K. P. Soman

A novel method for detecting R-peaks in electrocardiogram (ECG) signal

Abstract The R-peak detection is crucial in all kinds of electrocardiogram (ECG) applications. However, almost all existing R-peak detectors suffer from the non-stationarity of both QRS morphology and noise. To combat this difficulty, we propose a new R-peak detector, which is based on the new preprocessing technique and an automated peak-finding logic. In this paper, we first demonstrate that the proposed preprocessor with a Shannon energy envelope (SEE) estimator is better able to detect R-peaks in case of wider and small QRS complexes, negative QRS polarities, and sudden changes in QRS amplitudes over that using the absolute value, energy value, and Shannon entropy features. Then we justify the simplicity and robustness of the proposed peak-finding logic using the Hilbert-transform (HT) and moving average (MA) filter. The proposed R-peak detector is validated using the first-channel of the 48 ECG records of the MIT-BITH arrhythmia database, and achieves average detection accuracy of 99.80%, sensitivity of 99.93% and positive predictivity of 99.86%. Various experimental results show that the proposed R-peak detection method significantly outperforms other well-known methods in case of noisy or pathological signals.

Modern analog and digital communication systems development using GNU Radio with USRP

In this modern world many communication devices are highly intelligent and interconnected between each other. Any up-gradation of the hardware in the existing communication devices is not easier one. Compatibility of the new hardware with existing hardware is highly essential. But the new protocols may or may not support the older one. The solution for these problems can be provided by using the reconfigurable hardware design. The hardware can be reprogrammed according to the new change in technology up-gradation. The cost of commercially available hardware and software requirements for setting up such a module is very high. This can be solved by using Open source hardware and software such as Universal Software Radio Peripheral (USRP) and GNU Radio. This work demonstrates how the modern analog communication system like Community Radio Schemes and Radio Data System (RDS) and digital communication systems such as Simple Digital Video Broadcasting (DVB) and OFDM based data communication can be developed using the Open source hardware USRP1. This work will be helpful even for first year level of engineering students to easily implement any communication and control applications with cheaper cost.

Tamil Font Recognition Using Gabor Filters and Support Vector Machines

Tamil Font Recognition is one of the Challenging tasks in Optical Character Recognition and Document Analysis. Most of the existing methods for font recognition make use of local typographical features and connected component analysis. In this paper, Tamil font recognition is done based on global texture analysis. The main objective of this proposal is to employ support vector machines (SVM) in identifying various fonts in Tamil. The feature vectors are extracted by making use of Gabor filters and the proposed SVM is trained using these features. The method is found to give superior performance over neural networks by avoiding local minima points. The SVM model is formulated tested and the results are presented in this paper. It is observed that this method is content independent and the SVM classifier shows an average accuracy of 92.5%.

Rule Based Machine Translation from English to Malayalam

Here we propose a method for translating English sentences to Malayalam. This machine translation is done by rule based method. The core process is mediated by bilingual dictionaries and rules for converting source language structures into target language structures. The rules used in this approach are prepared based on the Parts Of Speech (POS) tag and dependency information obtained from the parser. There are mainly two types of rules used here, one is transfer link rule and the other is morphological rules. In this method, the transfer link rules are used for generating target structure. Morphological rules are used for assigning morphological features. The bilingual dictionary used here is English, Malayalam bilingual dictionary. By using this approach, a given English sentence can be translated to its Malayalam equivalent.

Secrecy of Cryptography with Compressed Sensing

This paper deals with a new image encryption scheme which employs both compressive sensing and Arnold scrambling method. The compressed sensing(CS) paradigm unifies sensing and compression of sparse signals in a simple linear measurement step. Compressed measurements are scrambled using Arnold transform. So this system provides more security to the data.

Ensuring Security to the Compressed Sensing Data Using a Steganographic Approach

This paper focuses on the strength of combining cryptography and steganography methods to enhance the security of communication over an open channel. Here the data to be send are secured by using the compressive sensing method and the Singular Value Decomposition (SVD) based embedding method. The data is encrypted using the compressive measurements of the data and the resultant data is embedded in the cover object using the SVD based water mark embedding algorithm. This approach helps to send the secret data after hiding in a cover image. The compressive sensing method helps to compress and encrypt the data in a single step. The proposed system provides more security to the compressed data. This scheme significantly reduces the attacks. This method is very useful to hide the secret images. The results demonstrate that the proposed system is highly efficient and robust.

A robust watermarking method based on Compressed Sensing and Arnold scrambling

Watermarking is a technique for information hiding, which is used to identify the authentication and copyright protection. In this paper, a new method of watermarking scheme is proposed, which uses both Compressed Sensing and Arnold scrambling method for efficient data compression and encryption. Compressive sensing technique aims at the reconstruction of sparse signal using a small number of linear measurements. Compressed measurements are then encrypted using Arnold transform. The proposed encryption scheme is computationally more secure against investigated attacks on digital multimedia signals.

Kernel based part of speech tagger for Kannada

The proposed paper presents the development of a part-of-speech tagger for Kannada language that can be used for analyzing and annotating Kannada texts. POS tagging is considered as one of the basic tool and component necessary for many Natural Language Processing (NLP) applications like speech recognition, natural language parsing, information retrieval and information extraction of a given language. In order to alleviate problems for Kannada language, we proposed a new machine learning POS tagger approach. Identifying the ambiguities in Kannada lexical items is the challenging objective in the process of developing an efficient and accurate POS Tagger. We have developed our own tagset which consist of 30 tags and built a part-of-speech Tagger for Kannada Language using Support Vector Machine (SVM). A corpus of texts, extracted from Kannada news papers and books, is manually morphologically analyzed and tagged using our developed tagset. The performance of the system is evaluated and we found that the result obtained was more efficient and accurate compared with earlier methods for Kannada POS tagging.

Exploiting GNU radio and USRP: An economical test bed for real time communication systems

Communication experiments using normal lab setup, which includes more hardware and less software raises the cost of the total system. The method proposed here provides a new approach through which all the analog and digital experiments can be performed using a single hardware-USRP (Universal Software Radio Peripheral) and software-GNU Radio Companion (GRC). Initially, networking setup is formulated using SDR technology. Later on, one of the analog communication experiments is demonstrated in real time using the GNU Radio Companion, RTL-SDR and USRP. The entire communication system is less expensive as the system uses a single reprogrammable hardware and most of the focus of the system deals with the software part.

Kernel Method for English to Kannada Transliteration

Language transliteration is one of the important area in natural language processing. Accurate transliteration of named entities plays an important role in the performance of machine translation and cross-language information retrieval processes. The transliteration model must be design in such a way that the phonetic structure of words should be preserve as closely as possible. This paper addresses the problem of transliterating English to Kannada language using a publicly available structured output Support Vector Machines (SVM). The proposed transliteration scheme uses sequence labeling method to model the transliteration problem. This transliteration technique was demonstrated for English to Kannada Transliteration and achieved exact Kannada transliterations for 87.28% of English names.