Adwitiya Sinha

Performance evaluation of data aggregation for cluster-based wireless sensor network

In wireless sensor network, data fusion is considered an essential process for preserving sensor energy. Periodic data sampling leads to enormous collection of raw facts, the transmission of which would rapidly deplete the sensor power. In this paper, we have performed data aggregation on the basis of entropy of the sensors. The entropy is computed from the proposed local and global probability models. The models provide assistance in extracting high precision data from the sensor nodes. We have also proposed an energy efficient method for clustering the nodes in the network. Initially, sensors sensing the same category of data are placed within a distinct cluster. The remaining unclustered sensors estimate their divergence with respect to the clustered neighbors and ultimately join the least-divergent cluster. The overall performance of our proposed methods is evaluated using NS-2 simulator in terms of convergence rate, aggregation cycles, average packet drops, transmission cost and network lifetime. Finally, the simulation results establish the validity and efficiency of our approach.

A Multi-Level Strategy for Energy Efficient Data Aggregation in Wireless Sensor Networks

In this paper, we have proposed energy efficient multi-level aggregation strategy which considers data sensing as continuous stochastic process. Our proposed strategy performs filtration of sensed data by removing the redundancy in the sensed data pattern of the sensor node using Brownian motion. Further, the filtered data at the sensor node undergoes entropy-based processing prior to the transmission to cluster head. The head node performs wavelet-based truncation of the received entropy in order to select higher information bearing packets before transmitting them to the sink. Overall, our innovative approach reduces the redundant packets transmissions yet maintaining the fidelity in the aggregated data. We have also optimized the number of samples that should be buffered in an aggregation period. In addition, the power consumption analysis for individual sensors and cluster heads is performed that considers the communicational and computational cost as well. Simulation of our proposed method reveals quality performance than existing data aggregation method based on wavelet entropy and entropy based data aggregation protocols respectively. The evaluation criteria includes—cluster head survival, aggregation cycles completed during simulation, energy consumption and network lifetime. The proposed scheme reflects high potential on practical implementation by improving the life prospects of the sensor network commendably.

Stable geographic forwarding with link lifetime prediction in mobile adhoc networks for battlefield environment

Mobile adhoc network (MANET) is one of the most relevant areas of research in wireless communication that has gained prevalence due to its diversity over large-scale highly mobile networks to small-scale networks having low mobility and power constraints. Adhoc networks are inherently envisioned to be highly vulnerable to dynamic changes because of mobile sensor nodes. Node mobility often results in breakage of communication links, thereby introducing additional overhead for establishing new routes and transmitting table updates, further causing rapid exhaustion of energy reserve. Especially in military applications, preventing communication disruptions is an important security concern for defence applications engaged in safeguarding national boundaries. This necessitates the need for efficient routing strategy for battlefield environments susceptible to frequent link failures due to random mobility of groups/individuals. In this regard, we have proposed an efficient stable geographic forwarding with link-lifetime prediction (SGFL) that utilizes the broadcast nature of wireless channel and multicasts with node mobility. During the next hop selection process, a node preferably selects the neighbours which are at the least distance from the destination with low mobility. Unlike position based opportunistic routing, our scheme allows selection of backup node that lies within the transmission range of selected neighbours. Link lifetime prediction with backup nodes enhances efficiency and reliability of routing in highly mobile and congested adhoc networks. Simulation results show that our proposed SGFL achieves better performance than existing counterparts in terms of packet delivery ratio, packet loss and end-to-end delay under high node density as well as increased traffic flow.

Prediction Models for Energy Efficient Data Aggregation in Wireless Sensor Network

In sensor networks, the periodically aggregated data often exhibit high temporal coherency. Huge energy consumption incurred in transmitting these redundant information results in network disconnection thereby leading to service disruption. In order to effectively manage the energy consumption in concurrent data gathering rounds, temporal data prediction model is proposed. The proposed model provides near accurate predictions that successfully restricts redundant transmissions. The communication energy conserved owing to successful predictions helps to increase the number of data cycles considerably. In addition, an energy prediction-based cluster head rotation algorithm is also presented for load balancing within clusters. Experimental outcomes show that the proposed prediction model significantly improves energy conservation by providing successful predictions per data gathering cycle. Results reveal lower magnitude of prediction error as compared to certain existing prediction methods.

Probabilistic Data Aggregation in Information-based Clustered Sensor Network

Efficient data aggregation helps in achieving maximum performance for complex interactive and sensing applications. In our proposed work, we have considered a heterogeneous sensor network which is partitioned into clusters. Each cluster is further divided into smaller information-based groups (also called similar groups) by the local processing center (LPC). The LPC acts as a cluster head and computes the probabilistic similarity of sensors by exploiting resemblance in the pattern of their sensed data. It further schedules the active and sleep duration of sensor nodes, such that only single node from every group remains active to participate in the aggregation cycle. The LPC gathers multi-characteristic dataset from the active clustered nodes and processes them by using probabilistic aggregation model. The proposed model normalizes the multi-characteristic data for deriving relative weights of each of the characteristic attribute. Finally, the aggregated information is transmitted to the global processing center. Our protocol is evaluated with wide range of performance metrics which includes aggregation gain, information accuracy, aggregation miss ratio, network lifetime and energy consumption.

An Entropic Approach to Data Aggregation with Divergence Measure Based Clustering in Sensor Network

In wireless sensor network, data fusion is considered an essential part for preserving energy. Periodic data sampling leads to enormous collection of raw facts, the transmission of which would rapidly deplete the sensor power. In this paper, we have performed data aggregation on the basis of entropy of the sources. The entropy is computed from the local and global probability models. The models provide assistance in extracting high precision data from the sensor nodes. Further, we have proposed an energy efficient method for clustering the sensor nodes. Initially the sensors sensing same category of data are placed within a distinct cluster. The remaining unclustered sensors estimate their divergence with respect to the clustered neighbors and ultimately join the least-divergent cluster. The performance of our proposed methods is evaluated using ns-2 simulator in terms of entropy, aggregation cycles and energy utilization. The simulation results confirm the validity and efficiency of our approach.

An integrated fusion protocol for congregating sensor data in wireless sensor network

Effective fusion of data, accumulated from the sensors, can be regarded as a direct proportional factor to the successful deployment of a wireless sensor network. Two important fusion properties in support of the concerned area are: Correlation and Aggregation. This paper proposes an energy efficient data fusion protocol, which apart from employing power saving aggregation schemes, also implements network throughput enhancing routines via correlation of sensor signals. The protocol operates in dual mode to provide a perfect balance between the workload distributions among the sensor nodes during several parameter calculations. The notion of Connected Correlation Dominating Set is used to find out the clusters of active alive sensors, which actually involves in the transmission of data. On the basis of an Energy Model, the cluster heads are determined. To keep track of the error parameters, Least Squares (LS) estimation method along with the Linear Predictive Model is taken into consideration. Moreover, we simulate our algorithm using the Network Simulator (NS), ns-2.34.

Intelligent Vehicular Monitoring System Integrated with Automated Remote Proctoring

Environmental pollution is one of the crucial challenges confronted by everyone, especially in metropolitan cities. Several authorized governmental agencies and scientists are trying to develop solutions for controlling this dreadful menace. Our research aims to address the challenge by remotely tracking level of hazardous gaseous substance emitted by vehicle within a specified region under certain governmental jurisdiction. Our models assist the concerned authority to proctor, the allowable level of emission, more importantly from vehicles that are older than specified number of years. Further, in this perspective, we have designed a module which will be installed in the exhaust system of a vehicle, so as to measure the accurate level of pollution with an additional feature of displaying the reading on the display panel of the vehicle. This not only would inform the driver of present levels of pollutants, but also warn them if the levels are violated the permitted values.

Congestion control for self similar traffic in wireless sensor network

Wireless sensor networks (WSNs) have inspired many research domains in recent years. Congestion is a major issue faced by such networks, which causes heavy loss in data transmissions. Congestion is caused due to several reasons, such as heavy traffic, link failure, node failure and many more. There are various techniques developed for combatting network congestion. In this paper, we have proposed a technique for prediction of the congestion before it happens and controlling the situation before it becomes worse. Congestion in the network is controlled by adjusting the traffic rate of sources. Source nodes change their transmission rate as soon as they receive the control signal. Our algorithm is developed especially for managing congestive situations created by self similar traffic. The self-similarty in network traffic is simulated Pareto distribution. Congestion in the network is detected by analyzing the buffer ratio of nodes. Further, the simulation results show that our algorithm outperforms other existing techniques in terms of packet delivery ratio and average number of packets dropped.

Real-time analysis and visualization of online social media dynamics

Social network emanates from socially interacting entities that are represented as nodes with identical properties. These nodes are interconnected through different forms of relationship among them. Social network analysis (SNA) establishes interactions and defines the relationships on basis of theoretical foundation of networks and graphs. Social media involves in exploring several ways people share ideas and information through mobile devices, e-mail, posts, microblogging etc. Connection among users gets strengthened as they like, dislike, follow, tweet, tag, or comment over the shared idea. Various network parameters are employed to analyze the real time datasets generated from online social media such as Twitter via application programming interfaces.