Journal of Advances in Computer Research
Volume:13 Issue: 3, Summer 2022

in Named Data Network (NDN), besides routing packets in the network, each router is able to cache and provide content. Depending on the caching policy, content can be cached in routers between the consumers and the servers. The NDN supports two streams. The flow of interest packets is propagated by the consumers in the network and passes through the nodes to the content provider routers or content producers. The flow of data packets is published from the content producer or the content provider, in reverse route. If the network leverage only one or more specific points to cache content as a provider, then the network’s load just focuses on them, and other parts of the network might be disabled. Consequently, by increasing the request rate, congestion might emerge.Fair content caching is proposed in this study to achieve load balancing in the network. Hence, the dynamic content storage approach is employed to balance the network load. In the proposed method, by balancing the load of each router in the network to the threshold limit, tries to balance the load in the entire network. In this approach, the amount of incoming traffic of each router is monitored, and if the load of each router exceeds the threshold limit, the router diverts its high-demand data to another part of the network that has a lower load share. Therefore, part of the requests related to some content will be sent to another part of the network to distribute the load more fairly.

Today, with the equipping of vehicles and streets with wireless equipment, a new dynamic network called vehicle ad hoc network (VANET) is emerging. This network can support two types of communication: communication between one vehicle and another vehicle and contact between a vehicle and fixed means of communication around the streets. This network can be used in many fields such as safety, business, entertainment, and emergency. One of the main challenges in this network is the routing process to transfer information from one node to another. The main problem is the lack of a stable infrastructure. Of course, there is a problem in the MANET network that many algorithms and protocols have been presented to solve, but due to the high dynamics in these networks, these protocols cannot be used for VANET. Therefore, today the presentation and improvement of the protocol Routing features in VANET networks have become an attractive issue for researchers interested in computer networks. One of the challenging issues in these protocols is the possibility of adapting them to the high dynamics of the in-vehicle network. Of course, efforts have been made in recent years in this field, including artificial intelligence techniques. This paper also tried to improve sending rate and reduce latency by using the reinforcement learning method to select the nodes. To evaluate the protocol, its implementation has been done with OMNeT++ simulation. According to the obtained results and compared with other protocols such as GSR, the proposed protocol performs better in intercity networks.

Signature identification plays an important role in many areas such as banking, administrative and judicial systems. For this purpose, in this paper, an automatic intelligent framework is developed by combining a deep pre-trained network with a recurrent neural network. The results of the proposed model were evaluated on several valid datasets and collected datasets. Since there was no suitable Persian signature dataset, we collected a Persian signature dataset based on US ASTM guidelines and standards, which can be very effective and profound for deep approaches. Due to the very promising results of the proposed model in comparison with recent studies and conventional methods, to evaluate the resistance of the proposed model to different noises, we added Gaussian Noise, Salt and Pepper Noise, Speckle Noise, and Local var Noise in different SNRs to the raw data. The results show that the proposed model can still be resistant to a wide range of SNRs; So at 15 dB, the accuracy of the proposed method is still above 90%.

Requirement engineering is one of the critical phases in the software systems development process. Functional Requirements (FR) and Non- Functional Requirements(NFR) are two of the fundamental requirements in software systems that are observed in the classifications of most of the researchers in this field. Conflicting and overlapping among the requirements in both intra and extra communications levels are some of the problems and challenges in the elicitation and prioritization phases. This paper defines and presents a decision strategy called requirements conflicts management strategy (RCMS). This strategy is defined to manage conflict and overlap of NFRs in the prioritization of the constraints satisfaction model for requirements prioritization, known as "CSOP + RP" model, in which the necessary constraints are also applied. RCMS is applied as a pre-processing phase by the requirement analyser and the results are delivered to the system manager in the "CSOP+RP" model. The composition of multiple components, such as conflict catalogue of NFRs and the relation table, the mapping model of NFRs to the domain systems, and precedence of weighted decision units in this strategy, leads to proper management of implicit and explicit knowledge conflicts and applying to overlap. The results show that using the proposed strategy leads to make a better decision, conflict management and overlapping control optimization. Therefore, ambiguities and influencing of NFRs and between NFRs and FRs reduces requirement ranking in the search-based prioritization approach. Furthermore, the system manager facilitated the final list of requirements prioritization.

(IoT) is one of the most important networks with many applications. In this network, the objects are capable of connecting to the network and sending information to the server and the server can control objects remotely. Nowadays temperature control by the IoT is very important and widespread and network sensors send the received temperature to the server at intervals. Temperature monitoring and surveillance systems are control systems that are created as a network based on the (IoT) by placing sensors in the desired environment. Procedures in these data collection models include assigning monitoring tasks to sensors, acquiring data transmission monitoring data, and controlling data accuracy. The Procedures in these data collection models include assigning monitoring tasks to sensors, acquiring data transmission monitoring data, and controlling data accuracy. Due to the huge growth of smart objects and their application, the need to collect and analyze sensor data has become one of the main challenges. The data set used in this paper contains records of temperature measured in a commercial building. In this paper, while comprehensively examining the methods of temperature control and monitoring in the (IoT), an attempt is made to provide a method that can perform a data aggregation related to temperature measurement based on the accuracy of the data sent by nodes in previous periods. In the proposed method of this research, the received data is stored as a Markov chain and by examining the data in the past periods, the accuracy of the current data can be obtained.

Research on network Community Detection (CD) has predominantly focused on identifying communities of densely connected nodes in undirected networks. Community structure is an integral part of a social network. Detecting such communities plays a vital role in a wide range of applications, including but not limited to cluster analysis, recommendation systems, and understanding the behavior of complex systems. Researchers have derived many algorithms from discovering the community structures of networks. Finding communities is a challenging task, and there is no single algorithm that produces the best results for all networks. Therefore, despite many elegant solutions, learning communities remain active research areas. CD is a challenging optimization problem that consists in searching for communities that belong to a network or graph under the assumption that the nodes of the same community share properties that enable the detection of new characteristics or functional relationships in the network. Many methods have been proposed to address this problem in many research fields, such as power systems, biology, sociology, and physics. Many of those optimization methods use modularity to identify the optimal network subdivision. This paper proposes a new CD approach based on Symbiotic Organisms Search (SOS) and Lévy Flight (LF). The LF distribution is used to prevent the stagnation of solutions in local minima. Extensive experiments compare the SOS-LF with other state-of-the-art algorithms on real-world social networks. Experimental results show that the SOS-LF is effective and stable.