices by guaranteeing the freshness of sensor data. The network environment and input tasks are variable with time. In this article, the environment of the problem and its limitations are expressed with formulas. This problem has been solved using the proposed genetic algorithm and reinforcement learning. The proposed solution has improved the dynamic environment of the problem for offloading data and tasks by considering energy and transferring calculations and data by considering their freshness. The results show an average improvement of 40% compared to the previous methods.</p>http://ijece.org/Article/43859Offloading Industrial Internet of Things (IIoT) genetic algorithm reinforcement learning.per Iranian Research Institute for Electrical Engineeringفصلنامه مهندسی برق و مهندسی کامپيوتر ايران16823745168237452024-12223197206articleAli Habibihabibi.ali@ut.ac.ir1Ramtin Khosravir.khosravi@ut.ac.ir2School of ECE, University of Tehran, Tehran, IranUniversity of Tehran<p style="direction: ltr;">With the increasing use of software in safety-critical applications, such as the automotive, defense, and medical industries, achieving high levels of assurance regarding the quality of these software systems is essential. Model-based testing is an automated test-case generation method that, on one hand, provides relative assurance by covering a formal description of the system&rsquo;s behavior, ensuring that various execution scenarios of the program are tested. On the other hand, by automating the generation of these test cases, it significantly reduces the cost of test production. In this research, a model-based testing framework is presented that utilizes a rule-based model and can generate test cases based on two criteria: rule coverage and active rule condition coverage. To generate test cases, this framework employs a search-based approach using a genetic algorithm. The proposed method enables the definition of a system with a large state space and the generation of test cases for it. The framework has been evaluated through a case study on an embedded industrial software, and the evaluation results demonstrate its applicability to real-world problems in the industry.</p>http://ijece.org/Article/44039Model-based testing search-based testing formal specificationper Iranian Research Institute for Electrical Engineeringفصلنامه مهندسی برق و مهندسی کامپيوتر ايران16823745168237452024-12223207216articleAhmad Baratianbaratian.ahmad.22@gmail.com1Esmaeil Kheirkhahe.kheirkhah@gmail.com2Department of Computer Engineering, Mashhad Branch, Islamic Azad niversity, Mashhad, IranDepartment of Computer, Mashhad Branch, Islamic Azad University, Mashhad, Iran<p style="direction: ltr;">Considering the growth of the population of cities and the number of vehicles that are increasing exponentially, a challenge in parking lots is the positioning of vehicles. In a smart parking system, the driver can park without delay and by spending less energy; But its requirement is to use sensors (empty parking spaces) and parking guides for this purpose. With the progress of research in the Internet of Things, researchers have provided promising solutions in the smart parking system based on wireless sensors. Among these researches is the use of the gray wolf algorithm (GWO) in the optimal positioning of wireless sensors in the Internet of Things parking environment. In this article, due to the search power and high convergence of the Grasshopper Optimization Algorithm (GOA), this algorithm is used for the first time in the positioning of wireless sensors in the parking lot. The grasshopper optimization algorithm is used to determine the best anchor nodes to collect data from other sensors; So that it can reduce the positioning error and energy consumption of the sensors and increase their lifespan. The results showed that the proposed method was able to achieve an average improvement of 5.92% in reducing the positioning error, 6.43% in reducing the amount of energy consumption and 23.6% in increasing the lifetime of the network compared to the gray wolf algorithm. Also, the proposed method has been able to have more time for the first node to die, and this is an important advantage in smart parking because the efficiency of all sensors in the parking environment is required.</p>http://ijece.org/Article/45070Grasshopper optimization algorithm smart parking system sensor positioningper Iranian Research Institute for Electrical Engineeringفصلنامه مهندسی برق و مهندسی کامپيوتر ايران16823745168237452024-12223217225articleNarges Zahirinargess.zahiri@gmail.com1Fereshte Dehghanifdehghani@kashanu.ac.ir2Salman Golisalmangoli@kashanu.ac.ir3Computer Engineering Department, Faculty of Electrical And Computer Engineering, University of Kashan, Kashan, IsfahanComputer Engineering Department, Faculty of Electrical And Computer Engineering, University of Kashan, Kashan, IsfahanComputer Engineering Department, Faculty of Electrical And Computer Engineering, University of Kashan, Kashan, Isfahan<p>The emergence of the Internet of Things (IoT) has intensified the focus on web service composition and the fulfillment of increasingly complex and diverse user requirements. IoT-based systems often encounter numerous service candidates with varying qualitative attributes, presenting a significant challenge in selecting an optimal combination. This problem, categorized as NP-hard, requires efficient approaches for resolution. This study proposes a near-optimal solution for web service composition in IoT environments by leveraging the NSGA-III multi-objective metaheuristic algorithm to identify the optimal Pareto front. To further enhance the quality and diversity of the solutions, an improved algorithm integrating NSGA-III with a novel fitness function is introduced. The proposed approach optimizes service composition using nine quality parameters, which are subsequently streamlined into three principal objectives for better computational efficiency. Experimental evaluations demonstrate that the proposed method outperforms the baseline NSGA-III algorithm in terms of the average performance of two out of three objectives. Additionally, the approach achieves an average of 11% higher coverage based on performance indices and exhibits superior solution distribution and dispersion compared to alternative algorithms.</p>http://ijece.org/Article/42589Evolutionary Algorithm Internet of Things multi-objective optimization optimal web service composition and selection quality-aware web servicesper Iranian Research Institute for Electrical Engineeringفصلنامه مهندسی برق و مهندسی کامپيوتر ايران16823745168237452024-12223226234articlemeisam kargarmeisamcho2@gmail.com1Marzieh Varposhtimvarposhti@sku.ac.ir2Leila Samimirsamimi@sku.ac.i3Department of computer engineering, Shahrekorf universityShahrekord UniversityDepartment of Computer Engineering, Shahrekord University<p style="direction: ltr;">The rapid development of intelligent hardware, Internet of Things (IoT), and the emergence of various applications has led to an unprecedented increase in mobile data traffic. Therefore, the efficiency of network resource utilization and bandwidth needs to be improved effectively. Currently, Device-to-Device (D2D) communication technology can provide an effective tool for enhancing 5G networks by enabling direct communication between devices. The use of D2D communication can reduce the load on the 5G network and improve service quality. One of the main issues in this regard is how to manage communication resources and select communication links. In this article, we examine the problem of managing D2D communication links for content transmission between communication devices and formulate it as a binary linear optimization problem. To solve this problem, we propose a method based on game theory, where considering user devices containing the desired files for transmission as players, we design an exact potential game and then propose a distributed learning algorithm to reach a Nash equilibrium. Simulation results confirm the satisfactory performance of the proposed algorithm.</p>http://ijece.org/Article/43039D2D communications Potential game Edge caching 5G networkper Iranian Research Institute for Electrical Engineeringفصلنامه مهندسی برق و مهندسی کامپيوتر ايران16823745168237452024-12223235242articleMahdi Sarbazimahdi.sarbazi@uok.ac.ir1Mohammad Fathifathisam@gmail.com2University of KurdistanUniversity of Kurdistan<p style="direction: ltr;">Software-defined networks (SDNs) are an emerging area in computer networks, enabling efficient resource management in the network by decomposing data and control plans. In SDNs, network controllers acting as the network operating systems, are responsible for serving application programs. Since a control plan consists of several controllers, the placement of network controllers is a challenging issue in complex networks. While the literature has explored the number of controllers and their placement in the network, several fundamental parameters remain unexplored. Therefore, this topic remains open for more investigations. In this paper, we survey the work in the literature on controller placement in SDNs and introduce the research challenges in this area. Additionally, we present potential future research directions to advance this field.</p>http://ijece.org/Article/44474Software-defined networks meta-heuristics controller placement machine learning.