C. Chandrasekar

Combining local and global feature for object recognition using SVM-KNN

In this paper, a framework for recognizing an object from the given image based on the local and global feature is discussed. The proposed method is based on the combination of the two methods in the literature, K-Nearest Neighbor (KNN) and Support Vector Machine (SVM). For feature vector formation, Hus Moment Invariant is computed to represent the image, which is invariant to translation, rotation and scaling as a global feature and Hessian-Laplace detector and PCA-SIFT descriptor as local feature. In this framework, first the KNN is applied to find the closest neighbors to a query image and then the local SVM is applied to find the object that belongs to the object set. The proposed method is implemented as two stage process. In the first stage, KNN is utilized to compute distances of the query to all training and pick the nearest K neighbors. During the second stage SVM is applied to recognize the object. The proposed method is experimented in MATLAB and tested with the COIL-100 database and the results are shown. To prove the efficiency of the proposed method, Neural Network model (BPN) is performed and the comparative results are given.

Dynamic spectrum sharing for heterogeneous wireless network via cognitive radio

The current utilization of the spectrum is quite inefficient; consequently, if properly used, there is no shortage of the spectrum that is presently available. Therefore, it is anticipated that more flexible use of spectrum and spectrum sharing between radio systems will be key enablers to facilitate the successful implementation of future systems. Cognitive radio, however, is known as the most intelligent and promising technique in solving the problem of spectrum sharing. In this paper, we consider the technique of spectrum sharing among users of service providers to share the licensed spectrum of licensed service providers. It is shown that the proposed technique reduces the call blocking rate and improves the spectrum utilization.

Channel selection for Selective Sensing using Cognitive Radio Nodes

the proposed work, if one service provider is found to be overloaded, it utilizes the spectrum of the other service provider which is found to be underutilized. Thus, CR nodes senses the best available channel for sharing and it provides the list of available channels to the overloaded service provider to utilize the under utilized spectrum efficiently. Spectrum sharing among service providers improve the spectral efficiency, probability efficiency of sensing and reduces the call blockage. If the number of service providers are increased to share the spectrum, this would reduces the high traffic patterns of the calls.

Effect of Quality Parameters in Efficient Routing Protocol - Grid FSR with Best QoS Constraints

Rapid advanced in information technology has made it possible to transmit the data in wireless links without the aid of any fixed infrastructure or centralized administrator. Wireless mobile ad hoc networks are self-creating, selfadministering and self-organizing entities. Thus a set of selfmotivated mobile wireless users is able to dynamically exchange data among themselves, even in the absence of a predetermined infrastructure and controller. In our research work, we present investigations on the behavior of the Proactive Routing Protocol FSR in the GRID by analysis of various parameters. The Performance metrics that are used to evaluate performance of the routing protocols are Packet Delivery Ratio (PDR), Network Control Overhead, Normalized Overhead, Throughput and Average End to End Delay. Experimental results reveal that FSR is more efficient in Grid FSR in all QOS constraints.FSR can be used in all Resource critical environments. Scalability in respect to QOS is effective in FSRlarge area routing protocol. Grid Fisheye state routing (GFSR) consumes less bandwidth by restricting the propagation of routing control messages in paths formed by alternating gateways and neighbor heads, and allowing the gateways to selectively include routing table entries in their control messages.PDR and Throughput are 100% efficient in Simulation Evaluation with NS2.

A new approach to reduce flooding in Grid Fisheye state routing (GFSR) protocol by propagation neighborhood

Mobile Ad-hoc Network (MANET) is the self organizing collection of mobile nodes. Ad hoc wireless networks have massive commercial and military potential because of their mobility support. Quality of Service (QoS) routing in mobile Ad-Hoc networks is challenging due to rapid change in network topology. In this paper, we focused to reduce flooding performance of the Fisheye State Routing (FSR) protocol in Grid using ns-2 network simulator under different performance metrics scenario in respect to number of Nodes and Pause-Time. The connection establishment is costly in terms of time and resource where the network is mostly affected by connection request flooding. The proposed approach presents a way to reduce flooding in MANETs. Flooding is dictated by the propagation of connection-request packets from the source to its neighborhood nodes. The proposed architecture promotes on the concept of sharing neighborhood information. The proposed approach focuses on exposing its neighborhood peer to another node that is referred to as its friend-node, which had requested/forwarded connection request. If there is a high probability for the friend node to communicate through the exposed routes, this could improve the efficacy of bandwidth utilization by reducing flooding, as the routes have been acquired, without any broadcasts. Friendship between nodes is quantized based on empirical computations and heuristic algorithms. The nodes store the neighborhood information in their cache that is periodically verified for consistency. Simulation results show the performance of this proposed method.

A comparative analysis of probabilistic broadcasting to reduce flooding with FSR (Fisheye State Routing) protocol and grid FSR using MANET

Broadcasting in MANETs has traditionally based on flooding, which simply swamps the network with large number of rebroadcast messages in order to reach all network nodes. Although probabilistic flooding has been one of the earliest suggested schemes to broadcasting, there has not been so far any attempt to analyze its performance behavior in a MANET environment. In an effort to fill this gap, this paper investigates using extensive NS-2 simulations the effects of a number of important system parameters in a typical MANET, including node speed, pause time, and node density and packet reach ability on the performance of probabilistic flooding. The results reveal that most of these parameters have a critical impact on the reach ability and the number of saved rebroadcast messages achieved by probabilistic flooding with comparison on FSR (Fisheye State Routing) Protocol and Grid FSR in MANET.

A strategy to reduce flooding in grid Fisheye State routing GFSR protocol with weighted rough set model using MANET

A mobile ad hoc networks consist of wireless hosts that may move often. Movement of host results to changes in path. The well-known Fisheye State Routing FSR protocol determines a route when no route exists or route breaks. To establish new path from source to destination, it broadcast control packets route request packets, which increases the network bandwidth consumption and to reduce flooding. As mobile ad hoc networks have limited bandwidth, it is important to reduce the flooding. This paper provides a protocol which uses the weighted rough set model to control the route request packets in the existing FSR protocol in GRID. Weighted rough set theory is a mathematical tool to deal with vagueness, uncertainty and it also considers the importance of the objects nodes.

Energy efficient routing to reduce flooding in weighted rough set model using MANET

A mobile ad-hoc networks consists of wireless hosts that may move often, movement of host results changes in path. The well known Fisheye state routing (FSR) protocol determines a route when no route exists or route breaks. To establish new path from source to destination, it broadcast control packets (route request packets), which increases the network bandwidth, as there is a great reduce in total energy consumed (TEC) and average energy consumed (AEC) for reduced flooding than other two techniques - probabilistic broadcasting algorithm, propagating neighbourhood method. As mobile ad-hoc networks have limited bandwidth, it is important to reduce the flooding. This paper provides a protocol which uses the weighted rough set model which dominates probabilistic broadcasting algorithm and propagating neighbourhood method to control the route request packets in the existing FSR protocol in GRID. Weighted rough set theory is a mathematical tool to deal with vagueness, uncertainty and it also considers the importance of the objects (nodes).

Image segmentation using nearest neighbor classifiers based on kernel formation for medical images

Image Segmentation is one of the significant elements in the part of image processing. It becomes most essential demanding factor while typically dealing with medical image segmentation. In this paper, proposal of our work comprises of formation of kernel for the medical images by performing the deviation of mapped image data within the scope of each region from the piecewise constant model and based on the regularization term based on the function of indices value of the region. The functional objective minimization is carried out by two steps minimization in image segmentation using graph cut methods, and minimization with respect to region parameters using constant point computation. Nearest neighbor classifiers are introduced to the benchmarked image data segmented portions. Among the different methods in supervised statistical pattern recognition, the nearest neighbor rule results in achieving high performance without requirement of the prior assumptions about the distributions from which the training sets are taken.

Location-aware service discovery in next generation wireless networks

The service discovery mechanism in next generation wireless network should be flexible to both location and environment change of the user which can be achieved by appropriately predicting the user mobility. As a result, effective user mobility prediction technique need to be designed for offering the services without affecting the user location. In this paper, we propose a location aware service discovery protocol in next generation wireless networks. This technique consists of three phases: Handoff triggering based on received signal strength of the base station (BS), Client mobility prediction as per its velocity and direction, BS selection with maximum available bandwidth and residual power. By simulation results, we show that our proposed approach minimizes the query latency.