Advanced Computing, Networking and Informatics- Volume 2

A Cellular Automata (CA) is a computing model of complex System using simple rules. In this paper the problem space is divided into number of cells and each cell can be constituted of one or several final state. Cells are affected by neighbors with the application of simple rule. Cellular Automata are highly parallel and discrete dynamical systems, whose behavior is completely specified in terms of local relation. In this paper CA is applied to solve a bridge traffic control problem. Vehicular travel which demands on the concurrent operations and parallel activities is used to control bridge traffic based on Cellular Automata technique.

Successful developments of effective real-time traffic management and information systems demand high quality real time traffic information. In the era of intelligent transportation convergence, traffic monitoring requires traffic sensory technologies. We tabulate various realistic traffic sensors which aim to address the technicalities of both point and mobile sensors and also increase the scope to prefer an optimal sensor for real time traffic data collection. The present analysis extracted data from Mobile Century experiment. The data obtained in the experiment was pre-processed successfully by applying data mining pre-processing techniques such as data transformation, normalization and integration. Finally as a result of the availability of pre-processed Global Position System (GPS) sensors trace data a road map has been generated.

In many applications of pervasive computing and communication, it is often mandatory that a certain service area be fully covered by a given deployment of nodes or access points. Hence, a fast and accurate method of estimating the coverage area is needed. However, in a scenario with a limited computation and communication capability as in self-organized mobile networks, where the nodes are not static, computation-intensive algorithms are not suitable. In this paper, we have presented a simple algorithm for estimating the area covered by a set of nodes randomly deployed over a 2-D region. We assume that the nodes are identical and each of them covers a circular area. For fast estimation of the collective coverage of

n

such circles, we approximate each real circle by the tightest square that encloses it as well as by the largest square that is inscribed within it, and present an

O

(

n

log

n

) time algorithm for computation. We study the variation of the estimated area between these two bounds, for random deployment of nodes. In comparison with an accurate digital circle based method, the proposed algorithms estimate the area coverage with only 10% deviation, while reducing the complexity of area computation significantly. Moreover, for an over-deployed network, the estimation provides an almost exact measure of the covered area.

Mobile agent is a program that can migrate autonomously from one environment to another. Many factors affect execution of mobile agents during its life cycle. Errors may occur on the server, or during communication. Error probability further increases with longer path. In mobile agent computing environment any component of the network - node, link, or agent may fail at any time, thus preventing them from continuing their execution. Therefore, fault-tolerance is a vital issue for the deployment of mobile agent systems. Here we propose a scheme to tolerate faults caused by malicious node behavior and link failure using agent cloning. The strategy is shown to prevent the agents from getting lost at irrational nodes (nodes that behave maliciously). The scheme is simulated using IBM Aglet platform and is found to be scalable when the no. of irrational nodes is fixed. Performance improves with more no. of agents.

Sink mobility achieves great success in network life time improvement in wireless sensor networks. In mobile sink movement, mobile sink moves through random path or optimized path in obstacle free area. Fixed or constrained path is used in obstacle resisting environment. In fixed or constrained path, sink move through a predefined path, which is designed by the end user. In these strategies, only fixed obstacles are avoided which are previously present in the monitoring area. These strategies are unable to avoid those obstacles which are randomly entered in the monitoring area. In this paper, we propose a quad-tree based online path detection strategy that detects any type of obstacles which enter within the network life time and design a shortest mobile sink movement path avoiding detected obstacles. The proposed scheme divides whole network into different small size region and detects region wise obstacle. On the other hand, proposed scheme constructs region wise shortest path for mobile sink data collection. Simulation results are presented to verify our proposed scheme.

Energy efficiency is one of the important issues in the Wireless Sensor Networks (WSN). In this paper, a decentralized Alive Nodes based Low Energy Adaptive Clustering Hierarchy (AL-LEACH) is presented, that considers number of alive nodes in the network to elect the cluster heads. Alive nodes are used to dynamically compute weights of random numbers. Random number is one of the important parameters to elect cluster heads for the Low Energy Adaptive Clustering Hierarchy (LEACH) protocol. Extensive simulations are carried out to compare our proposed approach AL-LEACH with Low Energy Adaptive Clustering Hierarchy (LEACH), Low energy adaptive clustering hierarchy with Deterministic Cluster-Head Selection (LDCHS) and Advanced LEACH routing protocol for wireless micro sensor networks (ALEACH). Simulation results show that AL-LEACH improves the network life time and number of packets received by Base Station (BS) through balanced energy consumption of the network.

Ad-hoc networks are vulnerable to blackhole attack. Blackhole attacker drops every incoming legitimate packet to disrupt on-demand routing as well as data delivery in ad-hoc network. The attacker drops received route request packets instead of forwarding them pretending to have valid route to destination. As a result, all data from source will be delivered towards blackhole attacker. In this paper, we have proposed a trusted on-demand routing approach to defend blackhole attacker depending on our trust model with different levels of trust computations. In our approach, blackhole attackers are identified and isolated on context of data forwarding. Simulation and analysis justify our proposal against blackhole attack for on-demand routing in Ad-hoc Network. Simulation results analyses and justifies our trusted proposal against blackhole attack for on-demand routing in Ad-hoc Network.

One of the basic requirements of Wireless Sensor Network is enhanced network lifetime. Usually Multi-hop communications between sensor nodes and the sink is more energy efficient than single hop communication. However, because the cluster heads (CHs) closer to the sink are loaded with heavy relay traffic, they exhaust much faster than other CHs in the network and ultimately the whole network becomes in operational before time. In this paper we propose a LevelWise Initial Energy Assignment (LWIEA) technique for better network lifetime.For this purpose the whole sensing region is separated in to multiple hierarchical levels and level wise initial energy of sensor nodes have been calculated before real sensor deployment. Applying LWIEA, the resultant heterogeneous sensor network with level wise different initial energy can handle different energy consumption rate in different level. Simulation results show that assigning the initial energies using LWIEA technique effectively improves the network lifetime up to 79% comparedto uniform initial energy assignment strategy with respect to the parameter Full Energy Consumption (FEC) in the network.

Wireless Sensor Networks (WSNs) are becoming integral part of today’s world due to their wide range of applications. WSNs are used in industrial applications such as factory automation and control, environmental monitoring etc. which are of low data rate but demand Quality of Service (QoS) in terms of reliability and timeliness along with energy efficiency. A WSN’s lifetime depends on the rate of consumption of energy by the sensors. One way to save energy is the judicious use of active and inactive periods in the duty cycle based Medium Access Control (MAC) protocols. The slotted IEEE 802.15.4 MAC protocol provides for it. The duty cycle mechanism however has the potential to affect the performance of the network in terms of QoS due to possible higher packet collision rate in the active periods with reduction in duty cycle. For such applications an appropriate selection of duty cycle that maintains the required QoS is vital while ensuring longevity of the network. Another important MAC parameter that has a role in the performance of WSNs based on IEEE 802.15.4/ ZigBee is the length of the Beacon Interval (BI). In this paper we present the results of our simulation experiments for determining appropriate Duty Cycle and Beacon Interval that maximizes energy efficiency while ensuring the QoS requirements of the application.

Cognitive radio networks solve the spectrum scarcity problem by dynamically utilizing the unused spectrums. To ensure secure and reliable communication, cognitive radio mobile ad hoc networks require more stringent and secure protocols due to their intrinsic nature. Tree based topology for cognitive radio network is widely used as it takes less time for join and leave operations for the users within the channel of the spectrum compared to other topologies. This paper presents a survey of tree based group key agreement schemes applicable to cognitive radio networks.

Energy saving is an important issue in wireless sensor network. Energy consumed in transmission is higher in the presence of interfering nodes due to more re-transmissions are required for a successful transmission. In this paper, interference aware power control PCTDMA for channel access is proposed. Energy consumption in the presence of interferes is investigated. Mathematical models for channel access and power control are derived. The model is simulated in MATLAB and compared with TDMA. The proposed model consumed low energy as compare to traditional TDMA.

Route Discovery in Ad-hoc Networks is an important process that is used to find the optimal path for data transmission. This route exploring process can be either proactive or reactive depends on the nature of the routing protocol. When a routing protocol initiates a route exploring process, it uses query packets to discover the optimal path. These packets are broadcasted over the network. Since they propagate in the network even after route has been found, these packets become problem of congestion in the network. So, we have proposed a cluster based approach along with Expending Ring Search (ERS) to control these route discovery packets in these networks. In our approach, we divided the whole network into clusters to achieve the scalability of the network and used modified BERS+ to reduce the retransmission of query packets in the network.

A mobile Ad Hoc Network (MANET) is an autonomous system of mobile nodes connected by a wireless link. A MANET does not have any access point or does not depend on any central administrator. In a large scale Ad Hoc Networks, there is a continuous node mobility may cause radio links to be broken frequently. This further leads to increase in packet dropping rate, end to end delay, reduction in the packet delivery rate and finally complete efficiency of network degraded. Therefore to overcome from these consequences we propose Ad hoc On Demand Route Repair Technique (AODVRRT) which introduces a new mechanism so that link failure can be predicted and accordingly perform a rapid local path repair. So by repairing the path this protocol not reduces the end-to-end latency and packet loss rate, but it also increases the efficiency of the network

.

Finally, simulation is done to measure performance.

A Wireless Sensor Network is a collection of many small sensor nodes. Every sensor node has a sensing range and a communication range. Coverage of a sensor node means the sensing region within which an event can be observed or detected. Most protocol designs for energy efficient coverage optimization maintain an adequate working node density. However, they ignore the residual energy level of the nodes. In this paper, we propose Random Backoff Sleep Protocol(RBSP) which ensures that the probability of neighbor nodes becoming active is inversely related to the residual energy level of the current active node. This will help in increasing the network lifetime by balancing energy consumption among the nodes. RBSP uses dynamic sleeping window, for the neighbor nodes, based on the amount of residual energy at an

active

node. Simulation results show that our scheme achieves more power saving and longer lifetime compared to Probing Environment and Adaptive Sleeping protocol(PEAS).

Faults are inevitable in Wireless Sensors Networks (WSNs) because of physical defects caused due to environmental hazards, imperfection or hardware and/or software related glitches. If faults are not detected and handled properly the consequences may be inexorable in case of safety critical applications. This paper presents a distributed fault diagnosis algorithm to handle both permanent and intermittent faults in WSNs. The proposed diagnosis algorithm is based on the comparison of test results and residual energy estimations by neighboring sensor nodes. The intermittent faults are handled by iterating the comparisons for

r

rounds. The basic

time-out

mechanism is adopted to handle permanently faulty sensor nodes.

In this paper, a localization technique is used to reduce the number of active sensors at an instant of time in a region for wireless sensor networks(WSNs). The active nodes in a region is calculated baed on the K-hop Independent Dominating Set(K-IDS). The proposed algorithm is evaluated using Castalia Simulator.The simulation results are analysed using network parameters namely,energy consumption,number of active nodes in a cluster and packet delivery ratio. The simulation results indicates that the proposed algorithm outperforms LEACH and MOCA protocls.

Orthogonal frequency division multiplexing (OFDM) is a promising technique for fourth generation (4G) broadband wireless communication systems. However its performance is degrade due to Doppler frequency drift or frequency drift between transmitter and receiver oscillator which causes frequency offset. This leads to a loss in the orthogonality between sub-carriers and results in inter-carrier-interference (ICI). In this paper we proposed a novel ICI self-cancellation scheme for ICI mitigation in OFDM system and compared it with the standard OFDM system and conventional self-cancellation scheme in terms of carrier-to-interference ratio (CIR) and bit-error rate. This scheme works in two very simple steps. At the transmitter side, one data symbol is modulated onto four sub-carriers with appropriate weighting coefficients. At the receiver side by linearly combining the received signals on these sub-carriers leads to a sufficient reduction in ICI. The simulation result shows that the proposed scheme outperforms the existing method.

An ultra-high bit-rate fiber optic hybrid system is proposed in this work. The proposed system utilizes the advantages of both wavelength division and time division multiplexing techniques. The performance has been investigated for various advanced data formats such as carrier-suppressed return-to-zero (CSRZ), duo-binary return-to-zero (DRZ) and modified duo-binary return-to-zero (MDRZ). MDRZ format is proved to be the best among all. The proposed system performance has been investigated at various bit rate of 10Gbps, 40Gbps and 100Gbps, for over the selected transmission lengths of 1650 Km, 400 Km, 100 Km respectively.

Peer-to-Peer based Video-on-Demand (P2P-VoD) applications are becoming very popular for scalable video distribution in both academic and commercial environments. Asynchronous arrival of peers who wish to watch videos from beginning, as well as willing to avail interactive services like jump, rewind, fast-forward etc. in these systems makes it challenging to design and deploy them. A lot of research has already been done on the architectural design issues of P2P-VoD systems. In this paper, we present a survey on approaches which address some existing design issues like alleviating the server stress due to asynchrony, building efficient P2P-VoD applications with interactive services and handling peer churn as well as the time-varying nature of network. In addition to these issues, we also discuss the challenges in deployment of P2P-based VoD systems with additional user experience features like on-demand watching of segmented scenes, on-the-fly creation of playlists etc. over best-effort Internet.

In Robust software-based router, the ultimate work is to implement a PC, which acts like a router. Although much work has done on software based router most has focused on satisfying specific requirements. Our work focuses towards the extension of software based router with routing algorithms to satisfy the QOS-based routing issues and requirements, and satisfy system parameters. The main assumption behind this work is to end up by incorporating intelligent based routers that can decouple application QOS needs, which can adopt itself to wide range of application requirements. End to end predictability comes from having the necessary intelligence in each place. Too much intelligence has historically been a problem. The issues of fairness are also considered in the evaluation of routing algorithms. Software based router is that the router needs to provide predictable treatment of packets ,queue delay and drop characteristics, assure certain rates, loss characteristics, or delay bounds to identified classes of traffic. Three admission control algorithms are proposed for providing strong end–to–end guarantees and a good candidate for a QoS routing scheme, where delay, bandwidth and jitter constraints need to be respected.

Rapid technological advancements are leading to a continuous reduction of integrated chip sizes. An additional steady increase in the chip density is resulting in device performance improvements as well as severely complicating the fabrication process. The interconnection of all the components on a chip, known as routing, is done in two phases: global routing and detail routing. These phases impact chip performance significantly and hence researched extensively today. This paper deals with the global routing phase which is essentially a case of finding a Minimal Rectilinear Steiner Tree (MRST) by joining all the terminal nodes, known to be an NP-hard problem. There are several algorithms which return near optimal results. Recently algorithms based on Evolutionary Algorithms (such as Genetic Algorithm) and based on Swarm Intelligence (such as PSO, ACO, ABC, etc.) are being increasingly used in the domain of global routing optimization of VLSI Design. Swarm based algorithms are an emerging area in the field of optimization and this paper presents a swarm intelligence algorithm, Artificial Bee Colony(ABC) for solving the routing optimization problem. The proposed algorithm shows noteworthy improvements in reduction of the total interconnect length. The performance of this algorithm has been compared with FLUTE (Fast Look Up Table Estimation) that uses Look Up Table to handle nets with degree up to 9 and net breaking technique for nets with degree up to 100. It is used for VLSI applications in which most of the nets have a degree 30 or less than that.

In this paper we have proposed a technique to minimize the cost of call routing with assigned cell in wireless network using Ant Colony Optimization (ACO). The two components that are considered for optimizing the call routing cost with assigned cell in wireless network are paging cost and handoff cost. It is assumed that the total network is divided into some location areas which are already known. When a terminal wants to set up a connection with another terminal, it will first search for the location of that. If the destination terminal is in the same location area it resembles to paging cost only, otherwise it resembles to both paging and handoff cost. Connection between two terminals in a wireless network can be established in a number of ways via different terminals. In this paper we have applied Ant Colony Optimization technique and other different algorithm to minimize the call routing cost. A comparative assessment of the execution time has been made among different call routing algorithm.

The Mobile Ad-hoc Network (MANET) is constructed based on wireless medium and it is of self organizing manner. MANET is simple to set up and has changing topology. The mobile Ad-hoc networks are in danger of different attacks because MANET operational environment is open and dynamic. In MANET, data transfer use the different Routing protocols. Selfish (Malicious) node work is completely different as compared to normal Mobile nodes. Malicious nodes have capability to remove or modify the Routing Information. It also sends the fake Route Request to access user’s data. It is responsible for attacks on the existing normal mobile nodes and creates receiver collision, restricted transmission power, fake misbehavior etc. Malicious node carries different types of attacks on the networks so it directly or indirectly effects the routing Performance. The intention of this work is to check and analyze the Network performance in malicious environment and provide prevention for the attack. Throughput and Delay are analyzed for Denial of Service (DoS) attack and prevention scenarios.

Mobile Ad Hoc Networks (MANETs) are a self-configuring network of mobile nodes characterized by multi hops and forming a dynamic wireless topology. Efficient routing protocols are the backbone of MANETs and enable the network to support various Quality of Service (QoS) parameters. Most of the routing protocols like Ad Hoc On Demand Distance Vector (AODV) send packets via a single route. However, failure of this single route results in decline of performance of MANETs. Providing a single backup route does not solve the problem completely as the failure of the backup route again leads to lower QoS parameters. This paper proposes AODV routing protocol with nth backup route (AODV nthBR) that provides source node with more than one back up routes in case of a link failure. The proposed scheme results in better throughput, improved packet delivery fraction and lesser end to end delay.

In order to reduce the pixel expansion of visual cryptography scheme (VCS), many probabilistic schemes were proposed. Let

P

 = {

P

1

,

P

2

,

P

3

,...,

P

n

} be the set of participants. The minimal qualified set for essential VCS is given by Γ

0

={

A

:

A

 ⊆ 

P

,

P

1

 ∈ 

A

and

|

A

| = 

k

}. In this paper we propose a construction of probabilistic essential VCS for sharing plural secret images simultaneously.

In recent years there has been substantial increase in both online conducted industrial espionage and hacking, resulting in heavy losses to various organizations across the globe. According to the U.S. officials’ estimations American companies in 2009 lost $50 billion alone due to cyber- espionage. The global losses due to internet hacking is estimated to be more than $1 trillion. Several techniques and methods are being used to protect data and network but all these techniques have been proved inefficient by the black hats. Then some organizations realized the need for counter attacking the attackers, but there approach doesn’t differentiate an innocent user from an attacker. These techniques mainly focus on tracing or counter attacking the suspected attacker on the basis of the IP address retrieved. But the actual attacker may spoof his IP address and therefore some other person may be affected by the counter attack. Moreover tracing an attacker on the basis of the spoofed IP is also a very difficult task. We have proposed a new technique for a counter attack which will efficiently differentiate between an attacker and a normal user. We mainly focus on entering the users system and verify his authenticity and ultimately making the task of tracing very simple.

The horizontal IaaS federation is an emerging concept where the virtual resources are delivered to the users from the federated entities. There are a few challenges to manage the virtual resources securely in federated environment as follows. First, the virtual resources are spread over the federating entities. Second, the federation is a special case of distributed system where the resources are collectively owned by the federating participants. This paper proposes a security framework for the management of federated virtual resources by splitting all the security related modules into two parts: local and global sub-modules.

With the advent of latest data mining techniques, preserving the privacy of individual’s data became a persistent issue. Every day tremendous amount of data is being generated electronically with increasing concern of data privacy. Such data when gets disseminated among various data analysts, the privacy of individuals may be breached, as the released information may be personal and sensitive in nature. Irrespective of the type of data whether numerical, categorical, mixed, time series etc, accurate analyses of such data with privacy preservation is a pervasive task. And due to the complex nature of time series data, analyzing such kind of data without harming its privacy is an open and challenging issue. In this paper we have addressed the issue of analyzing records with preserved privacy, and the data under consideration are expressed in terms of numerical time series of equal length. We have developed a data perturbation method with wavelet representation of time series data. Our experimental results show that the proposed method is effective in preserving the trade-off between data utility and privacy of time series.

BYOD (Bring Your Own Device) is a business policy to allow employees to bring their own devices at their work. The employee uses the same device in and out of the corporate office and during outside use, it may be connected to insecure internet and critical corporate data become public or when a device is used in an insecure environment, it may get infected by big threats and they may get activated when the device is used in organization’s environment and may harm the confidential information. In both cases the organization will be in losing side because if internal data become public it may hurt business strategies and future policies. If internal organization gets infected from outside attack, it will certainly hurt the business in any way. In this paper we are suggesting some approaches which can guard against these types of threat and secure the corporate data.

Google Chrome, the most popular web browser today, allows users to extend its functionality by means of extensions available in its own store or any third party website. Users can also develop their own extensions easily and add them to their browser. Vulnerabilities in browser extensions could be exploited by malicious websites to gain access to sensitive user data or to attack another website. A browser extension can also turn malicious and attack a website or steal user data. This paper proposes a framework which can be used by users and developers to analyse Chrome extensions. The technique presented here uses the permissions feature of chrome extensions and the flow of data through the extension’s JavaScript code to detect vulnerabilities in extensions and to check whether the extension could be malicious.

Intrusion Detection Systems (IDS) are becoming an essential component usually in network and data security weapon store. Since huge amount of existing off-line data and newly appearing network records that needs analysis, data mining techniques play a vital role in development of IDS. The key idea of using data mining techniques for IDS is to aim at taking benefit of classification capability of supervised learning based neural networks and clustering abilities of unsupervised learning based neural networks. In this paper, we propose an efficient intrusion detection model by amalgamating competent data mining techniques such as K-means clustering, Multilayer layer perception (MLP) neural network and support vector machine (SVM), which significantly improve the prediction of network intrusions. Since the number of clusters desired for intrusion detection problem is defined by user a priori and does not change, we employed K-means clustering technique. In the final stage, SVM classifier is used as it produces superior results for binary classification while compared to the other classifiers. We have received the best results and these are compared with results of other existing methods to prove the effectiveness of our model.

We have explored a new dimension in image steganography and propose a deft method for image– secret data – keyword (steg key) based sampling, encryption and embedding the former with a variable bit retrieval function. The keen association of the image, secret data and steg key, varied with a pixel dependant embedding results in a highly secure, reliable L.S.B. substitution. Meticulous statistical analyses have been provided to emphasize the strong immunity of the algorithm to the various steganalysis methods in the later sections of the paper.

In this paper, human visual system (HVS) characteristics are modeled using Mamdani fuzzy inference system (FIS) for robust un-compressed video watermarking technique in discrete wavelet transform (DWT) domain. The video sequence is decomposed into frames and converted into YCbCr color space. Two HVS characteristics namely edge sensitivity and contrast sensitivity are computed for each luminance component (Y) of the frame. These two computed values are fed as input to the FIS. The output of the FIS is a weighting factor which is used to embed the watermark into the frame. For embedding purpose a binary watermark is embedded into the LL3 sub-band coefficients of the video sequence. To study the robustness of proposed scheme various video processing attacks are performed. Experimental results show that proposed video watermarking scheme is highly robust and obtain good perceptual quality.

In this work we consider two protocols for performing cryptanalysis and security enhancement. The first one by Jiang et al., is a password-based authentication scheme. which does not use smart cards. We note that this scheme is an improvement over Chen et al.’s scheme shown vulnerable to the off-line dictionary attack by Jiang et al. We perform a cryptanalysis on Jiang at al.’s improved protocol and observe that it is prone to the clogging attack, a kind of denial of service (DoS) attack. We then suggest an improvement on the protocol to prevent the clogging attack.

The other protocol we consider for analysis is by Wang et al. This is a smart card based authentication protocol. We again perform the clogging (DoS) attack on this protocol via replay. We observe that all smart card based authentication protocols which precede the one by Wang et al., and require the server to compute the computationally intensive modular exponentiation are prone to the clogging attack. We suggest (another) improvement on the protocol to prevent the clogging attack, which also applies to the protocol by Jiang et. al.

This paper proposes a new image encryption algorithm that makes the use of high dimensional fractional order Chua’s chaotic system. Fractional order extension of the Chua’s system gives a much larger key-space than its original integer order version. The proposed image encryption algorithm uses a simple but excellent technique which is quite fast and the encrypted images are found to have very high entropy. The algorithm is shown to be highly robust and almost invulnerable to statistical attacks. Moreover, the algorithm is designed in such a way that it can be extended by incorporating other chaotic systems as well.

The communication in today’s scenario is mostly rely on web, it will be increases day by day means the dependency of the users for communication is increases on web browsers. So thinking about security during data communication like text and image files will be legitimate. There are several research work are in progress in this direction. In this paper we present an efficient RC4 based secure content sniffing for web browsers which supporting textual files(word, pdf, text), web files(.jsp,.php.html) and image files also. In our proposed work we send the text data and image files by applying RC4 encryption algorithm. Data is then partition in several parts for reducing the file overhead and then the data will be sending with the extra bit of 0 and 1 for identifying the attack. Means our work will secure the encryption mechanism from the traditional file including wide variety of file formats. The effectiveness of our approach is shown by the results.

The paper presents the application of the Computational Diffie-Hellman problem to ID-based signatures with pairings on elliptic curves in the random oracle model. It focusses on the security of the scheme.It also understands the fundamentals of provable security as applied in cryptography.

To allow a secure conversation among a group of members over a public network there is a need of group key agreement protocol which provide a group session key used in necessary cryptographic operations. Nowadays the protocols based on the certificateless public key cryptography (CL-PKC) creating more attraction for research because it does not require certificates to authenticates the public key as like ID- based cryptosystem and unlike ID based cryptosystem, it does not suffers from the key escrow problem. The almost all CL-PKC based group key agreement schemes in current literature are employ bilinear pairing in their operations. Since the relative computation cost of pairing is many times more than the elliptic curve point multiplication, so it motivates the researchers to propose pairing free protocols based on the CL-PKC. The present paper propose an efficient pairing free group key agreement protocol based on certificateless cryptography over elliptic curve group with their security and performance analysis. The analysis shows that the proposed protocol has strong security protection against various kinds of attack and involves comparatively lower computation and communication overheads than the other existing protocols.

Need of wireless technology increasing day to day due to the rapid development in information and communications technology. Threats and attacks are also growing in accordance with the increment in the usage of wireless communication technology. Especially Wireless Local Area Networks (WLAN) are less sensitive to the security attacks. This work combines the image processing and speech processing techniques. This work introduces Improved Bio-cryptic security aware packet scheduling (IBSPS) algorithm to enhance security levels in WLAN. To strengthen the authentication mechanism this work replaces the existing Enhanced Bio-cryptic security packet scheduling algorithm (EBSPS) with the IBSPS algorithm. In addition biometric encryption of features like finger print, Iris, Palm print and Face. IBSPS adds one more bio-cryptic security in the form of human voice as security level. Simulation were made on Matlab software and results prove that the proposed IBSPS security is stronger than the existing EBSPS and Bio-cryptic security aware packet scheduling (BSPS).

Secure authentication scheme is required to protect businesses and clients against attacks. Passwords are used as private identity for an individual. The password has to be protected from several threats like stealing, shoulder surfing, eavesdropping and guessing. Several work has been done to improve the traditional password based authentication such as biometric password authentication, graphical password scheme, and dynamic password scheme etc. Graphical passwords are strong resistance towards brute force and dictionary attacks. But it suffers to eavesdropping and guessing attacks. However, these schemes have been proved ineffective. In this paper, we have designed a hybrid system by combing the features of three different schemes such as textual password, recognition based password and recall based password. The result shows that proposed model overcomes eavesdropping and guessing more effectively than its counterparts.

The signatures are behavioral biometric characteristic used for authentication purpose. The verification of a signature while writing through the machine is called online signature verification. In this paper, we have implemented verification of signatures at sub-trajectory level. The verification has been performed using common threshold of features and writer dependent threshold. A set of fifty features are extracted of nature static, kinematic, statistical and structural properties. The experiments have been performed using SVC2004 (Signature Verification Competition) Task1 where forty user’s data include twenty genuine and twenty forgery signatures of each user. The achieved results indicate that verification at sub-trajectory level is a promising technique in online signature verification.

With the rise of communication through the Internet, there has also been a rise of interception of important messages, thus resulting in a greater risk of breach of privacy. Hence, constant research is going on for the development and improvement of techniques that can handle such attacks. Most of the existing algorithms use either the spatial domain or the frequency domain of the image for embedding the secret message. This paper introduces a steganographic algorithm that uses the frequency domain of the image for selecting the potential pixels and the spatial domain for embedding the message bits, thus making it robust against steganalytic attacks. This technique is also capable of withstanding statistical attacks. Our proposed algorithm embeds a maximum of 5 bits of the message per pixel in each image component thus making the embedding capacity very high. Besides embedding capacity, our technique also has a high embedding efficiency.

Cloud computing is an effort in delivering resources as a service. It represents a shift away from the era where products were purchased, to computing as a service that is delivered to consumers over the internet from large-scale data centers or clouds. As cloud computing is gaining popularity in the IT industry, academia appeared to be working in parallel for the rapid developments in this field. In a cloud computing environment now a days, the role of service provider is divided into two: Cloud Broker who manage cloud platforms and lease resources according to a usage-based pricing model, and service providers, who rent resources from one or many infrastructure providers to serve the end users. The aim of this research work is to deal with the scheduling of the requests on the basis of some parameters that we have identified to achieve the best optimal paths or cloud service provider allotment to the users. We have used rough set theory to generate the mathematical model. The algorithm is implemented in the cloud simulator CLOUDSIM in which cloudlets, datacenters, cloud brokers are created to perform the algorithms. Finally, we created a GUI for the user convenience so that both Cloud Service Provider and users can themselves analyze each other’s performance. We have reused some inbuilt packages of Cloudsim net beans to simulate the process.

To make cloud computing model Practical and to have essential characters like rapid elasticity, resource pooling, on demand access and measured service, two prominent technologies are required. One is internet and second important one is virtualization technology. Virtualization Technology plays major role in the success of cloud computing. A virtualization layer which provides an infrastructural support to multiple virtual machines above it by virtualizing hardware resources such as CPU, Memory, Disk and NIC is called Hypervisor. It is interesting to study how different Hypervisors perform in the Private Cloud. Hypervisors do come in Paravirtualized, Full Virtualized and Hybrid flavors. It is novel idea to compare them in the private cloud environment. This paper conducts different performance tests on three hypervisors XenServer, ESXi and KVM and explains the behavior and results of each hypervisor. In the experiment, CloudStack 4.0.2 (open source cloud computing software) is used to create a private cloud, in which management server is installed on Ubuntu 12.04 – 64 bit operating system. Hypervisors XenServer 6.0, ESXi 4.1 and KVM (Ubuntu 12.04) are installed as hosts in the respective clusters and their performances have been evaluated in detail.

Cloud computing thrives a new supplement of consumption and delivery model for internet based services and protocol. It provides large scale computing infrastructure defined on usage and also provides infrastructure services in a very flexible manner which may scales up and down according to user demand. To meet the QoS and satisfy the end users demands for resources in time is one of the main goals for cloud service provider. For this reason selecting a proper node that can complete end users task with QoS is really challenging job. Thus in Cloud distributing dynamic workload across multiple nodes in a distributed environment evenly, is called load balancing. Load balancing can be an optimization problem and should be adapting its strategy to the changing needs. This paper proposes a novel ant colony based algorithm to balance the load by searching under loaded node. Proposed load balancing strategy has been simulated using the CloudAnalyst. Experimental result for a typical sample application outperformed the traditional approaches like First Come First Serve (FCFS), local search algorithm like Stochastic Hill Climbing (SHC),another soft computing approach Genetic Algorithm (GA) and some existing Ant Colony Based strategy.

OpenStack is a cloud computing project aimed at providing infrastructure as a service (IaaS). In this paper we describe our experience in deploying OpenStack cloud over commodity hardware. We have made an effort to build a large computational facility by sharing the computational resources of our institute through the use of the OpenStack cloud platform. In this paper, we give an overview ofthe OpenStack cloud platform and various services offered by it. We describe two multi-node cloud architectures that we have implemented. In the first architecture, we have deployed the cloud over few machines connected by a closed network. The second architecture allowed us to use geographically separated nodes. We describe the steps required for installation of the cloud for eachof these architectures and provide automated scripts for the same. These automated scripts are available at the following website:http://vnit.ac.in/images/openstack/openstack_grizzly.rar.

Resource management in cloud environment poses unique challenges. Resources, in the form of

virtual machines

(VM) are to be provisioned on the fly while using the underlying infrastructure efficiently and still meeting the performance parameters. This involves collecting system resource statistics for decision making by other components of cloud environment. In this paper, the process of resource (VM) management in the cloud is mapped to

demand based

system wherein the VMs that require additional resources or need to relinquish their resources send requests to a centralized controller. Further, since resources are limited, dynamic resource allocation forms a classical optimization problem. This paper proposes a one-dimensional knapsack optimization solved using dynamic programming, to achieve efficient resource allocation. The performance of the proposed algorithm has been compared with brute force algorithm.

This article aims to propose a large-scale cloud architecture to serve for biometric system that enrols large population. In identification mode of biometric system, a query template is matched with all stored templates in the database and a match is said to occur with the one with which match-value becomes highest. Hence the identification time =

n

×

t

where

n

= database size and

t

= 1:1 matching time. As the database size

n

becomes sufficiently large, the identification time increases significantly. This leads to long response time of the system. However, achieving the

n

matching processes in parallel can bring down the total identification system from

nt

to

t

. This speeds up the proposed system

n

times than its sequential counterpart with the trade-off of the cost of resources for cloud and extra communication. The proposed architecture also takes care of threat to compromise secured data as they are passed to different nodes. This architecture passes inputs to cloud nodes hiding the identity-holder’s information so that stealing the identity data of an individual will not compromise the security of the system.

Due to advancement in mobile device technology, mobile devices are becoming an inevitable part of human lives. In context of running massive applications on mobile devices, users can’t utilize the potential of mobile devices in an efficient manner since mobile devices are constrained by processing power, memory requirements and battery capacity. To alleviate this resource scarcity problem of the mobile devices, mobile cloud computing is the most promising solution which combines the technologies from both the mobile computing and the cloud computing. The execution of heavy application on mobile devices is augmented by powerful and resource-abundant cloud servers. This is achieved by partitioning an application into tasks such that the computational intensive tasks are offloaded to cloud and after executing task on cloud, results are sent back to mobile device, referred to as computation offloading. In this paper, we have put forth a scalable, fault-tolerant framework for dynamically and optimally partition the application using our proposed genetic algorithm.

VLSI technology has made possible the integration of massive number of components into a single chip with the minimum power dissipation. But concerned by the wall that Moore’s law is expected to hit in the next decade, the integrated circuit community is turning to emerging nano-technologies for continued device improvements. Quantum dot cellular automata(QCA) is a technology which has the potential of faster speed, smaller size and minimum power consumption compared to transistor –based technology. In quantum dot cellular automata, the basic elements are simple cells. Each quantum cell contains two electrons which interact via Coulomb forces with neighboring cells. The charge distribution in each cell tends to align along one of two perpendicular axes, which allows the encoding of binary information using the state of the cell. These cells are used as building blocks to construct gates and wires. This paper utilizes these unique features of QCA to simulate symmetric functions. A general equation for the minimum number of gates required to an arbitrary number of input variables causing synthesis of symmetric function is achieved. Finally a general expression for the number of gates in benchmark circuits is also deduced. It provides significant reduction in hardware cost and switching delay compared to other existing techniques.

The audio/video stream retrieves from the storage server depends upon the cache refreshment polices. The replacement policy depends upon the efficiency of handle the cache hit ratio and cache miss ratio at real time. The cache size is limited with compare to the auxiliary memory size, and it is only the fraction of the auxiliary memory. The cache memory maintains two blocks one for the Least Recently Used (LRU) and other for the Least Frequency Used (LFU). The Least Recently Frequency used (LRFU) pages store into the cache memory. Since the size of the cache is limited by using the exponential smoothing parameter, dynamically the cache replaces the page with the smallest hit count from the LRU. The request page from the submitted request stream increment the hit counts for the already listed pages. In this paper, we present the LRFU polices and the impact of that polices for the limited cache size with a huge submitted stream of requests in a very small interval of time.

This paper represents a method of analysis of simple networks with the help of MATLAB-SIMULINK models and microcontroller based hardware. The mathematical foundation of this method is an algorithm based on orthogonal Hybrid Function.

The performance and scalability of the virtualized systems are affected by the size and speed of main memory. The memory deduplication is a prominent approach which increases memory savings by sharing of duplicate memory pages across virtual machines. Stability of shared pages is the important factor for page sharing mechanism. If sharing is short lived, it triggers CoW exception handler instantly which significantly impact the performance of memory deduplication process. The proposed approach uses

hinting

mechanism to evade instability pages from scanning and merging process, thereby enhancing the effectiveness and efficiency of memory deduplication process. The main advantage of proposed approach is that it does not need any guest OS modification and it is implemented using Kernel Samepage Merging (KSM) - a memory deduplication daemon in the linux kernel. The evaluation with several benchmark workloads shows that the proposed approach achieves a significant improvement over vanilla KSM memory deduplication process.

In this paper, we propose segregation and dispersal of

hot-zone

&

cold-zone

of a FAT filesystem over a hybrid-storage system for performance gains. Specifically, we propose

hFAT

, a high performance FAT32 filesystem design, that stores the most frequently accessed metadata of files on a solid-state storage drive while as actual contents on the magnetic drive. The idea is to eliminate the head positioning latency incurred by FAT filesystem operations while accessing metadata & userdata disk areas. After exercising the

hFAT

filesystem using

Sprite LFS small-file

benchmark, we found that

hFAT

design can reduce the latency incurred by FAT32 filesystem operations by a minimum of 25%, 10% and 90% during writing, reading and deleting a large number of small files respectively, if a solid-state storage device having latency lesser or equal to 10% of that of magnetic disk is used in addition.

Solving the Task Assignment Problem is important for many real time and computational scenarios where many small tasks need to be solved by multiple processors simultaneously. In this paper, a Heuristic and Parallel Algorithm for Task Assignment Problem is proposed. Results obtained for certain cases are presented and compared with the optimal solutions obtained by already available algorithms. It is observed that the proposed algorithm works much faster and efficient than the existing algorithms. The paper also demonstrates how the proposed algorithm could be extended to multiple distributed processors.

Instruction Level Parallelism (ILP) is not the new idea. Unfortunately ILP architecture not well suited to for all conventional high level language compilers and compiles optimization technique. Instruction Level Parallelism is the technique that allows a sequence of instructions derived from a sequential program (without rewriting) to be parallelized for its execution on multiple pipelining functional units. As a result, the performance is increased while working with current softwares. At implicit level it initiates by modifying the compiler and at explicit level it is done by exploiting the parallelism available with the hardware. To achieve high degree of instruction level parallelism, it is necessary to analyze and evaluate the technique of speculative execution control dependence analysis and to follow multiple flows of control. The researchers are continuously discovering the ways to increase parallelism by an order of magnitude beyond the current approaches. In this paper we present impact of control flow support on highly parallel architecture with 2-core and 4-core. We also investigated the scope of parallelism explicitly and implicitly. For our experiments we used trimaran simulator. The benchmarks are tested on abstract machine models created through trimaran simulator.

In this paper, we analyze the effect of interconnect on multi-core processors and have proposed a novel highly scalable on-chip interconnection mechanism for multi-core processors. As the number of cores increases, traditional on-chip interconnect like bus and crossbar proves to be low in efficiency as well as suffer from poor scalability. In order to get rid of the scalability and efficiency issues in these traditional interconnects, ring based design has been proposed. But with the steady growth in number of cores have rendered the ring interconnect too infeasible. Thus, novel interconnect designs are proposed for the future multi-core processors for enhancement in the scalability. In this paper, we analyze and compare the interconnect of two existing multi-core processors named Multi-core Processor with Internal Network(MPIN) and Mult-core processor with Ring Network(MPRN). We have also proposed a highly scalable and efficient interconnect named as

fabricated Implant in Interconnect

for multi-core processors. The placement of cores and cache in a network is proved to be highly crucial for system performance. The benchmark results are presented by using a full system simulator. Results show that, by using the proposed on-chip interconnect, compared with the MPIN and MPRN, the execution time are significantly reduced for three applications.

Data concentration is an important tool for various scientific and engineering applications. Recently, torus embedded hypercube have attracted much attention among researchers due to its inherent architectural property of two different interconnection networks. In this paper we present an algorithm to perform data concentration on torus embedded hypercube network. Our proposed algorithm takes

d

(5.5

n

+ 3 log

n

) time to perform data concentration of

d

(

d

<

N

) datum on torus embedded hypercube network having

N

(=

n

×

n

×

n

) processing elements. Our proposed algorithm can be compared with other data concentration algorithm designed for various other interconnection networks.

High level synthesis is the process of generating the register transfer level (RTL) design from the behavioural description. Time-constrained scheduling minimizes the requirement of functional units under a given time constraint and Resource-constrained scheduling minimizes the number of control steps under given resource constraint. A PB-SAT based approach which concentrates on operation scheduling, and also optimizes the number of resources and control steps is proposed here. Time-constrained and Resource-constrained based scheduling is formulated as a Pseudo-boolean satisfiability (PB-SAT) based problem and a SAT solver is used for finding the optimum schedule and minimum number of functional unit and control steps satisfying all constraints.

The present work proposes application of a set of orthogonal hybrid functions (HF) which evolved from the synthesis of orthogonal sample-and-hold functions (SHF) and orthogonal triangular functions (TF). This HF set is employed for determining the result of closed loop convolution and the result has been used for solving linear control system analysis and synthesis problems. The theory is supported by an example and the results are compared with the exact solution. It has been observed that for closed loop system identification, oscillation occurs due to numerical instability.

Cognitive Radio has been acknowledged to be the ultimate solution to meet the huge spectrum demand due to various state-of-the-art communication technologies

.

It exploits the underutilized frequency band of the legacy users for the unlicensed users opportunistically. This requires a sensible spectrum sensing technique, generally performed by binary hypotheses testing. Noise and signal plus noise distributions are important in this context. These are assumed to be Gaussian in the suboptimal energy detection technique whereas these assumptions may not be validated by practical data. In this paper, the signal plus noise distribution is approximated by four distributions, known as Kaplansky distributions that closely resemble with Gaussian distribution. Testing of hypothesis is performed by non-parametric Kolmogorov Smirnov test and power of the test is calculated for a specific false alarm probability. Numerical results are provided in support of our proposition.

As Global Positioning System (GPS) behaves irrationally due to poor satellite reception, we provide an overview of indoor position detection of objects to precisely find out positions of different objects. Here we come up with Radio Frequency Identification (RFID) as an important tool for detecting objects that are occluded from satellite visibility. We briefly describe underlying architecture of RFID technology. We show how many innovations RFID technique have made by combining with Ultrasonic sensors, Infrared sensors, Impulse-Radio Ultra wide band, and image sensors that lead to different models. We also discuss these approaches in this position determination scheme as well as provide an overview of the pros and cons of each system. We also compare many existing systems and based on the underlying drawbacks, we propose a novel system design, architecture and illustrate its usefulness.

Now-a-days the increasing burden of maintaining both the topic-specific software and the framework that houses the basic components has become uncomfortably expensive. Most of the framework for legacy software has not been migrated to modern programming languages or new computational platforms. The flexibility of using this legacy infrastructure becomes more difficult. This situation is complicated by the use of government created software tools, which may also be out of date or generally are awkward to modify/manage. These conditions have driven an unstoppable movement to COTSS-based tools. In view of the above it is proposed to develop an Electronic Warfare simulator using COTS tools like STAGE for EW Scenario Generation and VAPS XT for building Dynamic displays.

In modern Internet, different protocols generate numerous traffic types with distinct service requirements. Therefore the Internet traffic identification plays an important role to improve the network performance as part of network measurement and network management task. Primarily well-known port based method was used. But latest services uses random and uncertain port numbers reduces the accuracy of identification. Consequently “payload based approach” also known as “deep packet inspection”, used but still resulted less accuracy and required huge operational resources and are exposed to encrypted traffic flows. The recent techniques classify the application protocol based on statistical characteristics at packet level using network flow-based approach. Dealing with several datasets and millions of transaction of packets needs the use of Machine learning techniques for classification and identification of traffic. Our research shows the classification accuracy up to 99.7929%. In this paper we propose the statistical feature based approach for real-time network traffic classification. We compared the performance of three machine learning algorithms for the same. This mechanism of real time protocol identification confirms improved performance and reduced complexity.

Communities are inherent substructures present in social networks. Yet finding communities from a social network can be a difficult task. Therefore, finding communities from a social network is an interesting problem. Also, due to its use in many practical applications, it is considered to be an important problem in social network analysis and is well-studied. In this paper, we propose a maximum spanning tree based method to detect communities from a social network. Experimental results show that this method can detect communities with high accuracy and with reasonably good efficiency compared to other existing community detection techniques.

High speed and always-on network access is becoming commonplace around the world, creating a demand for increased network security. Network Intrusion Detection Systems (NIDS) attempt to detect and prevent attacks from the network using pattern-matching rules. Data compression methods are used to reduce the data storage requirement. Searching a compressed pattern in the compressed text reduces the internal storage requirement and computation resources. In this paper we implemented search process to perform compressed pattern matching in binary Huffman encoded texts. Brute-Force Search algorithm is applied comparing a single bit per clock cycle and comparing an encoded character per clock cycle. Pattern matching processes are evaluated in terms of clock cycle.

Web search is generally treated as a solitary service that operates in isolation servicing the requests of individual searchers. But in real world, searchers often collaborate to achieve their information need in a faster and efficient way. The paper attempts to harness the potential inherent in communities of like-minded searchers overcoming the limitations of conventional personalization methods. The community members can share their search experiences for the benefit of others while still maintaining their anonymity. The community based personalization is achieved by adding the benefits of reliability, efficiency and security to web search.

In the original version, the author names were captured incorrectly. It should read as:

Rupesh Jaiswal

1

and Shashikant Lokhande

2

In the original version, initially the first author name was printed as ”Pradeep Varma”. It should read as “Pradeep Varma Mudunuru”.