Showing papers by "Goutam Saha published in 2019"
06 Sep 2019
TL;DR: This paper first discussed the evolution of conventional IoT to the SDN‐based IoT, which can resolve many drawbacks of a conventional IoT system and focused on how the concept of blockchain can be converged with SDN-based IoT system to further improve its security aspects.
Abstract: Blockchain is a key technology that enables cryptocurrencies such as Bitcoin, Litecoin, etc. In recent years, researchers have ventured into tapping the potential of blockchain‐based ecosy...
57 citations
TL;DR: A new multi-channel PCG-based system to classify CAD-affected and normal subjects is proposed, and it does not require any additional reference signal, such as an electrocardiogram (ECG) signal.
36 citations
TL;DR: This paper incorporated one of the evolutionary algorithms known as Particle Swarm Optimization (PSO) which was used in estimating the parameters pertaining to ANFIS and found that proposed PSO-ANFIS performed better than ARIMA and conventional ANfIS with respect to the prediction accuracy of runoff.
Abstract: Rainfall-Runoff modeling plays a crucial role in various aspects of water resource management. It helps significantly in resolving the issues related to flood control, protection of agricultural lands, etc. Various Machine learning and statistical-based algorithms have been used for this purpose. These techniques resulted in outcomes with an acceptable rate of success. One of the pertinent machine learning algorithms namely Adaptive Neuro Fuzzy Inference System (ANFIS) has been reported to be a very effective tool for the purpose. However, the computational complexity of ANFIS is a major hindrance in its application. In this paper, we resolved this problem of ANFIS by incorporating one of the evolutionary algorithms known as Particle Swarm Optimization (PSO) which was used in estimating the parameters pertaining to ANFIS. The results of the modified ANFIS were found to be satisfactory. The performance of this modified ANFIS is then compared with conventional ANFIS and another popular statistical modeling technique namely ARIMA model with respect to the forecasting of runoff. In the present investigation, it was found that proposed PSO-ANFIS performed better than ARIMA and conventional ANFIS with respect to the prediction accuracy of runoff.
12 citations
TL;DR: A computer aided diagnosis (CAD) method has been proposed based on the Doppler blood flow spectrograms of lower limb arteries based on which smartphone application is implemented to provide a cost-effective, portable, user-friendly solution for Point-of-Care US system.
6 citations
TL;DR: Recurrent Neural Networks and its variants, namely, LSTM and GRU were tried for the purpose of forecasting from time series data and the results confirm that GRU has the best predictive ability in case of temporal problems.
Abstract: Most of the real world prediction problems are naturally associated with time component which requires time series data as input. Presently, machine learning approaches are used for forecasting pur...
6 citations
01 Jan 2019
TL;DR: A detail investigation of IoT issues which can be improved by incorporating Software-Defined Network is presented, finding that SDN-based IoT is much powerful than that of the exiting IoT system.
Abstract: Internet of Things (IoT) is gaining a significant amount of importance in the field of networking. It uses low-cost devices to provide the application to the users. IoT devices are resource-constraint device, so they suffer from the constraints of availability, reliability, flexibility, and security. Lots of research attention has been given to minimize these challenges. By integrating the features of Software-Defined Network (SDN), most of the IoT issues can be overcome. This paper presents a detail investigation of IoT issues which can be improved by incorporating SDN. The resources used for SDN are equipped with huge computational power. This particular feature can help IoT devices to overcome the constraint problem. The virtualization concept of SDN will help to resolve the scalability issues of IoT. Similarly, many other issues can be resolved at a very low cost. Researchers are finding that SDN-based IoT is much powerful than that of the exiting IoT system.
5 citations
16 Dec 2019
TL;DR: In this article, the authors have tried to highlight the evolution of conventional IoT architecture to SDN based IoT architecture which offers better provisions to resolve Big Data issues also, and also highlight the procedures, developed by investigators of the area, as to how Big data issues can be resolved in these architectures.
Abstract: IoT infrastructure is resource and energy constraint. The emergence of a smart world has led to the interconnection of diverse objects with the Internet which leads to the generation of a tremendous amount of data which can be referred to as Big Data. Since the IoT is going to be expanded in a exponential manner, Big Data Issues becomes a threat for its faithful operation. Other than this, the existing IoT architecture has limitations with respect to scalability, reliability, availability, etc. To overcome these limitations along with resolution of Big Data issues, investigators proposed different types of architectures for its improvement. In this paper, we have tried to highlight the evolution of conventional IoT architecture to SDN based IoT architecture which offers better provisions to resolve Big Data issues also. This review paper also highlights the procedures, developed by investigators of the area, as to how Big Data issues can be resolved in these architectures. Further research scopes arising out of the present endeavor are also been highlighted which will provide the direction of future research in the area.
3 citations
TL;DR: An efficient fault detection mechanism is formulated to identify multiple numbers of defective/faulty electrodes on an m × n biochip array, where m and n can be of any positive number.
Abstract: The involvement of Digital Microfluidic Biochips (DMFBs) in the field of disease detection, automated drug discovery, on-chip DNA (Deoxyribonucleic acid) analysis has become well-accepted d...
3 citations
01 Jan 2019
TL;DR: An exhaustive study of brain tumor from MRI images has made by using different techniques and a comparative study and performance evaluation of different techniques based on certain performance metrics are discussed.
Abstract: Segmentation of brain tumor from medical images is an interesting topic which is investigated by many researchers. It is important to locate tumor at an early stage so that it can be easily healed and can be used for further diagnosis. There are different imaging techniques which are used in segmentation of brain tumor. Among them, Magnetic Resonance Imaging (MRI) is most widely used radiological tool as it is radiation free in nature. For detecting the size, shape, and location of the tumor many segmentation algorithms were used. In this paper, an exhaustive study of brain tumor from MRI images has made by using different techniques. A comparative study and performance evaluation of different techniques based on certain performance metrics are also discussed in this paper.
2 citations
TL;DR: In this article, a study for identification of feasible sites for construction of water reservoirs at appropriate locations based on the integration of thematic data on slope, flow accumulation, lithology, drainage density, lineament density and lineament direction using the advanced tools of ArcGIS software.
Abstract: Aizawl city is experiencing seasonal water scarcity. Tlawng is the longest River in Mizoram which flows for about 234 km length in Mizoram in south - north direction and joins the River Barak in Cachar district of Assam state. As Tlawng River is close to the Aizawl city, it will be wise to make some safe and durable strategy for storing rain water safely in predefined reservoirs and use it during hour of needs for domestic and agricultural purposes. In fact, Aizawl city is located within the Tlawng River basin at an average elevation of about 1,200 metres above mean sea level. An attempt has been made in this study for identification of feasible sites for construction of water reservoirs at appropriate locations based on the integration of thematic data on slope, flow accumulation, lithology, drainage density, lineament density and lineament direction using the advanced tools of ArcGIS software. Thirty two ideal sites have been identified in the vicinity of Aizawl city based on the multi-criteria evaluation of the thematic layers.
1 citations
13 Dec 2019
TL;DR: An investigation has been made to design a protocol, termed as 6LoWSD protocol, which has the ability to integrate the two systems, and required experimentation have been carried out on simulation to ascertain the effectiveness of the proposed protocol.
Abstract: With advent of smart cities, the IoT infrastructure is expanding in an exponential manner. This emerging infrastructure is going to produce a huge amount of data to be processed. Besides the data processing bottleneck, energy efficiency, security, interoperability and scalability of the network standard poses another problem to the present Internet of Things (IoT) architecture. So this become an interesting research issue which aims to improve the present IoT architecture by incorporating Software Defined Network (SDN) system in it. SDN, a relatively new paradigm has the capability of resolving the mentioned complexities. If SDN can be incorporated in traditional IoT architecture this will help to resolve many IoT problems. Now the research issue becomes as how to design an effective protocol which can bridge between the traditional IoT and SDN system for effective interoperability. In SDN, control operation are designated in the middle-ware layer, where SDN controllers are placed. Controllers maintains the operation of the whole network. In this paper, an investigation has been made to design a protocol, termed as 6LoWSD protocol, which has the ability to integrate the two systems. Required experimentation have been carried out on simulation to ascertain the effectiveness of the proposed protocol. Here, 6LoWSD was simulated on Contiki platform. 6LoWPAN has been used for the IoT system and OpenFlow for the SDN system. The experimental results displayed quite satisfactory outcomes.
01 Jun 2019
TL;DR: This paper proposes to capture textural information through statistics of local binary pattern of the mel-filterbank energies through a framework that outperforms two mel-scale based benchmark systems.
Abstract: Context-aware devices and applications can benefit when audio from real-life environments is categorized into different acoustic scenes. Such categorization is referred to as acoustic scene classification (ASC). However, the scene labels are database dependent. For most of the ASC applications, rather than giving explicit scene labels (like home, park etc), a general estimate of the type of surroundings (e.g. indoor or outdoor) might be enough. ASC has been generally achieved with mel-scaled cepstral features by the state-of-the-art systems. The characteristics that differentiate one scene class from the other are embedded in the texture of the time-frequency representation of the audio. In this paper, we propose to capture this textural information through statistics of local binary pattern of the mel-filterbank energies. The experiments were conducted on two datasets having same scene classes but varying audio sample duration and unequal total amount of data. The proposed framework outperforms two mel-scale based benchmark systems.
TL;DR: Experimental results show that adopted approach can overcome the problems with conventional approaches and achieve segmentation result with an acceptable level of accuracy.
Abstract: Hydrologists use various machine learning approaches for segmenting water bodies for carrying out various panning activities. Conventional approaches are mainly clustering or classification based a...
01 Dec 2019
TL;DR: This work uses spectral entropy modulated speech-signal-based scale along with Gaussian filter to derive new cepstral feature for ASV task and finds improved performance of proposed features over baseline method in both clean and noisy conditions.
Abstract: In different speech-based applications, the speechsignal-based frequency cepstral coefficient (SFCC) based feature extraction method has been used successfully. The conventional method for extraction of such feature uses triangular filter banks similar to mel frequency cepstral coefficient (MFCC) features. In this work, we first present Gaussian filter based speech-signalbased frequency cepstral coefficient (GSFCC) based feature extraction technique derived from the speech-signal-based scale. Next, we use spectral entropy modulated speech-signal-based scale along with Gaussian filter to derive new cepstral feature for ASV task. We find improved performance of proposed features over baseline method in both clean and noisy conditions. The experiments were carried on NIST SRE 2001 database with simulated noise. In addition, a more recent real-world, noisy database VoxCeleb1 is used in the study. Finally, we do score level fusion of auditory filter based cepstral feature with datadriven filter based cepstral features. The performance obtained by the proposed features gives up to 15.71% and 15.37% relative improvement in equal error rate (EER) over the baseline method in NIST SRE and VoxCeleb1 databases, respectively.