Conference Papers
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506
Browse
18 results
Search Results
Item Lossless compression of digital mammography using fixed block segmentation and pixel grouping(2008) Kumar, R.; Koliwad, S.; Dwarakish, G.S.A mammography is a specific type of imaging that uses low-dose x-ray system to examine breasts. This is an efficient means of early detection of breast cancer. High resolution is a common characteristic of such images. Archiving and retaining these data for at least three years is expensive, difficult and requires sophisticated data compression techniques. In this paper an efficient method is proposed for lossless compression of mammography images. After performing de-correlation of the image using two efficient predictors, the residue image is divided into 4x4 blocks. The blocks with all-zero pixels are identified using one bit code. Later, Second order of pixel grouping is employed to the remaining blocks to increase the coding efficiency. Such blocks are coded using Base offset method. Special techniques are used to save the header information. The method is tested using 25 mammograms from the MIAS database, each having a resolution of 1024x1024 pixels with 8 bits/pixel. Experimental results indicate better compression ratio when compared to JPEG 2000, JPEG-LS, PNG and JBIG. © 2008 IEEE.Item Survey of dynamic resource management approaches in virtualized data centers(IEEE Computer Society help@computer.org, 2013) Bane, R.R.; Annappa, B.; Shet, K.C.Virtualization technology enabled hosting of applications and services in an isolated and resource guaranteed virtual machines (VMs). Typically single physical machine (PM) runs multiple virtual machines and application resource demands are changing with time. To achieve this, dynamic resource provisioning of physical machine resources to VMs in virtualized data center is necessary. Data center requires this provisioning should be elastic so that its cost can be minimized and service level objectives (SLO) can be met by allocating exact amount of resources. It invites two main challenges: (1) determining how many resources need to be allocated to the application where resource demand is dynamic and (2) prediction of the application resource need in advance so that resource allocation could be adjusted ahead of the actual need. In this paper we have given various ways of handling above mentioned challenges for dynamic resource management and their comparisons. © 2013 IEEE.Item Predicting an optimal sparse matrix format for SpMV computation on GPU(IEEE Computer Society help@computer.org, 2014) Bayyapu, B.; Guddeti, G.R.M.; Raghavendra, P.S.Many-threaded architecture based Graphics Processing Units (GPUs) are good for general purpose computations for achieving high performance. The processor has latency hiding mechanism through which it hides the memory access time in such a way that when one warp (group of 32 threads) is computing, the other warps perform memory bound access. But for memory access bound irregular applications such as Sparse Matrix Vector Multiplication (SpMV), memory access times are high and hence improving the performance of such applications on GPU is a challenging research issue. Further, optimizing SpMV time on GPU is an important task for iterative applications like jacobi and conjugate gradient. However, there is a need to consider the overheads caused while computing SpMV on GPU. Transforming the input matrix to a desired format and communicating the data from CPU to GPU are non-trivial overheads associated with SpMV computation on GPU. If the chosen format is not suitable for the given input sparse matrix then desired performance improvements cannot be achieved. Motivated by this observation, this paper proposes a method to chose an optimal sparse matrix format, focusing on the applications where CPU to GPU communication time and pre-processing time are nontrivial. The experimental results show that the predicted format by the model matches with that of the actual high performing format when total SpMV time in terms of pre-processing time, CPU to GPU communication time and SpMV computation time on GPU, is taken into account. The model predicts an optimal format for any given input sparse matrix with a very small overhead of prediction within an application. Compared to the format to achieve high performance only on GPU, our approach is more comprehensive and valuable. This paper also proposes to use a communication and pre-processing overhead optimizing sparse matrix format to be used when these overheads are non trivial. © 2014 IEEE.Item Language modelling and english speech prediction system to aid people with stuttering disorder(Association for Computing Machinery acmhelp@acm.org, 2015) Chandana, T.L.; Kalwad, P.S.; Pattanaik, S.; Guddeti, G.This paper proposes a novel method to predict the speech based on N-Gram language model for English Language. It also concentrates on how Speech Completion can be combined with stuttering detection to aid people suffering from this disorder to overcome psychological and social introversion. To the best of our knowledge, such systems exist only in Japanese language and hence, this paper is the first to introduce such an application for English language. The existing work in Japanese language uses a vocabulary tree structure for prediction in contrast to the n-gram language model used in this paper. The basic idea of the proposed work is to consider the user's speech input for detecting the repetition of words as stuttering. If this repetition of words is detected then, the next word can be predicted after eliminating the repeated word using the n-gram language model and the predicted word can be converted back to speech. Using this proposed methodology, we are able to achieve a prediction accuracy of 87% when a 10-fold test is carried out. © 2015 ACM.Item PPRP: Predicted Position based routing protocol using Kalman Filter for Vehicular Ad-hoc Network(Association for Computing Machinery acmhelp@acm.org, 2017) Jaiswal, R.K.; Jaidhar, C.D.New edition vehicles are equipped with Global Positioning System (GPS) device which provides the vehicle position in the form of latitude and longitude, this position is used as a location id of the vehicle at time t during routing in Vehicular Ad-hoc Network (VANET). The location ids are susceptible to have an error in position due to several factors such as line-of-sight, signal fading and tunnels just for an instance. Thus, Position based routing protocol experiences poor performance. To minimize the effect of position error, this work proposes a Predicted Position Based Routing Protocol (PPRP) for VANET. PPRP predicts the vehicle location based on previous and current location using Kalman Filter (KF) to improve the Packet Delivery Ratio (PDR), average delay and throughput. Before applying KF into routing its effectiveness is verified and found satisfactory results which advocate KF, to be used in routing. The proposed routing protocol is simulated on NS-3.23 simulator. VANETMOBISIM is used to get the vehicular mobility of 25, 50, 75 and 100 vehicles running on a city road network of 1000 ∗ 1000 m2 area. The performance of the proposed routing protocol is evaluated and compared with other prediction based routing protocol. Simulation is conducted for 250m and 500m transmission range using Winner-II and Two-ray ground propagation model with IEEE 802.11p standard. © 2017 ACM.Item On predicting the frequent execution patterns in information systems(Institute of Electrical and Electronics Engineers Inc., 2017) Manoj Kumar, M.V.; Thomas, L.; Annappa, B.Process mining research discipline offers a spectrum of techniques for analysing event logs. Event logs represent the history of process execution. This information can be used for monitoring, analysing and improving the operational processes. The currently available methods in process mining emphasise on constructing the static process model. These models depict various dimensions of the process under analysis. But, models can only represent the past execution history and can't be used to guide and control the prospectus execution of the process. There is a need for the methods and techniques which guide the future execution of process in the light of recorded information. This paper introduces a technique for identifying and predicting the frequent control-flow execution patterns in information systems. The proposed Position Weight Matrix proven to be efficient during experimentation and validation studies. © 2017 IEEE.Item Comparative study of coated and uncoated tool inserts with dry machining of EN47 steel using Taguchi L9 optimization technique(American Institute of Physics Inc. subs@aip.org, 2018) Mallesha, M.; Shivananda, N.H.EN47 steel samples are machined on a self-centered lathe using Chemical Vapor Deposition of coated TiCN/Al2O3/TiN and uncoated tungsten carbide tool inserts, with nose radius 0.8mm. Results are compared with each other and optimized using statistical tool. Input (cutting) parameters that are considered in this work are feed rate (f), cutting speed (Vc), and depth of cut (ap), the optimization criteria are based on the Taguchi (L9) orthogonal array. ANOVA method is adopted to evaluate the statistical significance and also percentage contribution for each model. Multiple response characteristics namely cutting force (Fz), tool tip temperature (T) and surface roughness (Ra) are evaluated. The results discovered that coated tool insert (TiCN/Al2O3/TiN) exhibits 1.27 and 1.29 times better than the uncoated tool insert for tool tip temperature and surface roughness respectively. A slight increase in cutting force was observed for coated tools. © 2018 Author(s).Item Accurate Router Level Estimation of Network-on-Chip Architectures using Learning Algorithms(Institute of Electrical and Electronics Engineers Inc., 2019) Kumar, A.; Talawar, B.The problem of intra-communication between the Intellectual Properties(IPs) due to the rise in the amount of cores on single chips in System-on-Chip(SoC). Network-on-Chips(NoCs) has emerged as a reliable on-chip communication framework for Chip Multiprocessors and SoCs. Estimating NoC power and performance in the early stages has become crucial. We employ Machine Learning(ML) approaches to estimate architecture-level on-chip router models and performance. Experiments were carried out with distinct topology sizes with various virtual channels, injection rates, and traffic patterns. Booksim and Orion simulators are used to validate the results. Approximately 6% to 8% prediction error and a minimum speedup of 1500 × to 2000 × were shown in the framework. © 2019 IEEE.Item A Support Vector Regression-Based Approach to Predict the Performance of 2D 3D On-Chip Communication Architectures(Institute of Electrical and Electronics Engineers Inc., 2019) Nirmal Kumar, A.; Talawar, B.Recently, Networks-on-Chips (NoCs) have evolved as a scalable solution to traditional bus and point-to-point architecture. NoC design performance evaluation is largely based on simulation, which is extremely slow as the architecture size increases, and it gives little insight on how distinct design parameters impact the actual performance of the network. Simulation for optimization purposes is therefore very difficult to use. In this paper, we propose a Support Vector Regression(SVR)-based framework, which can be used to analyze the performance of 2D and 3D NoC architectures. Experiments were conducted by varying architecture sizes with different virtual channels, injection rates. The framework proposed can be used to obtain fast and accurate NoC performance estimates with a prediction error 2% to 4% and minimum speedup of 3000 × to 3500×. © 2019 IEEE.Item A Conceptual Framework for Intelligent Management of Workloads in Cloud Environment(Institute of Electrical and Electronics Engineers Inc., 2020) Shishira, S.R.; Kandasamy, A.Cloud computing is an important paradigm for processing, computation, storage, and network bandwidth. Workloads are the amount of data given to the hardware resource for processing. Its behavior and properties play a major role in the efficient scheduling of requests to given resources. Also, it is very difficult to predict workloads nature if they are changing excessively. To address this issue, we propose a conceptual framework which can be used for efficient prediction and optimization of workloads in a cloud environment. Classification of optimization metrics based on the provider and consumer constraints are presented. In addition to this, some of the research gaps found during the study has been highlighted and also provided possible solutions in the cloud research domain. © 2020 IEEE.