ACCS India

31. Designing Application Specific Irregular Topology for Network-on-Chip
Authors:	Naveen Choudhary, M.S.Gaur, V. Laxmi & Virendra Singh
Abstract:	Network-on-chip (NoC) has been proposed as a solution for the communication challenges of System-on-chip (SoC) design in nano-scale technologies. Application specific SoC design offers the opportunity for incorporating custom NoC architectures that are more suitable for a particular application, and may not conform to regular topologies. In this work we propose to generate a custom NoC that maximizes performance under the given resource constraints. This being NP-hard problem, we present a heuristic technique based on genetic algorithm for synthesis of custom NoC architectures along with requisite routing tables with the objective to improve communication load distribution.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

32. Time Synchronization for an Efficient Sensor Network System
Authors:	Anita Kanavalli, Vijay Krishan, Ridhi Hirani, Santosh Prasad, Saranya K., P Deepa Shenoy and Venugopal K R
Abstract:	Time synchronization schemes in Wireless Sensor Networks have been subjected to various security threats and attacks. In this paper we throw light on some of these attacks. Nevertheless we are more concerned with the pulse delay attack which cannot be countered using any of the cryptographic techniques. We propose an algorithm called Resync algorithm which not only detects the delay attack but also aims to rectify the compromised node and introduce it back in the network for the synchronization process. In-depth analysis has been done in terms of the rate of success achieved in detecting multiple outliers i.e. nodes under attack and the level of accuracy obtained in the offset values after running the Resync algorithm.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

33. Rotational Invariant Texture Classification of Color Images using Local Texture Patterns
Authors:	A.Suruliandi, E.M.Srinivasan & K.Ramar
Abstract:	In this paper a new approach to extend Local Texture Patterns (LTP) texture model suitable for color images is presented. In this study, to extract spatial feature of a color mage, Gray-Local Texture Patterns (GLTP) is introduced. Contrast is another important property of images. To extract contrast features, Color-Local Contrast Variance Patterns (CLCVP) is proposed. However, much important information contained in the image can be revealed by joint distributions of individual features. Hence, GLTP/CLCVP is proposed as a textural feature extraction technique for classification of color images. The performance of the proposed features is carried out with rotational invariant classification of Outex texture database images. From the experimental results it is observed that GLTP/CLCVP is yielding higher classification accuracy rate of 99.32% for color images.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

34. Integer Sequence based Discrete Gaussian and Reconfigurable Random Number Generator
Authors:	Arulalan Rajan, H S Jamadagni & Ashok Rao
Abstract:	A simple random number generation technique based on integer sequences is proposed in this paper. Random integers with Gaussian distribution were generated and compared with that of the numbers generated using the proposed technique. The mean square error between the estimated probability density function and the obtained one is very negligible and is of the order of 10-6. Using the proposed technique, one can generate anywhere between 16,000 to 80,000 random integers between 1 and 100, with Gaussian distribution. Depending on the required range, the number of random numbers generated can be varied. The technique renders itself to a very simple hardware implementation that is dynamically reconfigurable on-the-fly to generate random variables with different distribution.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

35. Parallelization of Page Rank and HITS Algorithm on CUDA Architecture
Authors:	Kumar Ishan, Mohit Gupta, Naresh Kumar, Ankush Mittal
Abstract:	Efficiency of any search engine mostly depends on how efficiently and precisely it can determine the importance and popularity of a web document. Page Rank algorithm and HITS algorithm are widely known approaches to determine the importance and popularity of web pages. Due to large number of documents available on World Wide Web, huge amount of computations are required to determine the rank of web pages making it very time consuming. Researchers have devoted much attention in parallelizing PageRank on PC Cluster, Grids, and Multi-core processors like Cell Broadband Engine to overcome this issue but with little or no success. In this paper, we discuss the issues in porting these algorithms on Compute Unified Device Architecture (CUDA) and introduce efficient parallel implementation of these algorithms on CUDA by exploiting the block structure of web, which not only cut down the computation time but also significantly reduces of the cost of hardware required.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

36. Texture Classification using Local Texture Patterns: A Fuzzy Logic Approach
Authors:	E.M. Srinivasan, A. Suruliandi & K.Ramar
Abstract:	Texture analysis plays a vital role in image processing. The prospect of texture based image analysis depends on the texture features and the texture model. This paper presents a new texture model 'Fuzzy Local Texture Patterns (FLTP)' and 'Fuzzy Pattern Spectrum (FPS)'. The local image texture is described by FLTP and the global image texture is described by FPS which is an occurrence frequency of FLTP over the entire image. The efficiency of the proposed texture model is tested with texture classification. The results show that the proposed method provides a very good and robust performance.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

37. A Comparative Study of Different Packet Scheduling Algorithms with Varied Network Service Load in IEEE 802.16 Broadband Wireless Access Systems
Authors:	Prasun Chowdhury & Iti Saha Misra
Abstract:	In this paper, we conduct a comprehensive performance study of several packet scheduling algorithms such as Strict Priority, WFQ, RR, SCFQ, WRR found in literature under different queue management type like FIFO, RED, RIO, WRED in point-to-multipoint mode of OFDMbased WiMAX networks. We focus our work in comparing quality of service (QoS) of various queue scheduling schemes to suit for different specific environment i.e. different Service Loads. Extensive simulation is performed under QualNet- 4.5 for the performance of various scheduling schemes and is observed the best suitability for a particular service load condition.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

38. A Simulation Based Comparison of Gateway Load Balancing Strategies in Integrated Internet-MANET
Authors:	Rafi-U-Zaman Khaleel-Ur-Rahman Khan M.A.Razzaq A. Venugopal Reddy
Abstract:	The interconnection of wireless mobile ad hoc network with the wired Internet is called Integrated Internet-MANET. This interconnection is facilitated through gateways which act as bridges between the heterogeneous networks. Load balancing of these gateways is a critical issue in Integrated Internet-MANET. In this paper, two gateway load balancing strategies for Integrated Internet-MANET are proposed which are based on load balanced routing protocols called WLB-AODV and Modified-AODV. The proposed strategies have been simulated using thens-2 simulator. The simulation results indicate that the strategy based on WLB-AODV performs better than the one based on Modified-AODV.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

39. Parallel Hybrid Germ Swarm Computing for Video Compression
Authors:	K. M. Bakwad1, S.S. Pattnaik1, B. S. Sohi2, S. Devi1, B.K. Panigrahi3, M. R. Lohokare
Abstract:	This paper proposes Parallel Hybrid Germ Swarm Computing (PHGSC), for real time video compression. The convergence of Bacterial Foraging Optimization (BFO) is very slow because of fixed step size and its performance heavily degraded for real time processing. In this paper, initially the authors tried to increase the speed of BFO by updating bacteria positions parallel instead of serial, which is treated as Parallel Germ Computing. Further Parallel germ computing is hybridized with GLBest Particle Swarm Optimization (GLBestPSO) to improve global performance of PGC. The PHGSC is used to reduce computational time of motion estimation in video compression. The adaptive step size with prediction, zero motion vector and Von Neumann neighborhood topology implemented in PHGSC ensures the best matching block computationally very fast. The presented PHGSC saves computational time up to 93.36 % when compared with other published methods.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

40. Energy Efficient High Available System: An Intelligent Agent Based Approach
Authors:	Ankit Kumar, R.K.Senthil Kumar, B.S.Bindhumadhava
Abstract:	For achieving high availability in agent based applications which are of mission critical nature, a replica of the real agent system can be created. However even the replica of real agent system may fail due to various reasons like node failure, failure in communication link etc which may lead to the agent loss. Another major issue with such types of systems is the wastage of energy since the replica has to be in the active mode (full power mode) all the time which is really harmful for the environment. The need for improved energy management in these type of systems has become essential for many reasons like reduced energy consumption & compliance with environmental standards. To overcome these issues, we present an intelligent agent based approach for efficient energy management in these systems and also agent loss prevention by creating a replica 'on demand' of the real agent system using an efficient election algorithm (to find the best suited system for replication) designed for dynamic networks.
Access Full Text: (Members Only) (For Non ACCS members each paper is priced at Rs.100)

FIRST PREV ( Page 4 of 4 ) NEXT LAST