Abstract At present, the accuracy and robustness of unsupervised learning algorithm for multidimensional outlier detection are not good, this paper conducts a supervised learning for partial data by supervised BP neural network. Reliability of data is guaranteed by weighted entropy, then the author learns the relationship between non outliers and outliers through BP neural network, forecasts test data by trained BP neural network, finally... At present, the accuracy and robustness of unsupervised learning algorithm for multidimensional outlier detection are not good, this paper conducts a supervised learning for partial data by supervised BP neural network. Reliability of data is guaranteed by weighted entropy, then the author learns the relationship between non outliers and outliers through BP neural network, forecasts test data by trained BP neural network, finally compares the change of the data position before and after the prediction, take it as multidimensional data outlier distance, so as to complete this test. This article carries on the test experiment on the data set of multiple data sets. Compared with the traditional LOF and PSO methods, the experimental results show that the robustness and accuracy of the proposed algorithm for outlier detection of multidimensional data are better, which is suitable for all kinds of outlier detection of multidimensional data. Read More Read Less

Abstract This paper presents a continuously tunable optical band pass filter based on nested MZI designed by two arms multimode Machzehnder interferometer was proposed. The tunability of the filter is achieved by wavelength tuning of the multimode fiber. Tunability of the filter can be generated by applying Mechanical effect which is strain on the sensing arm of MM-MZI. It has been shown... This paper presents a continuously tunable optical band pass filter based on nested MZI designed by two arms multimode Machzehnder interferometer was proposed. The tunability of the filter is achieved by wavelength tuning of the multimode fiber. Tunability of the filter can be generated by applying Mechanical effect which is strain on the sensing arm of MM-MZI. It has been shown from the simulated results, it is done by Optiwave soft wear obtained that the shifted central wavelength from (1546.01) to (1546.021) nm, full width at half maximum from (279.321) to (273.681) pm, maximum gain is 22.55 dB and phase shift from(0) to (6.0808) rad and in the practical results obtained to the shifted central wavelength from(1545.762)to (1545.764)nm, FWHM from (197.529) to (283081)pm, maximum gain is 21.66 dB, and phase shift from (0) to (6.059) rad. Observed from the results, the relation between the shifted central wavelength and strain is nonlinear. Read More Read Less

Abstract Mobile applications have proliferated internet market due to abundant development of smart mobile phones and mobile devices. Testing of these mobile applications is crucial before they are made available to the customers. Testing mobile applications has remained a big challenge for software professionals because of diverse features of mobile phones as well as limitations which come with small size and portability. In this... Mobile applications have proliferated internet market due to abundant development of smart mobile phones and mobile devices. Testing of these mobile applications is crucial before they are made available to the customers. Testing mobile applications has remained a big challenge for software professionals because of diverse features of mobile phones as well as limitations which come with small size and portability. In this context testing of the application’s GUI is of utmost importance. GUI is tested for for efficiency, reliability and user friendliness. A number of testing models and techniques have been proposed to automate GUI testing to reduce testing time, but still there is no technique which can be considered as standard. This paper presents an analysis of available GUI testing techniques to expose their limitations and critical issues related with GUI testing automation, cross platform testing, GUI modeling, and automatic test cases generation and execution. To handle these issues TriTest, a model based approach is proposed which aims to automatically test functionality, compatibility, and performance of mobile apps. Read More Read Less

Abstract Wireless Sensor networks (WSNs) is collection of various sensor devices to capture the environment conditions. Node deployment, limited energy capacity, location of sensor devices, Quality of Services (QoS) and data aggregation are the critical design challenges in WSNs. To overcome these design challenges in WSNs, many techniques have proposed. Swarm Intelligence (SI) is one of the most appropriate technique to overcome the design... Wireless Sensor networks (WSNs) is collection of various sensor devices to capture the environment conditions. Node deployment, limited energy capacity, location of sensor devices, Quality of Services (QoS) and data aggregation are the critical design challenges in WSNs. To overcome these design challenges in WSNs, many techniques have proposed. Swarm Intelligence (SI) is one of the most appropriate technique to overcome the design challenges in WSNs.SI shows a current computational and behavioral similitude for taking care of disseminated issues that initially took its motivation from the biological illustrations gave by social insects like ants, termites, honey bees, wasp. Here, we will discuss many SI techniques such that Ant Colony Optimization (ACO), Elephant swarm Optimization (ESO), Hnee based optimization (HBO) and Particle Swarm Optimization (PSO) to make network energy efficient and improve the WSNs lifespan. Read More Read Less