Adaptive Best Mother Wavelet Based Compressive Sensing Algorithm for Energy Efficient ECG Signal Compression in WBAN Node
Rajashekar Kunabeva1, Manjunatha P2, Narendra VG3
1Rajashekar Kunabeva Research Scholar, Dept of ECE, JNNCE, shivamogga, Karanataka, VTU, India.
2Dr. Manjunatha. P, Professor & HOD, Dept of ECE, JNNCE shivamogga, Karanataka, VTU, India..
3Dr. Narendra.V.G, Associate Professor, Dept of CSE, MIT, MAHE, Manipal, India.
Manuscript received on 02 July 2019 | Revised Manuscript received on 05 July 2019 | Manuscript published on 30 August 2019 | PP: 685-692 | Volume-8 Issue-10, August 2019 | Retrieval Number: J88010881019/2019©BEIESP | DOI: 10.35940/ijitee.J8801.0881019
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In ambulatory ECG monitoring application energy efficient signal acquisition plays significant role in ensuring the lifetime of resource constrained WBAN node . Most of the Compressive Sensing (CS) algorithms employ fixed mother wavelet choice for decomposition phase, resulting in incorrect block-wise data representation thus yielding higher PRD, lower CR and subsequent faster energy consumption rate. To overcome this design issue a novel minimum PRD based adaptive best mother wavelet (ABMW) selection algorithm has been proposed individually for each block and tested for compression of ECG signals in emergent CS paradigm over three datasets. Performance metrics illustrate that the proposed algorithm supports true representation of the physiological events, is energy efficient and faster than its predecessors and has an average execution delay of 1.7 seconds for compression and recovery of 10 seconds ECG data. Simulation results show that proposed algorithm achieved average PRD of 1.141733, CR of 63.77417 and SNR of 40.63878. The proposed algorithm achieved average PRD of 3.59 , execution speed of 2.09 seconds ,CR of 62.32, SNR of 29.5dB and energy consumption is around 1.64E-04 which is very near to average energy consumption values for both MIT-BIH ,PTB datasets and 24-bit acquired ECG data.
Keywords: Adaptive mother wavelet selection, Energy efficiency, Compressive Sensing, Sparse representation, WBAN
Scope of the Article: Compressive Sensing