China
Africa
Explore chapters and articles related to this topic
High-Capacity Watermarking
Published in Frank Y. Shih, Digital Watermarking and Steganography: Fundamentals and Techniques, 2017
For watermark images, set partitioning in hierarchical trees (SPIHT) compression [24] is adopted for preprocessing. SPIHT is an image compression algorithm using an essential bit allocation strategy that produces a progressively embedded scalable bitstream. The algorithm first encodes the most essential DWT coefficients, followed by transmitting the bits. Therefore, an increasingly refined compression of the original image can be obtained progressively. An example of SPIHT is shown in Figure 9.18, where (a) lists the bitstream generated by applying SPIHT and (b) shows the original and reconstructed images.
Mathematical Preliminaries
Published in Frank Y. Shin, Digital Watermarking and Steganography, 2017
Set partitioning in hierarchical tree (SPIHT), introduced by Said and Pearlman [13], is a zerotree structure for image coding based on DWT. The first zerotree structure, called an embedded zerotree wavelet (EZW), was published by Shapiro in 1993 [14]. The SPIHT coding uses a bit allocation strategy to produce a progressively embedded scalable bitstream.
Multimedia transmission in MC-CDMA using adaptive subcarrier power allocation and CFO compensation
Published in International Journal of Electronics, 2018
The efficiency of the proposed residual CFO compensated MWFA over MWFA without CFO compensation is validated using different metrics. The required Eb/No to achieve the BER of 10–4 is 9 dB for the proposed method whereas 11 dB for MWFA. The proposed method offers less ICI power such as −31 dB at while MWFA offers −20 dB. The maximum capacity offered by the proposed residual CFO compensated MWFA is 91 Mbps where WFA without CFO compensation is 80 Mbps. The maximum MSE and PSNR obtained from the proposed method are 10 pixels and 82 dB, respectively, whereas MWFA has 35 pixels and 40 dB, respectively. The proposed method is applied in SPIHT image compression for high quality and offers the retrieved image with high PSNR. The above discussions illustrate that the proposed method outperforms the existing power allocation algorithms hence it is the suitable solution for resource allocation problems in future generation wireless standards.
Pre-Processing Based ECG Signal Analysis Using Emerging Tools
Published in IETE Journal of Research, 2023
Varun Gupta, Arvind Kumar Sharma, Pawan Kumar Pandey, Ritesh Kumar Jaiswal, Anshu Gupta
In [27], Ustundag et al. demonstrated the use of fuzzy thresholding and wavelet packet analysis for de-noising the ECG. Wavelet packet analysis was considered for the decomposition of ECG signals at various levels whereas threshold value was obtained by fuzzy s-function. In [28], Bushra et al. detected the QRS wave by decompositions with the help of continuous wavelet transform (CWT) and Lipschitz exponent of the components. Finally, the QRS complex was classified by clustering technique (K means). In [29], Widjaja et al. presented automated pre-processing of RR intervals on an ordinary ECG signal. The work was validated by using ECG signals of an hour duration collected from 20 pregnant ladies. R peaks of the signal were manually modified for spurious and missed detections. They obtained an overall error rate of 0.06% after pre-processing. In [16], Bayasi et al. extracted ECG features for detecting ventricular arrhythmia based on a unique set obtained from two consecutive cardiac cycles. For classifying normal and abnormal patient’s conditions, Linear Discriminant Analysis (LDA) was used. These conditions were assessed using 10-fold cross validations and secured a precision (Pr) value of 98.39%. In [30], Rajankar and Talbar have proposed wavelet-based ECG signal compression because it provides better compression performance. Wavelet-based compression technique outperforms the other existing techniques. But these techniques were sensitive to noise. In [31], Xingyuan and Juan used a wavelet for implementing a hybrid compression technique on an ECG signal. They used two correlations of heartbeat signals, 2-D wavelet transform, set partitioning hierarchical trees (SPIHT) and the vector quantisation (VQ) methods. In [32], Subramanian and Ramasamy proposed dual-tree complex wavelet transform (DT-CWT) for ECG signal compression. Along with DT-CWT, Set Partitioning in Hierarchical Tree (SPIHT) coding was also included for enhancing the compression ratio. They have reported a performance improvement of 35.19% over other existing methods.