Big data analysis has important applications in many areas such as sensor networks and connected healthcare. High volume and velocity of big data bring many challenges to data analysis. One possible solution is to summarize the data and provides a manageable data structure to hold a scalable summarization of data for efficient and effective analysis. This research extends our previous work on developing an effective technique to create, organize, access, and maintain summarization of big data and develops algorithms for Bayes classification and entropy discretization of large data sets using the multi-resolution data summarization structure. Bayes classification and data discretization play essential roles in many learning algorithms such as decision tree and nearest neighbor search. The proposed method can handle streaming data efficiently and, for entropy discretization, provide su the optimal split value.
Target tracking is a significant application of wireless sensor networks (WSNs) in which deployment of self-organizing and energy efficient algorithms is required. The tracking accuracy increases as more sensor nodes are activated around the target but more energy is consumed. Thus, in this study, we focus on limiting the number of sensors by forming an ad-hoc network that operates autonomously. This will reduce the energy consumption and prolong the sensor network lifetime. In this paper, we propose a fully distributed algorithm, an Endocrine inspired Sensor Activation Mechanism for multi target-tracking (ESAM) which reflecting the properties of real life sensor activation system based on the information circulating principle in the endocr
... Show MoreThis paper is attempt to study the nonlinear second order delay multi-value problems. We want to say that the properties of such kind of problems are the same as the properties of those with out delay just more technically involved. Our results discuss several known properties, introduce some notations and definitions. We also give an approximate solution to the coined problems using the Galerkin's method.
Magnetic levitation (Maglev) systems are employed in a wide range of applications and are therefore of significant practical importance, which has led to growing research interest. This paper presents the design of a terminal synergetic control (TSC) and feedback linearization-based proportional-integral-derivative plus second-order derivative (FL-PIDD2) controller for the Maglev system. For developing the control law of both controllers, the mathematical model of the Maglev system is converted into a canonical system where the expression of the nonlinearity is displayed in the last differential dynamic equation of the system. The determination of the TSC and FL-PIDD2 gains for achieving the desired dynamic response is carried out using the
... Show MoreThis paper develops a fuzzy multi-objective model for solving aggregate production planning problems that contain multiple products and multiple periods in uncertain environments. We seek to minimize total production cost and total labor cost. We adopted a new method that utilizes a Zimmermans approach to determine the tolerance and aspiration levels. The actual performance of an industrial company was used to prove the feasibility of the proposed model. The proposed model shows that the method is useful, generalizable, and can be applied to APP problems with other parameters.
A chemical optical fiber sensor based on surface plasmon resonance (SPR) was developed and implemented using multimode plastic optical fiber. The sensor is used to detect and measure the refractive index and concentration of various chemical materials (Urea, Ammonia, Formaldehyde and Sulfuric acid) as well as to evaluate the performance parameters such as sensitivity, signal to noise ratio, resolution and figure of merit. It was noticed that the value of the sensitivity of the optical fiber-based SPR sensor, with 60nm and 10 mm long, Aluminum(Al) and Gold (Au) metals film exposed sensing region, was 4.4 μm, while the SNR was 0.20, figure of merit was 20 and resolution 0.00045. In this work a multimode
... Show MoreAbstract
The research Compared two methods for estimating fourparametersof the compound exponential Weibull - Poisson distribution which are the maximum likelihood method and the Downhill Simplex algorithm. Depending on two data cases, the first one assumed the original data (Non-polluting), while the second one assumeddata contamination. Simulation experimentswere conducted for different sample sizes and initial values of parameters and under different levels of contamination. Downhill Simplex algorithm was found to be the best method for in the estimation of the parameters, the probability function and the reliability function of the compound distribution in cases of natural and contaminateddata.
... Show More
Data scarcity is a major challenge when training deep learning (DL) models. DL demands a large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate data to train DL frameworks. Usually, manual labeling is needed to provide labeled data, which typically involves human annotators with a vast background of knowledge. This annotation process is costly, time-consuming, and error-prone. Usually, every DL framework is fed by a significant amount of labeled data to automatically learn representations. Ultimately, a larger amount of data would generate a better DL model and its performance is also application dependent. This issue is the main barrier for