فهرست مطالب

Information and Communication Technology Research - Volume:7 Issue: 1, Winter 2014

International Journal Information and Communication Technology Research
Volume:7 Issue: 1, Winter 2014

  • تاریخ انتشار: 1394/01/23
  • تعداد عناوین: 7
|
  • Kourosh Dadashtabar Ahmadi, Neda Enami, Morteza Barari Page 4
    Cluster based routing are the most frequently used energy efficient routing protocols in Wireless Sensor Networks which avoid single gateway architecture through dividing of network nodes into several clusters while in each cluster, Cluster Heads work as local Base stations. However, there is several energy efficient cluster-based protocols in the literature, most of them use the topological neighborhood or adjacency as main parameter to form the clusters. This paper present a new centralized adaptive Energy Based Clustering protocol through the application of Self organizing map neural networks (called EBC-S) which can cluster sensor nodes, based on their energy level and coordinates. We apply some maximum energy nodes as weights of SOM map units; so that the nodes with higher energy attract the nearest nodes with lower energy levels. So a cluster may not necessarily contain adjacent nodes. The new algorithm enables us to form energy balanced clusters and equally distribute energy consumption on whole network space. Simulation results show the considerable profit of our proposed protocol over LEACH and LEA2C (another SOM based protocol); by increasing the network lifetime and insuring more network coverage.
    Keywords: energy based clustering, self organizing map neural networks, wireless sensor networks
  • Azadeh Omidvar, Karim Mohammadi Page 11
    Delay tolerant networks (DTN) are sparse wireless networks with intermittent connections due to limited energy, node mobility, propagation and etc. There are various real applications for DTNs such as wildlife tracking, military environment, deep space searching and etc. Traditional routing protocols fail in these networks due to intermittency. DTN protocols are based on store-carry-forward mechanism (SCF). In most of proposed methods, nodes replicate messages and give copies to nodes they encounter. This causes waste of network resources. In proposed algorithm, which is called nearest neighbor visit, for each message, source node has to find the connected neighbor which has the minimum geographic distance to destination. Next hop has to find neighbors which have recently met destination. Comparing NNV to ER and PROPHET, overhead has reduced on average by 85% compared to ER and 50% compared to PROPHET. Also, delivery ratio and delay are maintained in acceptable ranges.
    Keywords: Delay tolerant networks, DTN, message delivery delay, message delivery ratio, overhead
  • F. Mousavi Madani Page 21
    A surge of interest toward design and implementation of green networks are emerging in recent years. One obvious trend to reduce energy consumption of major active network components is to craft backbone network architecture that takes into consideration traffic grooming of low-rate IP traffic as well as power-aware virtual topology design schemes and RWA. Previous research efforts developed design strategies which assumed digital processing of incoming traffic flows at every node structure in the optical layer (DXCs). This architecture provides full wavelength conversion capability by the use of optical transponders (OEO convertors) thus reduces complicated RWA problem to a simple routing problem at the cost of sacrificing lightpath transparency. Moreover, introduction of transponders give rise to extra sources of power consumption which contrasts over goal. In this paper, energyminimized design of IP over WDM networks based on optical cross connects (OXCs) is investigated. Integer linear programming formulation as well as heuristics for two design strategies viz. multi-hop lightpath and direct bypass have been developed and their performance with regard to power consumption and relative bandwidth utilization are compared. Simulation results indicate superior performance of the proposed strategies with regard to previous studies.
    Keywords: component, Energy, minimized design, multi, hop lightpath, direct lightpath
  • Maryam Khani, Ali Ahmadi, Maryam Khademi Page 29
    Spatial-temporal coordination problem (STCP) plays a critical role in urban search and rescue (USAR) operations. Artificial Intelligence has tried to tackle this problem by taking advantage of multi-agent systems, GIS, and intelligent algorithms to enhance the task allocation by establishing collaboration between human agents and intelligent assistant agents. This paper presents a model based on cellular learning automata (CLA) to improve the teamwork interaction between human-agent teams in performing the distributed tasks. In this model, the main objective is to add the learning ability to the assistant agents in a way that they can guide human-agent toward the optimal decision(s). The effectiveness of the proposed model is evaluated on different scenarios of an earthquake simulation. Results indicate that the proposed model can significantly improve the rescue time and the maximum distance traveled by the rescue teams.
    Keywords: Spatial, Temporal Coordination, Human, agent Interaction, Multi Agent System, Cellular Learning Automata, Earthquake Emergency Response, GIS
  • Mansour Sheikhan, Mahdi Abbasnezhad Arabi, Davood Gharavian Page 41
    Artificial neural network is an efficient model in pattern recognition applications, but its performance is heavily dependent on using suitable structure and connection weights. This paper presents a hybrid heuristic method for obtaining the optimal weight set and architecture of a feedforward neural emotion classifier based on Gravitational Search Algorithm (GSA) and its binary version (BGSA), respectively. By considering various features of speech signal and concatenating them to a principal feature vector, which includes frame-based Mel frequency cepstral coefficients and energy, a rich medium-size feature set is constructed. The performance of the proposed hybrid GSA-BGSA-neural model is compared with the hybrid of Particle Swarm Optimization (PSO) algorithm and its binary version (BPSO) used for such optimizations. In addition, other models such as GSA-neural hybrid and PSO-neural hybrid are also included in the performance comparisons. Experimental results show that the GSAoptimized models can obtain better results using a lighter network structure.
    Keywords: emotion recognition, speech processing, neural network, connection optimization, structure optimization, gravitational search algorithm
  • Ahmad A. Kardan, Somayeh Modaberi, Seyede Fatemeh Noorani Page 53
    The learner model represents essential information about characteristics of learner. The Adaptive Educational Systems and Intelligent Torturing Systems use learner model to adapt required learning services according to characteristics of each learner. Hence, the accuracy of learner model is an important issue. A learner model is called “open” if its parameters could be inspected, discussed or changed by users. In this paper a novel method is proposed to improve accuracy of learner model based on learner knowledge and learner belief about his/her model. For this purpose the overlay learner modeling with Bayesian networks is used to represent learner knowledge. Then according to nature of open learner model, the learner model is presented as skill meter and learner could state his/her belief about it. Then the model is updated through proposed method. Finally the method is evaluated by use of a comprehensive test and t-student test. The results show our method improves accuracy of learner model.
    Keywords: Learner Model, Accuracy of Learner Model, Open Learner Model formatting
  • Hossein Abbasimehr, Mohammadjafar Tarokh Page 63
    Online product review websites as one of the examples of Web 2.0 websites allow users share their ideas and opinions about various products and services. Although online reviews as a user generated content can be considered as an invaluable source of information for both consumers and firms, these reviews vary greatly in term of quality and credibility. To tackle the problem of low quality reviews, we address reviewer credibility and propose a novel and feasible framework for ranking reviewers in terms of credibility. The proposed framework exploits four kinds of features including social network, profile, engagement and knowledge to quantify reviewer credibility dimensions and utilize a fuzzy inference system to calculate credibility scores of reviewers in a cognitive approach. To illustrate an application of the proposed method, we conduct an experimental study using real data gathered from Epinions. The proposed framework can support marketing departments in identifying the most credible reviewers.
    Keywords: Social Web, Online Reviews, Reviewer Credibility, Social Network, Shannon Entropy, Fuzzy Inference Systems