International Journal of Advances in Engineering Sciences
Not a member yet
159 research outputs found
Sort by
Modeling of Software Development Effort
To accurately model the inter-relationshipbetween the software project effort/cost and theeffort/cost drivers various modeling studies have beenconducted. These mainly include empirical and softcomputing and expert analogy based techniques,primarily aimed at accurate software estimation andefficient management of software projects. This paperpresents a Simulated Annealing (SA) based optimizationapproach using Matlab to minimize the variance insoftware development (SD) effort modeling of NASA SDprojects. A comparison of results based on SA approachwith non-linear regression and other available models isalso presented proving the estimation utility of the SAapproach
Analysis of Single-Event Upsets in Phase-Locked Loops
Exploration of space provides us with usefulinformation about the universe. Understanding the number ofissues and solving them lead to progress in this field. The sun ismajor source of radiation, thus the radiation experienced by thespace-craft in the outer space is one of the issue. The electroniccircuitry on-board the space-crafts are exposed to theseradiations and may result in reduction of the lifetime of thespace-crafts and even cause failure. Thus for successful and safeexploration of space it is necessary to analyze and understand the effects of radiation on any electronics circuitry. One of the main integral parts of many electronic systems is a phase-lockedloop(PLL) which is widely used in commercial and spacedeployedelectronic system to reduce phase delay associated withdistribution of clock signal. Single event upsets (SEUs) have been a growing concern in modern integrated circuits (ICs) whereincreased susceptibilities to SEUs have been reported as devicefeature sizes decrease and operating frequencies increase [1].Thus analyzing the effects of single event transients (SET) in PLLis important for all these applications. This paper summarizeseffects of radiation on different blocks in PL
Design of an Embedded System for Health Monitoring and Emergency services using wearable sensors
Our approach in this paper is to Develop and design an embedded system which can help anyone (includes child, elders, patients and normal persons) who is in emergency or critical situation. This embedded system is wearable sensor based which can monitor the health condition such as blood pressure, body temperature, heart activity and fall detection of the person wearing and if it detects any change in the health condition immediately it sends an alert signal to the nearby rescue team members such as fire, police, ambulance, doctors and care takers of the persons family members. In this paper we analyze the technology, feasibility, cost, user interface, development tools and issues involved and propose a user friendly economical embedded system for health monitoring and emergency using wearable sensor systems
Query Optimization Strategies in Distributed Databases
The query optimization problem inlarge-scale distributed databases is NPnature and difficult to solve. The complexity of theoptimizer increases as the number of relations andnumber of joins in a query increases.being carried out to find an appropriate algorithmto seek an optimal solution especially when the sizeof the database increases. Various OptimizationStrategies have been reviewed in this paper and thestudies show that the performance of distributedquery optimization is improved when Ant ColonyOptimization Algorithm is integrated with otheroptimization algorithms
Hardware Implementation of the Huffman Encoder for Data Compression Using Altera DE2 Board
This work will aim to describe hardware implementations of static and dynamic Huffman encoders written in VHDL.The flexibility of the design allows for hardware-based implementations using FPGAs. The proposed method will have the following properties: (1) high compression for test responses is expected because Huffman coding is well-known as a minimum block coding, (2) zero-aliasing compression can be achieved.(3) It would also reduce transmission time, storage space, translation table space and encoding time
An Efficient Parallel Algorithm for Finite Exponential Function Using Modified Mesh of Trees
Exponential functions are so useful in real world situations. Three of the most common applications of exponential functions are population growth, exponential decay and compound interest. In this paper we present a parallel algorithm for finite exponential series. We map this parallel algorithm on n n Modified Mesh of Trees[11], however the number of terms in exponential series should be equal to +1. This parallel algorithm requires 10log n + O(1) steps for computing finite exponential series of +1 terms
Middleware Privacy Protection in Databases
In this paper, a solution has been developed for mitigating privacy violation in databases, as per legislatives requirements. The research work developed a modified P3P to use it as a privacy enforcement language while in parallel, the peer research community developed standard privacy enforcement languages like E-P3P and EPAL. The privacy policy enforcement layer, presented in this paper, is based on EPAL rather than the modified P3P that was originally developed in the absence of an acceptable privacy enforcement language. It further presents a modified SQL to cater to the privacy requirements for managing privacy constrained records. The privacy access layer presented demonstrates the equivalence of ECA and EPAL and therefore proposes a ECA based privacy access layer. However, since EPAL and E-P3P have evolved as a widely acceptable standard for privacy enforcement language. The Privacy Violation Detection and Monitoring layer termed as PRIVDAM is required since it is hypothesized that just as systems are breached despite implementation of information security control, similarly, privacy violations can happen despite privacy middleware being implemented. Such violation can happen due to hacking or due to compromise in the access to the privacy constrained records due to human error like password sharing. This also presents the findings of applying the solution in the health domain. The health domain was selected, as it is one of the most sensitive areas and has mature laws, such as HIPPA, governing the privacy of records in the medical domain. The overall effectiveness of the solution was found to be very good on the given data. The results showed that the solution provides excellent privacy protection, privacy violation detection with low false positives and acceptable privacy monitoring
Data mining Techniques in E - Business
Ever rising competition and changing demands of customer it is difficult to satisfy the people with different backgrounds. E- business with its various advantages of low cost, high efficiency, time saving has assumed importance these days. But it is confronted with many problems such as how to optimize the structure of website and the way of service according to the user access behavior. This paper presents an architectural framework for personalized e-business service recommendation using data mining technique
A Research Study on Mobile Internet Protocol Wireless Network
This Paper presents the way to implement a test bed for the measurement of various parameters that might affect wireless network performance and to validate results obtained from the simulator environment to that of real networking environment. This validation is of great importance to correctly predict current and future uses of the Internet technology in wireless domain. This validation will also help to make recommendations on research and standardization tasks to improve acceptance of Internet network and transport layer protocols in wireless environments
A fresh loom for Multilevel feedback Queue scheduling Algorithm
CPU scheduling is a vital phenomenon of operating system. At present, numerous CPU scheduling algorithms are existing like FCFS(First come first serve), SJF (shortest job first),SRTF( Shortest remaining time first) , Priority Scheduling,(RR)Round Robin scheduling , MLQ( multilevel queue). Efficiency and performance are not remaining satisfactory of these algorithms. MLFQ( Multilevel feedback queue) be one of most potential strategies, for CPU scheduling .It  is further extension of multi-level queue scheduling algorithm while multilevel queue scheduling is results of combination of basic scheduling algorithms such as FCFS and RR scheduling algorithm. Therefore, research on these algorithms remains continuing till today. This paper, suggested a novel approach which will improve the performance of MLFQ (CPU) scheduling algorithm