scispace - formally typeset
Search or ask a question
Author

M. Palanivelan

Bio: M. Palanivelan is an academic researcher from Rajalakshmi Engineering College. The author has contributed to research in topics: Orthogonal frequency-division multiplexing & Spectral efficiency. The author has an hindex of 3, co-authored 22 publications receiving 38 citations.

Papers
More filters
Proceedings ArticleDOI
01 Feb 2017
TL;DR: In this article, distance based interference mitigation (DBIM) algorithm is proposed to mitigate interference where the locations of user equipment (UEs) and evolved NodeB (eNB) are identified and distances between them are estimated using Haversine Formula.
Abstract: Device-to-device (D2D) communication brings many advantages in underlay cellular networks such as improving spectrum efficiency, energy efficiency and cellular capacity. However, D2D transmission creates interference to the cellular user equipment (CUEs) and other D2D pairs. So, proper resource allocation handling is required to suppress the interference introduced by the addition of D2D user equipment (DUEs) in the existing cellular networks. In this paper, Distance Based Interference Mitigation (DBIM) algorithm is proposed to mitigate interference where the locations of user equipment (UEs) and evolved NodeB (eNB) are identified and distances between them are estimated using Haversine Formula. In this work resources are allocated to UEs based on distance constraints. Further, the significance of D2D communication and simulations to justify the proposed scheme is also presented.

10 citations

Journal ArticleDOI
TL;DR: The proposed model has shown considerable improvement in downgrading processing time overcoming the issues of cumbersome hyper-parameter tuning and huge data demand of the deep learning algorithms with high classification accuracy.
Abstract: In this paper, a novel Hybrid Deep Ensemble (HDE) is proposed for automatic speech disfluency classification on a sparse speech dataset. Categorizations of speech disfluencies for diagnosis of speech disorders have so long relied on sophisticated deep learning models. Such a task can be accomplished by a straightforward approach with high accuracy by the proposed model which is an optimal combination of diverse machine learning and deep learning algorithms in a hierarchical arrangement which includes a deep autoencoder that yields the compressed latent features. The proposed model has shown considerable improvement in downgrading processing time overcoming the issues of cumbersome hyper-parameter tuning and huge data demand of the deep learning algorithms with high classification accuracy. Experimental results show that the proposed Hybrid Deep Ensemble has superior performance compared to the individual base learners, and the deep neural network as well. The proposed model and the baseline models were evaluated in terms of Cohen’s kappa coefficient, Hamming loss, Jaccard score, F-score and classification accuracy.

10 citations

Journal ArticleDOI
TL;DR: In this paper, the Deep Long-short term memory Autoencoder (DLAE), a regularized deep learning model, is proposed for the automatic severity assessment of phonological deviations which are crucial in the classification of long-term memory errors.
Abstract: In this paper, the Deep Long-short term memory Autoencoder (DLAE), a regularized deep learning model, is proposed for the automatic severity assessment of phonological deviations which are crucial ...

6 citations

Journal ArticleDOI
25 Oct 2018-PLOS ONE
TL;DR: Receive Diversity based Transmission Data Rate Optimization (RDTDRO) scheme is proposed to improve the network lifetime and delay efficiency of Multi level-Quadrature Amplitude Modulation (M-QAM) based WBAN.
Abstract: Wireless Body Area Network (WBAN) has become the emerging technology due to its ability to provide intelligent and cost-effective healthcare monitoring solution. The biological sensors used in WBAN are energy-constrained and required to be functional for a longer duration. Also, the sensed data should be communicated in reasonable time. Therefore, network lifetime and delay have become the primary concerns in the design of WBAN. In this paper, Receive Diversity based Transmission Data Rate Optimization (RDTDRO) scheme is proposed to improve the network lifetime and delay efficiency of Multi level-Quadrature Amplitude Modulation (M-QAM) based WBAN. In the proposed RDTDRO scheme, minimum energy consumption is ensured by optimizing the transmission data rate with respect to a given transmission distance and number of receive antennas while satisfying the Bit Error Rate (BER) requirements. The performance of proposed RDTDRO is analyzed in terms of network lifetime and delay difference and is compared with conventional Baseline and Rate optimized schemes. The results show that at a transmission distance of 0.3 m, the proposed RDTDRO scheme with a receive diversity order of 4 achieves 1.30 times and 1.27 times improvement in network lifetime over conventional Baseline and Rate optimized schemes respectively. From the results, it is also evident that at a transmission distance of 0.3 m, the proposed RDTDRO scheme with a receive diversity order of 4 is delay efficient as it achieves delay difference of 0.75 μs and 0.29 μs over conventional Baseline and Rate optimized schemes respectively.

6 citations

Journal ArticleDOI
TL;DR: This work proposes a segmentation algorithm for identifying brain tumour regions using the fuzzy integrated active contour model, which is more robust than the classical snake methods based on the gradient.
Abstract: Magnetic resonance imaging used for diagnosis, localisation, and volume quantisation of tumours helps radiologists set treatment plans. Medical image segmentation is vital in detecting tumours. We ...

6 citations


Cited by
More filters
Journal ArticleDOI
TL;DR: This paper reveals that the NOMA techniques have evolved from single-carrier NomA (SC-NOMA) into multi- carrier NOMa (MC-N OMA), and comprehensively investigated on the basic principles, enabling schemes and evaluations of the two most promising MC-NomA techniques, namely sparse code multiple access (SCMA) and pattern division multiple access(PDMA).
Abstract: Non-orthogonal multiple access (NOMA) is one promising technology, which provides high system capacity, low latency, and massive connectivity, to address several challenges in the fifth-generation wireless systems. In this paper, we first reveal that the NOMA techniques have evolved from single-carrier NOMA (SC-NOMA) into multi-carrier NOMA (MC-NOMA). Then, we comprehensively investigated on the basic principles, enabling schemes and evaluations of the two most promising MC-NOMA techniques, namely sparse code multiple access (SCMA) and pattern division multiple access (PDMA). Meanwhile, we consider that the research challenges of SCMA and PDMA might be addressed with the stimulation of the advanced and matured progress in SC-NOMA. Finally, yet importantly, we investigate the emerging applications, and point out the future research trends of the MC-NOMA techniques, which could be straightforwardly inspired by the various deployments of SC-NOMA.

104 citations

Journal ArticleDOI
TL;DR: A QoS-aware load balancing strategy (QALB) for software defined Wi-fi networks (SD-Wi-Fi), as a solution to address the problem of Wi-Fi congestion among the OpenFlow enabled APs (OAPs).
Abstract: High density Wi-Fi networks require load balancing in order to ensure quality of service (QoS) In the traditional Wi-Fi networks, the wireless stations learn the access points (APs) load and make the association decisions themselves leading to uneven load distribution among the APs The new paradigm, software defined networking (SDN), has a centralized architecture, which allows to manage, measure and control high density Wi-Fi networks easily In this paper we propose a QoS-aware load balancing strategy (QALB) for software defined Wi-Fi networks (SD-Wi-Fi), as a solution to address the problem of Wi-Fi congestion among the OpenFlow enabled APs (OAPs) The SDN controller selects a load level up to which the association decisions are made by the OAPs without consulting the controller The wireless stations from an overloaded OAP are handed to an underloaded OAP by considering multi-metrics such as the packet loss rate, received signal strength indicator (RSSI) and throughput An emulation platform and a large-scale-low-cost testbed with the same settings are constructed to evaluate the performance of our load balancing strategy The results show that in comparison to four non-static schemes such as, channel measurement based access selection scheme (CMAS), (DL-SINR) downlink-signal to interference plus noise ratio AP selection scheme (DASA), mean probe delay scheme (MPD) and RSSI scheme, the proposed QALB, optimizes the throughput up to 16%, reduces the average frame delay up to 19%, minimizes the number of re-transmissions by 49%, reduces the number of handoffs by 15%, improves the degree of load balancing by 22% and minimizes the re-association times by 38%, in high density SD-Wi-Fi

24 citations

Journal ArticleDOI
TL;DR: In this article, the authors survey the field of brain tumor MRI images segmentation and summarize multi-modal brain tumor image segmentation methods, which are divided into three categories: conventional segmentation, segmentation based on classical machine learning methods, and segmentation method based on deep learning methods.

22 citations

Journal ArticleDOI
TL;DR: This work proposes computation offloading and resource allocation optimization for AVR (CoroAVR) algorithm with hybrid sensing data fusion of cooperative perception to process the real-time data in fog-edge computing for maximizing system throughput and spectrum utilization on the premise of ensuring the quality of task completion.
Abstract: Augmented vehicular reality (AVR) is one of the key technologies to realize intelligent transportation in the future, which can significantly improve traffic safety and transportation efficiency of autonomous driving. However, available computation and spectrum resources of vehicles are not well utilized to meet the requirements of cooperative perception of autonomous vehicles. In order to meet the needs of sharing sensing data to cooperatively perceive the surrounding environment, executing delay-sensitive and computationally intensive tasks of autonomous driving applications, we propose computation offloading and resource allocation optimization for AVR (CoroAVR) algorithm with hybrid sensing data fusion of cooperative perception to process the real-time data in fog-edge computing for maximizing system throughput and spectrum utilization on the premise of ensuring the quality of task completion. First, the minimum signal to interference plus noise ratio needed to complete the task is obtained with the given maximum computing resources. Second, the optimal power allocation is carried out by using convex optimization theory, and the throughput gain is calculated, and the offloading decision is made by comparing the throughput gain. Finally, the channel allocation problem is solved by using the maximum matching algorithm of the bipartite graph, and the computation resource allocation is studied. The simulation results show that the performance of the proposed algorithm is better than that of the contrast algorithm.

17 citations

Posted Content
TL;DR: This work investigates methods for combining multiple self-supervised tasks-i.e., supervised tasks where data can be collected without manual labeling to tackle the training data bottleneck in disfluency detection.
Abstract: Most existing approaches to disfluency detection heavily rely on human-annotated data, which is expensive to obtain in practice. To tackle the training data bottleneck, we investigate methods for combining multiple self-supervised tasks-i.e., supervised tasks where data can be collected without manual labeling. First, we construct large-scale pseudo training data by randomly adding or deleting words from unlabeled news data, and propose two self-supervised pre-training tasks: (i) tagging task to detect the added noisy words. (ii) sentence classification to distinguish original sentences from grammatically-incorrect sentences. We then combine these two tasks to jointly train a network. The pre-trained network is then fine-tuned using human-annotated disfluency detection training data. Experimental results on the commonly used English Switchboard test set show that our approach can achieve competitive performance compared to the previous systems (trained using the full dataset) by using less than 1% (1000 sentences) of the training data. Our method trained on the full dataset significantly outperforms previous methods, reducing the error by 21% on English Switchboard.

15 citations