Refine your search
Collections
Journals
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All
Abinaya, R.
- Improving QOS using Artificial Neural Networks in Wireless Sensor Networks
Abstract Views :151 |
PDF Views:0
Authors
R. Abinaya
1,
S. Kamakshi
1
Affiliations
1 School of Computing, SASTRA University, Thanjavur, Tamilnadu, IN
1 School of Computing, SASTRA University, Thanjavur, Tamilnadu, IN
Source
Indian Journal of Science and Technology, Vol 8, No 12 (2015), Pagination:Abstract
The service provided by the Wireless Sensor Networks (WSN) should provide better quality as Quality of the network plays an important role in the improving the performance of the system. The dynamic topology and resource constraints of WSN are highly challenging in achieving QoS. Thus in this paper Quality of Service (QoS) is improved for some of the QoS parameters like packet loss and congestion. Categorization of nodes as qualified and unqualified is done using the parameters. The quality node then dynamically forms a network. A novel method to improve the QoS based on Artificial Neural Network (ANN) is used to train the unqualified nodes to make them as quality nodes. The structure of Artificial Neural Networks provides less complexity compared to other Computational Intelligence (CI) tools as wireless sensor networks are prone to many constraints like memory and energy it reduces the computation cost and time. Our Simulation results shows that the proposed system has better performance in improving the QoS by increasing the network lifetime and reducing the packet loss ratio.Keywords
Artificial Neural Networks, Congestion Rate, Qualified Nodes, Quality of Service, Packet Loss Ratio, Unqualified Nodes, Wireless Sensor Networks- Proxy Based Buffer Overflow Attack Blocker
Abstract Views :162 |
PDF Views:5
Authors
Affiliations
1 Department of Computer Science & Engineering, Annamalai University, Chidambaram, Tamil Nadu, IN
1 Department of Computer Science & Engineering, Annamalai University, Chidambaram, Tamil Nadu, IN
Source
Programmable Device Circuits and Systems, Vol 3, No 9 (2011), Pagination: 497-501Abstract
All web servers that need to be guarded from buffer overflow attack must be registered with the proxy server, so that all user requests are directed towards proxy server, which in turn handles each request and processes the request using disassembling algorithm, if that request is found as legitimate request proxy server requests original web server else the request is blocked. This proxy server disassembles and extracts instruction sequences from a request, and then analyzes instruction sequences to find executable machine code. A machine code is a sequence of machine instructions in the form of hexadecimal executed by the machine in response to a service request. The proxy server is based on disassembling process. The BeaEngine is used for disassembling. The proxy server blocks the buffer overflow attack requests from reaching the web server.Keywords
Attack Model, Buffer Overflow, Experiments, System Design.- An Intelligent Street Light System based on Piezoelectric Sensor Networks
Abstract Views :132 |
PDF Views:0
Authors
Affiliations
1 Department of Electronics and Communication Engineering, Thiagarajar College of Engineering, Madurai –625015, Tamil Nadu, IN
1 Department of Electronics and Communication Engineering, Thiagarajar College of Engineering, Madurai –625015, Tamil Nadu, IN
Source
Indian Journal of Science and Technology, Vol 9, No 43 (2016), Pagination:Abstract
Objectives: The project aims at saving energy by detecting the vehicle movement on highways and switching on the block of street light ahead of it and simultaneously switching off the trailing lights. Methods: This project requires a sensor to detect the day and night and an array of piezoelectric sensors to detect the vehicle movements and accordingly switching on the lights ahead of it. So when there are no vehicles on the highway, all the lights remains off. Sensors used on either side of the road-senses vehicle movement and sends logic commands to microcontroller to change the intensity of the lamps accordingly. Findings: In current years, the street light control is automated by changing the resistance using the light-sensitive device. The reliability of this method is less. Though it reduces the man power, it does not conserve energy spent during night time. It cannot meet the needs of the growing street lamp information and intelligent management. The aim of smart street lights is to save energy, and to save cost. In our project we make use of the piezoelectric sensors, which is cost effective. ZigBee technology is used to communicate between the lights and internet of things to calculate parameters as well as store the information about the vehicles. Statistical analysis is also done on the data obtained from the sensors. Experiments were conducted on the road and with different types of vehicles. Sufficient data were obtained and were consolidated to come to the conclusions. All these data were processed using MATLAB Software to obtain the graphical results. The shape of the graph clearly gives us some strong conclusions. Improvements: The entire setup can be connected to the IOT (Internet of Things) for processes like data extraction, warning vehicles and is particularly useful for places like a 4-way junctions etc.Keywords
Control Systems, Intelligent, Piezoelectric Sensors, Street Light System, Sensing Network.- Wireless Communication Module to Replace Resolver Cable in Welding Robots
Abstract Views :167 |
PDF Views:0