China
Africa
Explore chapters and articles related to this topic
Lossless Coders
Published in Keshab K. Parhi, Takao Nishitani, Digital Signal Processing for Multimedia Systems, 2018
Ming-Ting Sun, Sachin G. Deshpande, Jenq-Neng Hwang
The run-length coding discussed above is relatively easy to implement by using a counter to count the number of repeated symbols. Huffman encoders and decoders, however, are more complicated. The operation of Huffman encoding and decoding involve bit-level manipulation, thus, the implementation on general purpose processors may be relatively slow for real-time video processing. In Huffman encoding, a symbol to be encoded is used to look up a code-book to obtain the variable-length code-word and the corresponding code-length. Based on the code-length information, a concatenation circuit concatenates the variable-length code-words into a continuous bit-stream and segments the concatenated bit-stream into fixed-length words so that they can be stored in a fixed-length memory after the Huffman encoder. After the concatenation of the variable-length code-words, there is no longer clear symbol-boundaries. In Huffman decoding, the bit-stream has to be decoded from the beginning to obtain the decoded symbols one by one. The decoding of the 2nd code-word cannot start before the first code-word and the corresponding code-lengths are decoded. Due to this inherent serial decoding nature, it is difficult to use traditional high-speed implementation techniques such as parallel processing to achieve high-speed Huffman decoding. There are many different approaches proposed for high-speed Huffman decoding [16]–[23]. Some are more suitable for implementations using hardware or special processors. In the following, we only discuss some approaches which are suitable for implementation in both software and hardware.
Basics of Electrical Communication Systems
Published in P. S. Neelakanta, ATM Telecommunications, 2018
This is an improved version of the simple variable bit length coding discussed above. It also relies on the concept of entropy. Huffman code offers an organized technique to find the best possible variable length code for a given set of messages. The procedure to construct a Huffman code involves two steps, namely reduction and splitting.
V
Published in Philip A. Laplante, Comprehensive Dictionary of Electrical Engineering, 2018
van der Pol oscillator van der Pol oscillator an oscillator or oscillating system described by the equation dx d2x +x =0 - µ(1 - x 2 ) dt dt 2 This equation is mentioned in almost any book on oscillators or on nonlinear mechanics. The reason is that this equation is relatively simple, yet it successfully lumps together two rather complex processes associated with oscillators, namely, the process of generation of periodic waveform and the process of automatically stabilizing the amplitude of this wave. The attempts to improve the solutions of the van der Pol equation and to apply to design of low-distortion oscillators can be traced to recent publications. Vander Lugt filter encoded optical mask for representing, in the Fourier-transform domain, the reference or library functions needed in an image correlator; encoding is performed holographically. vanishing point the point in the perspective projection plane in which a system of 3-D parallel lines converge. Vapnik-Chervonenk (VC) dimension a measure of the expressive power of a learning system with binary or bipolar outputs. For neural networks it is closely related to the number of weights in the network. For single-layer networks it is simply equal to the number of weights (including biases) but for multilayer and other networks, analytic expressions for VC dimension are not available. vapor cooling a cooling technique for power vacuum tubes utilizing the conversion of hot water to steam as a means of safely conducting heat from the device and to a heat sink. VAR See volt-ampere reactive. as a tuning element or as a nonlinear frequency multiplier. varactor diode a diode designed to have a repeatable and high capacitance vs. reverse voltage characteristic. A two terminal semiconductor device in which the electrical characteristic of primary interest is the voltage dependent capacitance. varactor tuner a tuning circuit at the input of a television receiver that uses a varactor diode. The tuning capability comes from the characteristic of a varactor, or varicap, to function as a voltagesensitive capacitance. variable bit rate (VBR) describes a traffic pattern in which the rate at which bits are transmitted varies over time; such patterns are also referred to as bursty. VBR sources often result from compressing CBR sources, for example, a 64 kbps voice source in its raw form has a constant bit rate; after compression by removing the silence intervals, the source becomes VBR. variable frequency drive electric drive system in which the speed of the motor can be varied by varying the frequency of the input power. variable length code to exploit redundancy in statistical data, and to reduce average number of bits per word luminance levels having high probability are assigned short code words and those having low probability are assigned longer code words. This is called variable length coding or entropy coding. See also entropy coding. variable loss machine loss that changes with a change in the mode of machine operation such as loading, temperature and current. For example, in a transformer, the winding losses are a function of the load current, while the core losses are almost independent of the load current. variable polarity plasma arc welding (VP-PAW) a welding process that produces
Region of interest-based adaptive segmentation for image compression using hybrid Jaya–Lion mathematical approach
Published in International Journal of Computers and Applications, 2021
B. P. Santosh Kumar, K. Venkata Ramanaiah
In general, digital image is characterized via three major kinds of redundancies: coding redundancy, spatial redundancy as well as psychovisual redundancy. The compression models [15–18] utilize all the mentioned redundancies for compressing evaluated by values of another pixel. The redundancy of coding is also termed to be the statistical redundancy, which is referred as the utilization of variable length code for matching the original image’s statistics. In addition, the Psychovisual redundancy is on the basis of information’s human perception. Eliminating them generates distortion of uncompressed images, hence the step is skipped while doing image compression under lossless category. In lossy compression model, there is loss of data or information. However, lossless approach is reversible, which represents the image signed along least feasible count bits without losing the significant data or information, and the compression ratio attained is less, including JPEG-LS, LOCO-I, and JPEG2000 (5/3).
An efficient two-digit adaptive delta modulation for Laplacian source coding
Published in International Journal of Electronics, 2019
Zoran Peric, Bojan Denic, Vladimir Despotovic
Slope overload distortion problem can be addressed using Adaptive Delta Modulation (ADM) (Jayant & Noll, 1984; Sayood, 2005), where the step size adaptation is performed according to changes in the input signal. An example of ADM is the Constant Factor Delta Modulation (CFDM), where one- or two-bit memory is introduced to estimate the step size at each sampling instant (Sayood, 2005). In Continuously Variable Slope Delta Modulation (CVSDM) the step size of the approximated signal is progressively increased or decreased, in case the same state has been observed three times in a row. Various multi-bit and multi-digit DM systems were proposed for variable step size estimation (Denic, Peric, & Despotovic, 2017; Prosalentis & Tombras, 2007; Tombras, 1990a, 1990b). The adaptive 2-digit system has been discussed in Tombras (1990a), that uses both ‘memory’ and ‘look ahead’ step size estimation and generates output codeword consisting of a binary and a ternary digit. The implementation of 2-digit ADM has been presented in Tombras (1990b), while a proper line coding format for a 2-digit ADM is provided in Tombras (1992). The modification of such system has been proposed in Prosalentis and Tombras (2007) that eliminates the need for a ternary digit and produces 2-bit output codeword. The ADM algorithm with three-level quantization designed for variable length code-words has been studied in Denic et al. (2017), where the adaptation is framewise. The algorithms based on Sigma Delta Modulation (SDM) have also received significant attention (Aldajani & Sayed, 2001; Frick, Dadouche, & Berviller, 2016; Prosalentis & Tombras, 2008, 2009; Zhang & Shi, 2018; Zierhofer, 2000). Basically, the SDM system is characterized by an integrator added in front of the ordinary DM, followed by a differentiator in front of the DM demodulator.
Fast Pattern Matching in Compressed Text using Wavelet Tree
Published in IETE Journal of Research, 2018
Surya Prakash Mishra, Rajesh Prasad, Gurmit Singh
The Huffman compression [12] generates variable length code by using greedy approach. It is greedy towards frequency of the symbols presents in the source text. In this approach, codes for symbols with high frequencies will have represented with less number of bits and code for low-frequency symbols will have represented with relatively a higher amount of bits. Huffman trees are used to generate the code. The details about construction of Huffman code is given in [12]. Example 1 illustrates the Huffman code for a text.