نشریه مهندسی برق و مهندسی کامپیوتر ایران
سال نوزدهم شماره 2 (پیاپی 63، تابستان 1400)

Cellular learning automaton is an intelligent model as a composition of cellular automaton and learning automaton. In this study, an extended algorithm of cellular learning automata is proposed based on transfer learning as the TL-CLA algorithm. In this algorithm, transfer learning is used as an approach for computation deduction and minimizing the learning cycle. The proposed algorithm is an extended model based on merit function and attitude vector for transfer learning. In the TL-CLA algorithm, the value of the merit function is computed based on the local environment, and the value of the attitude vector is calculated based on the global environment. When these two measures get the threshold values, the transfer of action probabilities causes the transfer learning from the source CLA to the destination CLA. The experimental results show that the proposed TL-CLA model leads to increment the convergence accuracy as 2.7% and 2.2% in two actions and multi-action standard environments, respectively. The improvements in convergence rate are also 8% and 2% in these two environments. The TL-CLA could be applied in knowledge transfer from learning one task to learning another similar task

Applying a hierarchical routing approach based on clustering technique and mobile sink has a great impact on reducing energy consumption in WSN. Two important issues in designing such an approach are cluster head selection and optimal allocation of mobile sinks to critical regions (i.e., regions those have low remaining energy and thus, high risk of energy hole problem). The limited number of mobile sinks should be utilized due to a high cost. Therefore, allocating the limited number of mobile sinks to the high amount of requests received from the critical regions is categorized as a NP-hard problem. Most of the previous studies address this problem by using heuristic methods which are carried out by sensor nodes. However, this type of solutions cannot be implemented in real WSN due to the sensors’ current technology and their limited processing capability. In other words, these are just theoretical solutions. Consequently, a semi-central genetic algorithm based method using mobile sink and clustering technique is proposed in order to find a trade-off between reduction of computation load on the sensors and increasing accuracy. In our method, lightweight computations are separated from heavyweight computations. While, the former computations are carried out by sensors, the latter are carried out by base station. Following activities are done by the authors: 1) cluster head selection by using effective environmental parameters and defining cost function of cluster membership, 2) mathematical modeling of a region’s chance to achieve mobile sink, and 3) designing a fitness function to evaluate the fitness of each allocation of mobile sinks to the critical regions in genetic algorithm. Furthermore, in our activities minimizing the number and length of messages are focused. In summary, the main distinguishing feature of the proposed method is that it can be implemented in real WSN (due to separation of lightweight computations from heavyweight computations) with respect to early mentioned objectives. The simulation results show the better performance of the proposed method compared to comparison bases.

Network-on-chip is a router-based paradigm that determines the path of packet passing from the source to destination by a routing pattern through simplified protocols of the public data communication network. Sometimes, it is impossible to send packets from source to destination due to the communication problems caused by network elements in NoC such as routers and faulty links. In most cases, fault-tolerant algorithms select a reliable path using definite criteria. Therefore, in this paper, a reliable path is selected using a multi-criteria decision making technique through an adaptive approach according to the density status received from the adjacent nodes along with the path length so that when a failure occurs, a reliable path with similar QoS features is replaced by rating different paths among network nodes. The weight path selection strategy in NoCs to detect the minimal output port and multi-criteria decision making approach with VIKOR method has improvement over the basic routing algorithm in terms of delay and throughput. The algorithm hardware overhead has a reasonably low cost that maintains scalability for large scale On-Chip networks implementations.

Nowadays, cloud computing industry has been transformed to a new supply chain between cloud service providers and service requesters. To this end, cloud service provisioning in datacenters is procured via virtualization platforms in which it can potentially increase the utilization of computing resources at infrastructure level. Inefficient virtual machine placement (VMP) schemes lead lower system utilization, increase of resource dissipation, and high rate of power consumption. Therefore, this paper formulates VMP problem on physical machines of cloud datacenters to a multi-objective optimization problem with minimization of both power consumption and resource dissipation perspectives which is computationally NP-Hard. Since the most meta-heuristic algorithms are designed for continuous optimization problems and are also susceptible to get stuck in local optimum, to figure out this combinatorial problem an optimization algorithm based on simulated annealing algorithm commensurate with discrete search space of stated problem is extended, so that the possibility of getting stuck in local optimum is reduced. To validate the proposed approach, several scenarios are introduced and conducted. Reported results from simulation of different scenarios show that the proposed approach outperforms against other existing approaches in terms of reduction in power consumption, resource dissipation, and the number of active server in use.

The rapid growth of Location-based Social Networks (LBSNs) is a great opportunity to provide personalized recommendation services. An important task to recommend an accurate Point-of-Interests (POIs) to users, given the challenges of rich contexts and data sparsity, is to investigate numerous significant traits of users and POIs. In this work, a novel method is presented for POI recommendation to develop the accurate sequence of top-k POIs to users, which is a combination of convolutional neural network, clustering and friendship. To discover the likeness, we use the mean-shift clustering method and only consider the influence of the most similarities in pattern’s friendship, which has the greatest psychological and behavioral impact rather than all user’s friendship. The new framework of a convolutional neural network with 10 layers can predict the next suitable venues and then select the accurate places based on the shortest distance from the similar friend behavior pattern. This approach is appraised on two LBSN datasets, and the experimental results represent that our strategy has significant improvements over the state-of-the-art techniques for POI recommendation.

In forensic dentistry, age is estimated using dental radiographs. Our goal is to automate these steps using image processing and pattern recognition techniques. With a dental radiograph, the contour is extracted and features such as apex, width and tooth length are determined, which are used to estimate age. Optimizing the resolution of radiographic images is an important step in contour extraction and age estimation. In this article, the aim is to improve the image resolution in order to extract the appropriate area and proper segmentation of the tooth, which makes it possible to estimate age better. In this model, due to the low resolution of radiographic images, in order to increase the accuracy of extracting the desired area of each tooth (ROI), the image resolution increases using spatial entropy based on the spatial distribution of pixel brightness, along with another increasing resolution method, like the Laplacian pyramids. Increasing the resolution of the image leads to the extraction of appropriate ROI and the removal of unwanted areas. The database used in this study is 154 adolescent panoramic radiographs, of which 73 are male and 81 are female. This database is prepared from Babol University of Medical Sciences. The results show that by using fixed tooth segmentation methods and only by applying the proposed effective method to improve image resolution, the extraction of appropriate ROI increased from 66% to 78% which shows a good improvement. The extracted ROI is then delivered to the segmented block and the contour extracted. After contour extraction, age is estimated. The age estimation using the proposed method is closer to the manual age estimate compared to the method that does not use the proposed algorithm to increase the image resolution.

Unlike classical cryptography, where security is based on computational complexity, quantum cryptography has unconditional security, which is based on physical constraints. So far, the semi-quantum version of many of the problems of secure quantum communication protocols has been proposed. In this study, we examined semi-quantum protocols that allow users to access a secret message directly without distributing the key. An important factor used to analyze the performance of secure quantum direct communication protocols is efficiency. In this study, the proposed semi-quantum secure communication protocol against various quantum attacks has been investigated. In the proposed scheme for decoding the confidential message by the receiver, a sequence of single photons is required, which is first generated by the controller. The proposed protocol has a yield of 50%, which is higher than the previous protocol, which has a yield of 66.6%.

Graphs are common data structures which widely used for information storage and retrieval. Occasionally some vertices of a graph contain specific features or information, which we value in their effect. We consider modeling this effect formally, and we devise two super-fast algorithms to approximate the colored average degree. In the first method, we assume the information of each vertex is available; hence, the provided algorithm works with a 2+ϵ approximation factor. Eventually, we waive this assumption and find another algorithm with the same approximation factor, which computes the answer in the sublinear expected time.