نشریه محاسبات نرم
سال چهارم شماره 1 (پیاپی 7، بهار و تابستان 1394)

In vehicle surveillance systems, one of the appropriate methods for recognition are 3-D models. Several methods have been proposed for this purpose. Feature based methods are most significant and widely used. In this paper, is proposed an algorithm within recognition framework. Proposed algorithm is considered information of image and model edges as feature. A block descriptor has been used extract edges information to feature vector. Every feature vectors provide arrangement and layout in neighbourhood of edge point. Image and model feature vectors are compared using nearest neighbour method and measuring compliance are stored in a score matrix. Finally, the model has the most points in the image is detected as vehicle type. The experimental result is shown the proposed algorithm in terms of speed and accuracy offers better performance than the algorithms SURF and FREAK.

Agent-oriented software engineering is developing a new field of computer science in terms of agent-oriented methodologies, systematic approach to the analysis, design, implementation and maintenance of multiple offers. One of the major challenges in the agent- oriented software engineering is that in spite of numerous methodologies have been introduced in this area, there are still some gaps in the different phases of methodologies. In agent- oriented methodologies due to drawbacks in the different phases and the great impact these shortcoming have on the quality and efficiency of software projects a mechanism should be provided to introduce a hybrid methodology (IP) including INGENIAS and RICA methodologies, in this paper this mechanism has been presented. For this purpose the design process as well as encoding methodology of INGENIAS have been considered, INGENIAS good at design process but the stage of analysis is this methodology is not complete because it fails to model the system roles, on the other hand since RICA methodology is good at the stage of modeling system roles, a methodology combining the advantages of both methodologies has been provided. In which the stage of analysis and design methodology of INGENIAS and the process of modeling roles of RICA have been used. In other to increase the efficiency of the proposed methodology, models of capability, planning and knowledge have been added to the existing model and to explain its various phases, a case study (housing sale system) has been used, also to assess the ability of the proposed methodology, this methodology has been evaluated based on the criteria of concepts and pragmatism.

The importance of the reliability in circuits, especially the effect of cosmic ray and the faults caused by the particles hit are becoming increasingly important as the CMOS technology progresses from sub-micrometer to nanometer scale. In this paper a static latch presented which is resistant to soft error caused by energetic particles hit to the surface of the chip and suitable for high reliability applications. The hardening method is based on deploying multiple feedback paths in the opaque mode of the latch. The HSPICE post-layout simulation results in 65nm CMOS technology reveal that the proposed structures besides to single and multiple node resilience and reduced error-rate, introduce more than 13 percent reduction in delay and power consumption compared to similar structures.

New generation of wireless networks, LTE and WiMAX, supports many services which consume a lot of resources (such as VOIP, Video Conference, Digital Video, Multimedia streams and online Multi-player Games). Supporting multi-media services in wireless communication systems provide new resource allocation challenges. Because of high loads in downlink, efficient resource allocation is vital in downlink rather than uplink. In LTE networks, different services need different Quality of Services (QoS). The aim of this article is evaluation of scheduling policies for real time and non-real time services in terms of three parameters. These parameters are throughput, fairness, and packet loss rate. The evaluations and comparisons of these parameters are done by simulation of the scheduling algorithms for a video traffic as a real time service and a web browsing traffic as a non-real time service.

Aspect-oriented engineering has been introduced as a new development approach based on modularization in order to increase the maintainability and reusability of software. Aspect-oriented software engineering uses the concept of aspect to modularize crosscutting concerns. Today, this approach is considered in the both research and industrial environments. Different methods has been proposed for implementing the various phases of software development based on aspect orientation. Each of these methods accompanied with privileges, restrictions, and a variety of functionality. With respect to the diversity of these methods, a comprehensive, complete, and systematic literature review needs to be proposed. The purpose of this paper is to review the aspect-oriented development methods and determine their comparative differences in order to obtain the major challenges in this field. To do so, these methods are classified based on the software development phases. The strengths and limitations of them are also described. The results make it possible to select the most appropriate method in each phase of software development for engineers and researchers in this field.

In disease gene identification and classification, users are only interested in classifying one specific class, disease genes, without considering other classes (non-disease genes). This situation is referred to as one-class classification. Existing machine learning-based methods typically use known disease gene as positive training set and unknown genes as negative training set to build a classifier. Since there are not any non-disease gene set, in this paper we apply OCSVM (one-class support vector machines) method for one-class classification of genes to identify disease genes. Our experimental results show the superiority of our proposed method in terms of better precision, recall, and F1-measures than existing methods.