فصلنامه پدافند الکترونیکی و سایبری
سال یازدهم شماره 4 (پیاپی 44، زمستان 1402)

Influence operations are considered as one of the types of information operations along with cyber network operations and electronic warfare operations. The spread of online social networks has provided a suitable platform for influence operations. Therefore, it is essential to be able to evaluate influence operations in online social networks. The ability to evaluate influence operation requires the existence of a conceptual model that is connected to a consistent theoretical structure on the one hand, and has the ability to be implemented computationally on the other hand. The aim of this article is to present a conceptual model that provides a basis for computational modeling to evaluate influence operations in online social networks. In this article, after an in-depth and exploratory study in the research literature, using the meta-combination method and interviews with experts, the theoretical framework of cyber cognitive power and the characteristics, concepts, dimensions and components of the conceptual model are presented. Two-step Delphi method was used to evaluate the validity of the presented model. The proposed model was confirmed in terms of validity and reliability and based on the Delphi method, it was agreed by more than eighty percent. In this research, a new conceptual formulation for influence operations in online social networks based on the conceptualization of cognitive capital and cyber cognitive power is presented. The proposed conceptual model, which is presented descriptively, has provided the possibility to evaluate the success indicators of the influence operation. The advantage of the proposed conceptual model is that it evaluates influence operation based on the concept of power and provides computational evaluation of influence operation. Also, this model provides the possibility to identify and detect influence operations in online social networks.

Class-imbalanced datasets are common in many real-world domains, such as health, banking, and security. Machine learning researchers have recently focused on the classification of such datasets, where the costs of different types of misclassifications are unequal, the classes have different prior probabilities, or both. The performance of most standard classifier learning algorithms is significantly affected by class imbalance, where the algorithms are often biased toward the majority class instances despite recent advances in deep learning. However, there is very little empirical work on deep learning with class imbalance.To address this issue, we propose an incremental weighted cross entropy loss function. The proposed method involves gradually increasing the weight of the minority class as the training progresses, until it reaches the specified amount at the end of the training. Through experiments, we demonstrate the convergence and efficiency of the proposed method. The results of experiments on three datasets, including artificial datasets, human activity recognition dataset, and CIFAR-10, demonstrate the convergence and performance of the proposed method. The proposed method is compared with decision tree-based AdaBoost, Cross Entropy-based convolutional neural network, weighted Cross Entropy -based CNN, SMOTE method, and ensemble CNNs method. With accuracy gains of 94.6%, 92.92%, and 69.23% on the three datasets (CIFAR-10 with 5% imbalance rate), the proposed method outperformed the other methods. Additionally, the accuracy on the artificial dataset was 17.77% higher than the traditional decision tree-based AdaBoost method.

Delay tolerant networks are a group of wireless networks that suffer long-term and frequent disconnections due to the scattered distribution of nodes and their movement. In these networks, which are a type of mobile networks, there may not be a connected path from the sender to the receiver, so message delivery is done by opportunistic routing based on the store, carry and Forward pattern. Routing is one of the main challenges in these networks. Routing consists of the steps of choosing a replay node, buffer management and data transmission. Various methods have been proposed to deal with the routing challenge. Most of these methods are only about replay node selection and few have discussed about buffer management and alternative packet selection. Considering that the problem of choosing an alternative package from the buffer is decision-making, it seems that using game theory can be effective in improving buffer management. In this paper, at each step of the routing process and when two nodes meet after determining the sender and receiver nodes, a method is presented that, based on game theory, decides with which packet from the receiving node's buffer if the buffer of the received packet is full, to be replaced. The simulation results of this method using THE ONE simulator show that buffer management with the proposed method increases the delivery ratio and reduces the average delay and overhead ratio .

To evaluate the software quality it is required to measure factors affecting the qualityof the software . Reliability and number of faults are examples of quality factors. If these factors are measured during the software development life cycle, more efficient and optimal activities can be performed to improve the software quality. The difficulty is that these factors could be measured only at the ending steps of the life cycle. To resolve the difficulty, these factors are indirectly measured through some software metrics which are available at the early stages of life cycle. These metrics are used as the input to fault prediction models and software components which may be faulty are the output of these models. Prediction of fault prone modules is a well known approach in software testing phase. When a module is predicted to be faulty, apparently more efforts have to be paid for correcting it. In addition to the module, all its dependent modules require specific consideration. When modifying a module all its dependent modules may be affected. The difficulty is that current known metrics for fault prediction do not reflect this situation. To resolve the difficulty, in this thesis, new metrics are introduced. Our experimental results show that the more the dependees of a module are changed, the more fault prone the module will be

The Internet of Things is an emerging information architecture based on the Internet that develops interaction between things and services in a safe and reliable environment. In fact, the purpose of this structure is to reduce the distance between the things of the physical world and information systems. In the Internet of Things, it is expected that intelligent devices will become active members in business and informational and social processes, so that they are able to interact between themselves and the external environment through the exchange of data and sensed information. In fact, the Internet of Things is a network of devices in which various things can communicate with other equipment with the help of computers and through Internet connections.  Recommendation technologies can help to more easily identify relevant artifacts and thus will become one of the key technologies in future IoT solutions.  The main task of recommender systems is to recommend service providers that meet the different needs of users. The paper porposes a Support Vector Machine (SVM) based algorithm and the TOPSIS multi-criteria decision-making model in order to create an effective recommender system and provide suggestions to users based on their preferences and increase user satisfaction. The experimental results show that the proposed recommender system can produce a series of objective recommendations that are effective based on accuracy and variety, novelty and high coverage. Finally, the results confirm the improvement in making recommendations .

In wireless communications, the channel is available to everyone and therefore, any receiver in the telecommunication coverage area can receive the transmitted signal. In such cases, unauthorized users may use this feature to eavesdrop the information, disrupt data transmission, reduce network performance, and so on. Recently, using compressive sensing in physical layer security has attracted the attention of many researchers. Compressive sensing is very useful when the signals are sparse or compressible such as in image signal processing or image recognition. This is because it can be considered as an encryption system, sampling, compression and encryption, while maintaining a secret matrix. In this paper, we examine the physical layer security in a cooperative wireless communication based on several consecutive relays. In such a model system, the equivalent channel matrix is used as the secure measurement matrix. Since the majority of research work in this field consider relay hardware to be ideal and free of hardware impairments, in this research work, this practical assumption is made in the physical layer security with the help of compressive sensing. We consider a cooperative communication with multiple-hops. Finally, we will propose a solution to improve the secrecy performance of such a system. Simulation results show that in worst case, the screcy rate increases at least 10 percent and this would be increased to 120 percent in the best case. Moreover, computational complexity in terms of simulation run time is increased by 22 percent.

Cyberphysical systems have achieved rapid development over the past decade due to their widespread applications in various domains, and are typically composed of sensing, communication, computing, and actuation capabilities. However, the communication network of physical security systems may be attacked by cyber attacks, which significantly destroys the performance of the system. Recently, DOS denial-of-service attacks (1) have attracted a lot of attention. In this article, a solution to deal with DoS attacks on the power sharing control system of the electrical energy distribution network, including distributed generation, is presented. So that a time-varying distributed estimator has been used. Based on the feature of service interruption communication duration, the convergent design conditions of observer parameters are obtained using Lebag2 integral theory and average dwell time method. It is also shown that the observer errors converge. Finally, the presented coping method is placed on the virtual impedance control line 3 of the power sharing control system and will neutralize service interruption attacks

The principal component analysis (PCA) method is one of the well-known dimensional reduction methods, The PCA has many applications in big data analysis from various fields. PCA is an essential method for image processing that is used directly or after several stages of preprocessing and in combination with other methods. Face recognition methods based on principal component analysis have many applications in face detection and recognition. In this paper, we present a cost-effective algorithm for human face recognition based on principal component analysis, which combines the Mahalanobis distance with the PCA method, the ability to detect faces in the shortest possible time for low-quality and black and white images. The architect of this method is modular, and every part of it can be hybridized with other methods. The proposed method is expressed and discussed in terms of parameters for determining the complexity and computational efficiency. Overall, it can be said that the method presented compared to other methods can process images with very low resolution and color depth, is able to recognize the face based on the B&W images, has no need for robust and costly computer systems, has a modular structure, and customizable based on distance (For example, a 30 percent increase of recognition rate from 49 % to 79 % in some implementations).

Today, images are used as powerful communication tools and sources of information. In certain applications, such as medicine, justice, and forensics, images serve as evidence. Therefore, the validity of an image is crucial. With the spread and availability of image editing tools, people can easily manipulate images to their advantage. They follow political, cultural, economic, and social issues by adding or removing elements from images, often distributing misinformation. Consequently, forgery detection is one of the most important and challenging topics in the field of computer vision. This research aims to identify forgery and healthy images and pixels using a hybrid deep learning network. In the proposed method, three pre-trained networks—VGG16, MobileNet, and EfficientNetB0—are employed in three different branches. To detect forgery at both the image and pixel levels, the output feature maps from these branches are merged in a concatenate layer. Subsequently, a global average pooling layer and a scoring layer are used to identify forgery and healthy images. Additionally, feature maps combined from the three branches are utilized to create a heat map image for forgery detection. Notably, pixel forgery detection is performed solely using the heat map image generated from the combined network, without relying on ground truth images that specify the forgery area during training. The proposed method is evaluated on the well-known CoMoFod dataset, demonstrating satisfactory performance against forgery images with various geometric transformations and post-processing operations

Frequency hopping spread spectrum (FHSS) communications are used widely in military and commercial communications. Therefore, estimating the frequency hopping parameters of signals is of great importance. This article examines the problem of separating and estimating the parameters of multiple synchronous and time-interfering frequency hopping wideband signals using receivers with low bandwidth and low sampling rate. Due to minimal knowledge about the transmitted signals, the problem is analyzed in semi-blind mode. For this purpose, a hybrid method based on array processing is developed based on the modulated broadband converter (MWC) and processing in the time-frequency domain by considering the spatial information of the signal. The above method consists of two steps: the first step is receiving frequency hopping signals by uniform linear arrays (ULA) and passing through the MWC with minimal components. the next step, the joint diagonal (JD) method is applied to the space-time-frequency distributions (STFDs) of the MWC output data. This leads to the estimation of the separator matrix and finally to the separation of the signals and the extraction of the frequency hopping pattern. The combined MWC-STF-JD method is the idea used in solving the problem, where MWC features and processing in the STF domain are used simultaneously. The results of the estimation of the desired parameters, compared to other traditional methods of source separation such as Compressed Sensing (CS) analysis based on evaluation indices such as root mean square error (RMSE), have led to improved performance in lower signal-to-noise ratio (SNR). So that the amount of error in the range of SNR=4 dB reaches the minimum error faster by 6 dB. Among the other features of this method, it is possible to reduce the complexity of STFDs processing calculations because of the low sampling rate MWC and the possibility of implementing the design with minimal components.

This research has been done with the aim of providing a routing method in Internet of Things networks based on fuzzy logic and clustering for the use of smart transportation. Today, transportation is one of the basic issues and major problems in cities and this issue bigger cities such as Tehran have become an important concern, hence transportation systems have been created and expanded. But rapid population growth, migration to cities, increasing car ownership and the use of personal vehicles are the challenges that urban transportation systems are facing. Meanwhile, passenger transportation systems can play a significant role in improving the city's traffic situation. Cluster-based routing protocols are one of the most important ways to reduce energy consumption in wireless sensor networks. In this thesis, a new clustering protocol using bee colony optimization algorithm is presented and a new method to reduce energy consumption and. Lifetime improvement in wireless sensor network routing includes three stages of clustering sensor nodes into a number of clusters using fuzzy clustering algorithm, selecting optimal cluster heads and data transmission. The results of implementing the proposed method and comparing its performance with other methods the former showed the proper performance of this method in this field

With the expansion and popularity of social networks among people, special attention has been focused on the activities, reactions, and feelings of people in these networks compared to other circles. Analyzing this volume of unstructured textual information of users requires new and optimal methods of text mining and natural language processing. The purpose of sentiment analysis is to examine a large volume of opinions about an entity by the machine and provide a summarized report of the sentiment expressed in it to the user. To achieve this goal, statistical techniques, data mining, and natural language processing are used. In this article, a new method based on Archimedes' optimization algorithm is presented so that users' opinions can be obtained in less time and with higher accuracy. Also, in order to eliminate one of the main challenges in sentiment analysis, in the feature extraction phase, all ironic sentences are collected and entered into the collection as sentences with a specific class, in order to eliminate this big challenge. This method can be applied to different languages and tries to increase the accuracy and speed of previous algorithms. The evaluation results of the data set show that the proposed method has an accuracy of 0.967.