Performance Comparison of AODV and DYMO Routing Protocols, for Congestion Detection in VANET
Vehicular Ad Hoc Networks (VANET) is a subclass of Mobile ad hoc networks which provides a distinguished approach for Intelligent Transport System (ITS). VANET’s provide communication between vehicles moving on the roads. Congestion detection and control in VANET is essential and necessary for smart ITS in order to avert an accident. Special considerations are required to develop more sophisticated techniques to avoid and detect congestion. The constrained resources of the Wireless Sensor Networks (WSN) must be considered while devising such technique to achieve the maximum throughput. In this paper we have implemented congestion detection in a collision avoidance scenario in two routing protocols, AODV and DYMO, on the basis of sent and received beacons and packets dropped while nodes communicate with each other or the Base Stations. This paper also discusses and compares the advantages/disadvantages of AODV and DYMO based on battery usage, energy used in transmit mode, throughput, message received and sent. The various ways of congestion control are further discussed.
Published by: Rajashree Dutta, Ranjana Thalore
Author: Rajashree Dutta
Paper ID: V3I3-1319
Paper Status: published
Published: May 18, 2017
Image Transfer Learning For Image Classification of Flowers Using Tensor Flow
Human Being can process images easily. It’s very easy for us to distinguish between various objects that we come across in our daily life. These tasks which are quite rudimentary for the human brain tend to be very hard problems when the computers are involved in solving the problem. The image recognition is easy to us because we have been trained to recognize the objects through their images since infancy. Machine Learning fields over the last few years have made exceptional progress in this field. But training the computer for the same thing requires a lot of time and effort. This is where Transfer Learning comes into play. We will be using transfer learning on the basic TensorFlow library to train our module on Oxford 17 VGG and Oxford 102 VGG flower data sets. We have majorly used Google’s Inception v3 model and applied it on Oxford data set to categorize flowers. This gave an overall accuracy of 94.8%.
Published by: Ujjwal Rangarh, Tanmay Trehan, Shalini .L
Author: Ujjwal Rangarh
Paper ID: V3I3-1221
Paper Status: published
Published: May 17, 2017
Performance Comparison of AODV and DYMO Routing Protocols for Boundary Detection in 3-D Wireless Sensor Networks
In order to track and detect continuous nature objects in wireless sensor networks, a large number of sensor nodes are involved. These continuous objects like biochemical diffusions, forest fires, oil spills usually spread over larger area. The nodes that sense the phenomena need to communicate with each other for exchanging the information and also send sensing information to sink, possibly by passing through many intermediate nodes. For many geometry based algorithms triangulation serves as the basis for wireless sensor networks. In this paper, we propose a distributed algorithm that produces a Delaunay triangulation for an arbitrary sensor network, for communication between the sensor nodes. The information of occurrence of an event is then passed to a control node. The communication is done both with and without using relay nodes and a comparison is made between the two methods in terms of battery, total unicast messages received, throughput and delay. The two protocols in which we have worked are AODV and DYMO.
Published by: Aditi Pandey, Ranjana Thalore
Author: Aditi Pandey
Paper ID: V3I3-1320
Paper Status: published
Published: May 17, 2017
Segmentation of Breast Images Using Gaussian Mixture Models
Breast cancer originates when cells grow uncontrollably in the breast resulting into a tumour that can be felt as a lump or observed on x-ray. The tumour is malignant that is cancerous if the cells invade surrounding tissues or spread to distant areas of the body. Breast Cancer can be observed both in women as well as men. Image processing aims at divide one picture into different types of classes or regions, recognition of objects and detecting of edges, etc that is done after the image is segmented. The main aim of this paper is to detect and separate background and foreground by using Gaussians Mixture Model, the parameters of the model and training data are learned by EM-algorithm. Pixel labelling corresponding to each pixel of true image is done by Bayesian rule. This hi labelled image is constructed during of running EM-algorithm.Numerical method of finding maximum a posterior estimation is done by using EM-algorithm and Gaussians mixture model which we called EM-MAP algorithm.
Published by: Miss. Ankita Satyendra Singh, Prof. M. M. Pawar
Author: Miss. Ankita Satyendra Singh
Paper ID: V3I3-1324
Paper Status: published
Published: May 17, 2017
Modelling and Control System Design for Electro Magnetic Suspension System
when any vehicle runs on the road, it is subjected to bump and potholes which result in vibration that affects the passenger comfort and vehicle handling. The suspension system is an assembly of tires, springs, shock absorbers and linkages that connect a vehicle to its wheels and is used to absorb and damp-out the road shocks. In designing of suspension, it is characterized by the suspension which has low spring stiffness, low damping rate which results in large suspension travel, while suspension having high damping rate is good for better handling that results in small suspension travel. These two contradictory needs the requirement of suspension with varying damping rate. This paper describes the design and simulation of the magnetic suspension system for the quarter car. Here in the designed suspension system, the damping value can be varied by changing the electric current supply in the electromagnet. Basically, this system works on Levitation technology, in which pre-defined air gap is maintained due to magnetic repulsion. Here two types of the controller are designed that is look-up table and fuzzy rules to maintain the predefined air gap. Here, when the bump comes, the air-gap fluctuation is sensed by the air sensor and this changed value of gap send to the controller and with respect to this air-gap value the current value from the look-up table or from fuzzy is decided by the controller and sends to voltage amplifier. Voltage amplifier should send that much current toward the electromagnet to vary the repulsion force which results in maintaining the predefined required air gap. Both Look-up tables and fuzzy rules are compared and validate the results.
Published by: Y. Yojana Reddy
Author: Y. Yojana Reddy
Paper ID: V3I3-1326
Paper Status: published
Published: May 17, 2017
Review on Cognitive Spectrum Sensing By Met Heuristics
cognitive radio is widely expected to be the next Big Bang in wireless communications. Spectrum sensing, that is, detecting the presence of the primary users in a licensed spectrum, is a fundamental problem for cognitive radio. As a result, spectrum sensing has reborn as a very active research area in recent years despite its long history. In this paper, spectrum sensing techniques from the optimal likelihood ratio test to energy detection matched filtering detection, cycle stationary detection, eigenvalue-based sensing, joint space-time sensing, and robust sensing methods are reviewed. Cooperative spectrum sensing with multiple receivers is also discussed. Special attention is paid to sensing methods that need little prior information on the source signal and the propagation channel. Practical challenges such as noise power uncertainty are discussed and possible solutions are provided. Theoretical analysis of the test statistic distribution and threshold setting is also investigated. Spectrum sensing problem has gained new aspects with cognitive radio and opportunistic spectrum access concepts. It is one of the most challenging issues in cognitive radio systems. In this paper, a survey of spectrum sensing methodologies for cognitive radio is presented. Various aspects of the spectrum sensing problem are studied from a cognitive radio perspective and multi-dimensional spectrum sensing concept is introduced. Challenges associated with spectrum sensing.
Published by: Rahul Kumar, Kanchan Bala Jaswal
Author: Rahul Kumar
Paper ID: V3I3-1274
Paper Status: published
Published: May 17, 2017
