مجله رایانش نرم و فناوری اطلاعات
سال سوم شماره 2 (تابستان 1393)

Farid-Hranilovic (FH), in an interesting way, found a capacity-achieving discrete input distribution for free space optical (FSO) channel by numerically maximizing the inputparameter (β) dependent mutual information between channel input and the scaled output. In this paper, first, by using a simple mathematical inequality, we find an upper bound for FH input-scaled output mutual information and then maximize the obtained upper bound to reach to a third order equation for the optimum β as β*. Our equation (i) determines β* exactly in contrary to the FH work where β* is found numerically through an exhaustive search and also, (ii) is consistent with the estimated equation for β* in the FH work. Our upper bound is shown to be tighter than the proposed upper bound in the FH work that is found through sphere packing argument at very high SNRs. Using numerical illustrations at different SNRs, we compare our β*s, mass point spacing as ℓ*, and upper bound with previous works.

Multi-path routing is an important technique for reliable data forwarding in prone to failure wireless sensor networks (WSNs), which it leads to consume more energy. In this paper, we propose a new routing mechanism that combines multi-path routing with network coding (NCR). This combination decreases the number of required path and the total times of transmission in WSNs. In the proposed algorithm the number of control messages which exchange between nodes for route discovering, has been reduced. Simulation results show that NCR is an energy-efficient technique that improves the parameters of similar multi-path routing protocols.

In this paper, a new hierarchical fuzzy rule-based classification system based on evolutionary boosting algorithm is proposed. The rules of the proposed system are created in different levels hierarchically in a way that their membership functions are in different sizes based on the level. The flexible linguistic value definition helps the proposed classifier system to generate coarse and fine fuzzy subspaces simultaneously, which causes that the problem space to be covered in a proper manner. Each hierarchical fuzzy rule of the proposed system is formed by a running genetic algorithm, in which each chromosome represents a hierarchical fuzzy rule. After running the genetic algorithm, the best chromosome is chosen as a weak hypothesis for boosting algorithm. This process is repeated for other rules until all of the needed rules are learned iteratively. The performance of hierarchical fuzzy rules generated by evolutionary boosting algorithms is evaluated by comparing the performance of the proposed algorithm and other classification methods,especially Adaboost approximate classifier and hierarchical fuzzy rule based classification systems on a set of benchmark classification tasks. Experimental results show that the proposed algorithm accomplishes high-quality results in comparison with these classification algorithms.

This paper proposes the application of novel natural based algorithm called water cycle algorithm (WCA) on economic load dispatch (ELD) problem with multiple fuel types. In practical operation of power systems, the fuel cost function characteristics of generating units which are supplied with multiple fuel sources, have piecewise quadratic shapes whichmakes the problem of finding the global optimum more difficult when using any mathematical approaches. The proposed algorithm is based T is based on how the streams and rivers flow downhill toward the sea and change back and has been applied on a 10 unit system with multiple fuel options as two case studies, considering and neglecting valve point loading effect and also with various load demand values. This 10 unit system has also been duplicated to challenge the algorithm with large scale 30, 60 and 100 unit case studies. The results demonstrate the excellent convergence characteristics of the proposed method.

The aim of designing a controller for a teleoperation system is to achieve robust stability and optimal performance in presence of time delay in communication channel and modeling errors. Transparency is used as an index to evaluate the performance of the teleoperation system. High performance means that human operators can feel high accuracy reaction force from environments. This paper proposes an evolutionary approach to tune the parameters of a mixed H2/H∞ fixed structure controller via Memetic Particle Swarm Optimization (Memetic-PSO). According to our knowledge, this is the first research to apply mixed H2/H∞ problem via Memetic-PSO algorithm, especially in control of delayed SISO systems. The main advantageous of the proposed method is to find a suitable controller that minimizes the performance index of error signal under the robust stability and robust performance conditions. In Memetic-PSO, a local PSO model is used to disperse the particles in two different sub-regions. An adaptive local search (LS), which applies two different LS operators, is employed to refine the quality of particles. Simulation results demonstrate the effectiveness of the proposed technique.

We present a new next generation domain of Medical Meta Search Engine called MMSE. MMSE is a medical meta search engine designed for the users with no medical expertise. It is enhanced with the domain knowledge obtained from Unified Medical Language System (UMLS) to increase the effectiveness of the searches. The power of this system is based on the ability to understand the semantics of web pages and the user queries. MMSE transforms a keyword search into a conceptual search. In this system, medical concepts are extracted from user query by UMLS web service, at first. Then the expanded query is sent to Yahoo, Bing and Gigablast search engines. The relevant results of user query are shown as ranked and clustered results. To evaluate our proposed system, the experiments were divided in three parts: expansion of user query, re-ranking results, overall performance of the system. The results of our experiments indicate that, expanding the user query with domain knowledge, such as using the synonyms and partially or contextually relevant terms from UMLS, increase dramatically the relevance of an answer set produced by MMSE and the number of retrieved web pages that are relevant to the user request.