Sumanth Yenduri’s research while affiliated with University of Southern Mississippi and other places

Image classification is one of the major aspects in digital image analysis of remotely sensed data. In this paper, we present the effects on classification accuracy if improved thermal data are used instead of raw thermal data. We use two methods, Artificial Neural Networks (ANN) and Maximum Likelihood Approach (MLH) to demonstrate our purpose. Using each method different combinations of raw and improved data are tested to classify in order to compare the accuracies. As a final note, the findings are discussed.

In this paper, we demonstrate the use of a simplistic time based routing protocol called Simple Resilient Multi-hop Routing (SRMR) for a multi sink wireless sensor network (WSN). SRMR will allow the network to maximize efficiency by using a combination of performance metrics. For this simulation we will use the approximate distance measure based upon travel time of messages between sensor nodes. With multi sink architecture, we show efficient routing of data from the network to the closest sink based upon the selected metric. In doing so, we will be able to allow the network to become more resilient to network isolation, excess energy usage, and suboptimal throughput.

Asynchronous JAVASCRIPT and XML (AJAX)-based web applications (web apps) are different from traditional applications as they rely on stateful asynchronous communication and runtime manipulation of the DOM (document object model) tree. In this paper, we try to validate a method proposed by Mesbah et al. (2012) for testing AJAX applications automatically using two case studies. We try to identify AJAX-specific errors. Our goal is to replicate their experiment on two different web applications and comment on the performance of their approach. We use ‘Crawljax’ (designed by Mesbah et al.) to create a state-flow graph. First, we check the DOM validity for every state and, second, we check the navigation of the states. Finally, we present the results from our case studies.

In this study, we examine approaches to the problem of assembling large, contiguous sections of genetic code from short reads generated from laboratory techniques. We explore the Eulerian Path approach in detail, utilizing a de Bruijn Graph, and demonstrate current software technologies and algorithms using a sample genome. We investigate the input parameters of Velvet and discuss their implications.

In this paper we apply a Bayesian Network Methodology to a riverine model study. We show that continuous variables, following Ebert-Uphoff's method, would improve this model, and we modify an existing share-ware Bayesian Network to simulate continuous variables. We also present a new metric for the Bayesian Network.

In this paper, we address the efficient utilization of a wireless sensor network (WSN) to improve the overall data throughput and to cut down the cost of running it. Sensor nodes closest to the sink node expend more energy as they send more routing messages, which decreases their life and introduces dead spots. Having dead spots will lead to the eventual isolation of the sink nodes from the network. We attempt to solve this problem by distributing multiple sink nodes. This will enable us to utilize the network at its maximal capacity as the sensor nodes start to die off. By adding multiple sink nodes, we will switch between them and choose the one that will yield optimal network coverage after the initial sink node is no longer viable to service or cannot service the network optimally. However, at any point of time, there will be only one active sink node. We will explore in three ways. First, we use a single sink node (with no switching) until it is isolated from the network. Second, we use a combination of highest percentage of network coverage combined with the highest areas of node density on the network to derive the optimal use of the network. We call this the Cluster Algorithm for Sink Selection (CASS). Finally, we try to keep the network optimal by switching to the sink node that is farthest from the current sink (assuming to gain the maximum network reach). We will analyze and discuss our findings with respect to the data through put and life of the network.

In this paper we describe an application with large geographic data sets that was improved using Latent Semantic Analysis (LSA) in combination with word stemming. The results are consistent with other published works, and demonstrate value added skill.

The node capture attack on wireless sensor networks (WSNs) can be broken into three stages: node capture, node redeployment, and insider attacks. Existing detection techniques of the attack are mostly implemented in the second stage. Recent discovery proved the feasibility of quick and effective detection in the first stage. We propose a hello message based first stage detection scheme, which is faster, easier to execute, and more reliable than existing approaches. Two implementations FSD and SEFSD are provided. FSD is completely decentralized. SEFSD utilizes the base station to achieve energy efficiency and better security. Simulation shows the scheme outperforms existing techniques while illustrating SEFSD outperforms FSD in terms of message and energy overhead.

We have previously shown that the first stage scheme is effective in the detection of physical capture attacks in wireless sensor networks. Like most existing detection schemes, our scheme relies upon message passing. Regular message passing consumes considerable energy. In this paper the first stage scheme is integrated with a novel sleep/wakeup mechanism, which is completely pre-scheduled based upon a node's internal clock. The scheduling tolerates minor clock alignment errors, so no frequent synchronization is needed. In addition, a sink-based dissemination technique is adopted, which replaces frequent and overlapped flooding of many regular nodes with sparse sink-triggered flooding. The adoption further saves energy expense. A LDCFSD protocol has been implemented in simulation. Simulation results have confirmed above conclusion.

The node capture attack in wireless sensor networks (WSNs) can be decomposed into three stages: physically capture of node, redeployment of compromised node, and rejoin the network for various insider attacks. A well accepted belief - that the physical capture is easy to implement and that its detection is difficult - has directed majority of research effort to defense in stages three and two. The belief was recently proved false [4]. The discovery made first stage detection an attractive tactic. This paper proposes a new approach to detect the attack at the first stage. The detection is based upon the discovery of missing and malfunction of nodes due to the physical capture. The approach is simple, reliable, energy-efficient, completely local, and completely distributed. It can be used along with other approaches at stages two and three. Hence it could be employed in wide range of applications. In simulation we compare our proposal to one of latest successful research in the area. [2] Simulation results have proved that our approach is more effective and more efficient for in detection of node capture attacks.