Analysis of Babble Noise based on MFCC in Stuttered Speech
Authors: Salma Jabeen, Dr. K.M. RaviKumar
Abstract
This paper describes the effect of 0dB and 10dB babble noise on stuttered speech. 100 samples are collected from the subjects (stutterer), among which 80 samples are used to prepare database dictionary and 20 samples for testing. The samples are segmented and pass through Mel Frequency Cepstral Coefficient (MFCC) feature extraction and statistical estimation to prepare dictionary, then testing samples compared with the database. The results obtained are 100%. When the same testing samples are added with the babble noise of 0dB and 10dB, the result degrades; the degraded results are enhanced by using sparse matrix enhancement technique. The results obtained after enhancement are 60-95% depending on the samples used for analysis.
Methane Potential Of Coffee Waste And Food Waste Using Gas Chromatography
Authors: Dhanasree A. K, M.A Chinnamma
Abstract
The purpose of the experiment was to study biochemical methane potential of two substrates namely coffeewaste left in coffee filter and food waste from hostel mess for bio-fuel production. Methane can be obtained from anaerobic decomposition of organic waste or energy crops and may turn out to be promising option for disposal of agro-waste. The anaerobic digestion of food waste and coffee waste produces biogas, a valuable energy resource Anaerobic digestion is a microbial process for production of biogas, which consist of Primarily methane (CH4) & carbon dioxide (CO2) and contains small amount of hydrogen sulphide (H2S) and ammonia(NH3), as well as trace amounts of other gases.The parameters analyzed were pH, TS/VS ratio, Alkalinity and VFA. The total gas produced was found by installing a simple liquid displacing system and the percentage of methane in the gas was found by gas chromatography.
Meta-Search Engine based on Semantic Similarity for E-Learning Application
Authors: Harini Natarajan, Anita C.S
Abstract
With the sharp growth in internet usage, the amount of data stored in web is outsized. The web users simply query their needs to get appropriate results. This is enabled through search engines. Search engines are software, particularly designed to search information from World Wide Web (WWW).Such traditional engines go in vain while searching the semantic web. Thus metasearch engines were developed, which aims in utilizing the traditional search engines to enhance the search results to retrieve links that are highly related to user query. A semantic metasearch engine is proposed which includes semantic analysis of user queries and Meta search engine benefits. The proposed system uses WordNet tool to mine synonymous words for the given user query. Then these words are passed to the Google search engine to retrieve top 10 links. The tag of each link is pre- processed to obtain the keywords and ontology is created. The genetic algorithm is used to rank the links which has high word fitness to the user query. Finally after harmonization, the keywords are passed to Google to get significant results. The resultant links prove to be greatly related to user query and are also mostly active links. This metasearch engine can be implemented in E-learning domain to aid the students and researchers.
Low complexity detection and PAPR reduction in SFBC BICM-OFDM system
Authors: Renjitha Mariam Rajan, Teena Rajan
Abstract
This paper is concerned with a multi input multi output 2X2 antenna system based on almounti transmission. For higher data rate transmission orthogonal frequency division multiplexing is also incorporated with MIMO system. One of the main drawback of MIMO-OFDM system is the high decoding complexity which increases with the number of transmitted symbol per code word. This paper proposes a lower complexity detection scheme based on list sphere decoding using tree search. Here a list of candidates is selected based on tree search which is done in an unconstrained fashion and these list of candidates is considered for the detection process there by reducing the computational complexity. SFBC BICM OFDM system will suffer from high Peak average power ratio so PAPR reduction based on clipping and amplification is also done.
2x2 MIMO OFDM Systems for Adaptive Channel Estimation Technique
Author: Mr. S. Ramkumar
Abstract
In this paper analysis of least squares (RLS) algorithm and LMS algorithm is proposed for multiple input multiple output (MIMO) orthogonal frequency division multiplexing (OFDM) systems. These two algorithms are used as adaptive channel estimation techniques. CE methods uses adaptive estimator which are able to update parameters of the estimator continuously, so that the knowledge of channel and noise statistics are not necessary. This LMS CE algorithm requires knowledge of the received signal only. So by designing this we are going to analyze the terms of the mean squares error (MSE), and bit error rate (BER) on mat lab software. Simulation results demonstrated that, the LMS CE method has better for MIMO OFDM systems. In addition, the utilizing of more multiple antennas at the transmitter and/or receiver provides a much higher performance compared with fewer antennas.