Showing papers by "Mayank Dave published in 2012"

A Review of Various VANET Data Dissemination Protocols

Abstract: VANETs (Vehicular Ad hoc Networks) are upcoming wireless network environment for Intelligent Transportation Systems (ITS). Most VANET applications are built upon the data push communication model, where information is disseminated to a set of vehicles. The diversity of the VANET applications and their potential communication protocols needs a systematic literature survey. In this paper VANET characteristics and challenges are discussed. Application based various broadcasting data dissemination protocols are surveyed separately and their fundamental characteristics are revealed. In the end a tabular comparison of all the protocols is given.

A novel technique for digital image watermarking in spatial domain

Abstract: Information technology has eased the duplication, manipulation and distribution of digital data in recent times which has resulted in the demand for safe ownership of digital images. A very crucial concern for the content owners and distributors is copyright protection and content authentication. The solution to these problems is Digital Watermarking. The watermarking is the process of embedding a signal in to other signal robustly and invisibly at the same time, the embedded signal is called watermark and the other signal is called cover or host signal. In this paper we presents a brief overview of digital image watermarking techniques in spatial and frequency domain, proposed a watermarking algorithm in spatial domain by using the Least Significant Bit (LSB) method. Every technique has some advantages and disadvantages but the LSB method is to very good compromise between performance, robustness, computational cost and quality of embedding.

Filterbank optimization for robust ASR using GA and PSO

Abstract: Automatic speech recognition (ASR) systems follow a well established approach of pattern recognition, that is signal processing based feature extraction at front-end and likelihood evaluation of feature vectors at back-end. Mel-frequency cepstral coefficients (MFCCs) are the features widely used in state-of-the-art ASR systems, which are derived by logarithmic spectral energies of the speech signal using Mel-scale filterbank. In filterbank analysis of MFCC there is no consensus for the spacing and number of filters used in various noise conditions and applications. In this paper, we propose a novel approach to use particle swarm optimization (PSO) and genetic algorithm (GA) to optimize the parameters of MFCC filterbank such as the central and side frequencies. The experimental results show that the new front-end outperforms the conventional MFCC technique. All the investigations are conducted using two separate classifiers, HMM and MLP, for Hindi vowels recognition in typical field condition as well as in noisy environment.

Integration of multiple acoustic and language models for improved Hindi speech recognition system

Abstract: Despite the significant progress of automatic speech recognition (ASR) in the past three decades, it could not gain the level of human performance, particularly in the adverse conditions. To improve the performance of ASR, various approaches have been studied, which differ in feature extraction method, classification method, and training algorithms. Different approaches often utilize complementary information; therefore, to use their combination can be a better option. In this paper, we have proposed a novel approach to use the best characteristics of conventional, hybrid and segmental HMM by integrating them with the help of ROVER system combination technique. In the proposed framework, three different recognizers are created and combined, each having its own feature set and classification technique. For design and development of the complete system, three separate acoustic models are used with three different feature sets and two language models. Experimental result shows that word error rate (WER) can be reduced about 4% using the proposed technique as compared to conventional methods. Various modules are implemented and tested for Hindi Language ASR, in typical field conditions as well as in noisy environment.

Using host criticalities for fault tolerance in mobile agent systems

Abstract: Monitoring is a crucial factor for smooth run of distributed systems such as mobile agent based system. Various activities in such systems require monitoring such as performance analysis and tuning, scheduling strategies and fault detection. In this paper we present monitoring and fault tolerance technique for mobile agent based systems. We present mobile agent based fault prevention and detection technique where the team of mobile agents monitor each host in mobile agent based system. This research focuses on building an automatic, adaptive and predictive determining policy where critical host agents are identified in advance by monitoring agents, to avoid their failures. The novelty of proposed approach is constant collection and updating of local as well as global information of the system. This policy is determined by calculating weights; taking into account the criticality of the hosts by their monitoring agents which keep updating the weights of hosts. These weights act are used for decision making of checkpointing. These monitoring mobile agents act together to detect undesirable behaviors and also provide support for restoring the system back to normalcy. We also report on the result of reliability and performance issue of our proposed approach.

A novel technique for digital image watermarking in frequency domain

Abstract: New opportunities have been explored in social, business, entertainment and scientific fields due to the development of high speed computer networks and that of, internet in particular. Ironically, the cause for the development is apprehensive because of the use of digital formatted data. Digital media has several advantages over analog media such as high fidelity copying easy editing and high quality. Software products which hide information within digital audio, images and video files have been introduced to address these growing concerns. One of the techniques of data hiding is digital watermarking. The watermarking is the process of embedding a signal in to other signal robustly and invisibly at the same time, the embedded signal is called watermark and the other signal is called cover or host signal. In this paper we are presents a brief overview of digital image watermarking techniques in spatial and frequency domain, advantages of frequency domain over spatial domain techniques and proposed a watermarking algorithm in frequency domain by using the discrete wavelet transform.

Recent Trends in Speech Recognition Systems

Abstract: Ways of improving the accuracy and efficiency of automatic speech recognition (ASR) systems have been a long term goal of researchers to develop the natural language man machine communication interface. In widely used statistical framework of ASR, feature extraction technique is used at the frontend for speech signal parameterization, and hidden Markov model (HMM) is used at the back-end for pattern classification. This chapter reviews classical and recent approaches of Markov modeling, and also presents an empirical study of few well known methods in the context of Hindi speech recognition system. Various performance issues such as number of Gaussian mixtures, tied states, and feature reduction procedures are also analyzed for medium size vocabulary. The experimental results show that using advanced techniques of acoustic models, more than 90% accuracy can be achieved. The recent advanced models outperform the conventional methods and fit for HCI applications.

Stability Based Routing Scheme for Vehicular Networks

Abstract: The vehicles are significant entities inside a vehicular network, capable of exchanging safety and non-safety messages with the neighbouring vehicles and RSU (road-side unit). The rapidly changing connectivity, error prone wireless channels, and low vehicle density in vehicular networks can make the routing schemes perform less efficiently. It leads to two main issues: to find the best routing strategy, to adapt the routing scheme in challenging settings of vehicular networks. In this paper, we propose a scheme to construct a stable path between source and destination node by using route request and route reply packets to ensure better packet delivery ratio, lower end-to-end delays and reduced message overheads. On the basis of simulation results, we conclude that the distance parameter or the traffic model alone may not be sufficient to derive the stability factor of any link and should be incorporated with the stability factor derived from the propagation model.

Self Organized Replica Overlay Scheme for P2P Networks

Abstract: Peer-to-Peer (P2P) systems are widely used for data sharing applications in an autonomous and decentralized mode. P2P systems are suitable for large- scale distributed environments in which nodes can share resources other than data such as computing power, memory and network bandwidth. Some of important parameters that affect the performance of P2P systems are peer availability, data availability, network overhead, overlay structure, churn rate, and data access time. In this paper a self organized replica overlay scheme "Improved Hierarchical Quorum Consensus" (IHQC) for P2P systems is proposed. This scheme organizes replicas in a Self Organized Hierarchical Logical Structure (SOHLS) that has special properties. The scheme improves performance of the system by reducing search time to form read/write quorums, reducing probability of accessing stale data, improving degree of intersection among consecutive quorums and reducing network overhead. This scheme is highly fault tolerant (tolerate up to

Real Time Approach for Data Placement Using Distributed Cellular Framework Based Clustering for Large Scale Sensor Networks

Abstract: Any large scale real world Wireless Sensor Network (WSN) based application would require the sensor network to provide more features besides the energy consumption. The issues of the scalability, fault tolerance, efficient data placement as well as retrieval and real time communication are all key requirements besides energy efficiency. The existing real time approaches are not energy efficient. Moreover they are only suitable for specific types of applications and do not work well in large scale applications therefore there is a need for an effective and efficient approach which can work in any type of situation and application. In this paper, we have proposed a distributed cellular approach for our proposed real time data placement model for WSNs. It is assumed that the sensor nodes are time synchronized and aware of their locations in their deployment area. The usage of Action and Relay Stations (ARS) has been proposed for data dissemination and action in the wireless sensor network.