Sudhabindu Ray

A novel energy efficient chain based hierarchical routing protocol for wireless sensor networks

The nodes in a sensor network are severely constrained by energy. Reducing the energy consumption of the nodes to prolong the network lifetime is considered a critical challenge while designing a new routing protocol. In this paper we propose a new power-aware, adaptive, hierarchical and chain based protocol — CCPAR (Clustered Chain based Power Aware Routing) that utilizes the periodic assignments of the cluster head role to different nodes based on the highest residual battery capacity for ensuring the even dissipation of power by all the nodes. Transmission from a single cluster head to the base station in each round and the distribution of the data aggregation workload among all the nodes, save the cluster heads from early exhaustion. The use of data aggregation also reduces the amount of information to be transmitted to the base station. By chaining the nodes in each cluster and using a separate chain for the cluster heads, CCPAR offers the advantage of small transmit distances for most of the nodes and thus helps them to be operational for a longer period of time by conserving their limited energy. The simultaneous construction of multiple chains in different clusters reduces the time for chain construction as well as the length of each of the chains. These shorter length chains solve the problem of excessive delay in transmission for the distant nodes. Use of a fresh set of parameter values in each round provides the users the flexibility to change these values in a way to control the power consumption. The introduction of MAX threshold enables CCPAR to be quickly responsive and thus highly suitable for time critical applications. From the performance evaluation we observe that CCPAR outperforms other protocols in terms of energy saving and longevity of the network.

Analysis of QOS Parameters for DSDV and DSR in Hybrid Scenario

The area of mobile ad hoc networking has gained widespread interest of the network research community in recent years because of the ability of the MANETs to provide truly distributed and decentralized communication facility on ad hoc basis. But due to the bandwidth constrained links and limited transmission range of the mobile nodes, MANETs suffer from the inability to cover a wider network area. It is, however, possible to combine an infrastructure-less ad hoc network with a fixed one to form a hybrid network which can cover a wider area with the advantage of having less fixed infrastructure. Due to the hybrid nature of these networks, routing is considered a challenging task. Several routing protocols have been proposed and tested under various traffic conditions. However, the simulations of such routing protocols usually do not consider the hybrid network scenario. In this work we have carried out a systematic simulation based performance analysis of the two prominent routing protocols: Destination Sequenced Distance Vector Routing (DSDV) and Dynamic Source Routing (DSR) protocols in the hybrid networking environment using NS2. We have analyzed the performance differentials on the basis of three QOS parameters – packet delivery fraction, average end-to-end delay and normalized routing load under varying pause time with different number of sources.

SAR Analysis for handheld mobile phone using DICOM based voxel model

In this paper, Specific Absorption Rates (SAR) inside the human head and hand have been analyzed for a handheld mobile phone operated at GSM 900 band. Both the head and hand are modeled electrically using Digital Imaging and Communication in Medicine (DICOM) formatted CT Scan voxel data considering the electrical parameters of different internal anatomical structures. Three-dimensional Finite Difference in Time Domain (FDTD) method has been used to simulate SAR induced in the head and hand. Maximum peak 1-g and 10-g SARs of 1.33 W/kg and 0.388 W/kg are found at antenna resonance frequency of 930 MHz for 0.6 W applied input power which are below the limits set by ANSI/IEEE and FCC, respectively.

Implementation and performance analysis of the gateway discovery approaches in the integrated MANET-Internet scenario

Most of the research in the field of ad hoc network is limited to stand-alone isolated networks. In order to facilitate the users with the huge pool of resources together with the global services available from the Internet and for widening the coverage area of the MANET there is a growing need to integrate these ad hoc networks to the Internet. For this purpose we need gateways which act as bridges between these two different protocol architectures. The gateway discovery in hybrid network is considered as a critical and challenging task and with decreasing pause time and greater number of sources it becomes even more complex. Due to the scarcity of network resources in MANET, the efficient discovery of the gateway becomes a key issue in the design and development of future hybrid networks. In this paper the AODV reactive routing protocol is extended to support the communication between the MANET and the Internet. We have described the design and implementation of the various gateway discovery approaches and carried out a systematic simulation based performance evaluation of them using NS2 under different network scenarios. The performance differentials are analyzed on the basis of three metrics — packet delivery fraction, average end-to-end delay and normalized routing load.

SAR analysis in a realistic grounded human head for radiating dipole antenna

Specific Absorption Rate (SAR) averaged over 10 g, 1 g and 0.1 g mass of head tissue induced inside a realistic grounded human head in the near-field of a half-wave radiating dipole antenna has been computed using Finite Difference in Time Domain (FDTD) method. The head is modeled using voxel based anthropomorphic phantom data considering the electrical parameters of different anatomical internal structures of the human body. Results obtained from the simulation show that maximum local SAR increases significantly for smaller mass average. Maximum 0.1-g SAR of 3.75 W/kg is obtained for 4 mm distance between head and antenna with 0.6 W radiated power.

Suspended planar patch antenna for wireless energy transfer at 2.45GHz

A suspended planar patch antenna is presented for ISM band (2.4-2.485) GHz applications. This antenna can be used for wireless energy transfer as proposed in this paper. The radiating patch of the suspended patch antenna is placed over a ground plane at a height and is fed by a coaxial probe. The dielectric medium between the radiating planar patch and the ground plane is air. Suspended planar patch antenna design equations are validated by simulated and experimental results. The radiation patterns are measured in E-plane and H-plane. The analysis is performed using Method of Moments (MOM) simulation software (Zeland- IE3D version 12.0) and the experimental results are obtained from the network analyzer and radiation measurements are done in anechoic chamber. The proposed antenna provides S11 of -45dB, gain of 9.78 dBi with radiation efficiency of 99% at 2.45GHz resonant frequency. Wireless energy transfer is verified by measuring transmitted and received power using power meter with Friis transmission equation.

Improved Closed Form Large Injection Perturbation Analytical Model on the Output Spectrum of Unlocked Driven Oscillator—Part I: Phase Perturbation

This paper provides an original closed-form analytical model of frequency-pulling phenomenon in unlocked-driven nonlinear oscillators. The formulas are more general and accurate in estimating the spectral components and degree of phase perturbation of the unlocked driven oscillator under weak as well as strong injection signal level. The accuracy and superiority of the proposed analytical model in predicting injection locking of oscillators in strong injection signal level over prior literature have been established analytically and experimentally.

Trust Based Secure and Energy Efficient Clustering in Wireless Sensor Network: A Bee Mating Approach

In this paper we have proposed a trust based secure and energy efficient clustering algorithm in wireless sensor network using Honey Bee Mating Algorithm (TBCR-BMA). The proposed TBCR-BMA deprives the malicious node to act as cluster head, thus prolong the life time of the network. Moreover, we reveal that this proposed scheme outperforms the most popular hierarchical Low Energy Adaptive Clustering Hierarchy (LEACH) and Advertisement timeout driven bee’s mating approach to maintain fair energy level in sensor networks (TBCMA) in terms of average residual energy of nodes and total energy of the network.

Design and Analysis of the Gateway Discovery Approaches in MANET

The demand for any time anywhere connectivity has increased rapidly with the tremendous growth of the Internet in the past decade and due to the huge influx of highly portable devices such as laptops, PDAs etc. In order to provide the users with the huge pool of resources together with the global services available from the Internet and for widening the coverage area of the MANET, there is a growing need to integrate the ad hoc networks to the Internet. Due to the differences in the protocol architecture between MANET and Internet, we need gateways which act as bridges between them. The gateway discovery in hybrid network is considered as a critical and challenging task and with decreasing pause time and greater number of sources it becomes even more complex. Due to the scarcity of network resources in MANET, the efficient discovery of the gateway becomes a key issue in the design and development of future hybrid networks. In this paper we have described the design and implementation of the various gateway discovery approaches and carried out a systematic simulation based performance study of these approaches using NS2 under different network scenarios. The performance analysis has been done on the basis of three metrics - packet delivery fraction, average end-to-end delay and normalized routing load.

Lightweight Trust Model for Clustered WSN

Sensor network’s safety measures are built on an unrealistic trusted environment, because proposed trust models for WSNs are unsuited for resource supplies and have high computation overhead. This paper proposes a lightweight and realistic trust model for clustered WSNs (LTM). This model has been designed on the dynamic nature of actual trust building mechanism to meet the resource constraint of tiny sensor nodes. A trust metrics priority to emphasize on the important tasks of a sensor node is introduced here. A dynamic trust updating algorithm that ensures brisk drop and sluggish rise of trust is also proposed. Additionally, a self-adaptive weighted method is defined for trust aggregation, to avoid misjudgment of aggregated trust calculation. The proposed trust model also provides better resilience against vulnerabilities. We have tested the feasibility of our trust model with MATLAB.