N. S. Jayant and P. Noll. Digital Coding of Waveforms: Principles and Applications to Speech and Video. Prentice Hall, 1984.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... 3 22 xmc[9] 3 60 Nc[3] 7 23 xmc[10] 3 61 bc[3] 2 24 xmc[11] 3 62 Mc[3] 2 25 xmc[12] 3 63 xmaxc[3] 6 26 Nc[1] 7 64 xmc[39] 3 27 bc[1] 2 65 xmc[40] 3 28 Mc[1] 2 66 xmc[41] 3 29 xmaxc[1] 6 67 xmc[42] 3 30 xmc[13] 3 68 xmc[43] 3 31 xmc[14] 3 69 xmc[44] 3 32 xmc[15] 3 70 xmc[45] 3 33 xmc[16] 3 71 xmc[46] 3 34 xmc[17] 3 72 xmc[47] 3 35 xmc[18] 3 73 xmc[48] 3 36 xmc[19] 3 74 xmc[49] 3 37 xmc[20] 3 75 xmc[50] 3 38 xmc[21] 3 76 xmc[51] 3 Table 2: Ordering of GSM variables Octet Bit 0 Bit 1 Bit 2 Bit 3 Bit 4 Bit 5 Bit 6 Bit 7 0 1 1 0 1 LARc0.0 LARc0.1 LARc0.2 LARc0.3 1 LARc0.4 ....

....count, sampling rate, and bit rate. 4.5.14 PCMA and PCMU PCMA and PCMU are specified in ITU T Recommendation G.711. Audio data is encoded as eight bits per sample, after logarithmic scaling. PCMU denotes mu law scaling, PCMA A law scaling. A detailed description is given by Jayant and Noll [16]. Each G.711 octet shall be octet al..igned in an RTP packet. The sign bit of each G.711 octet shall correspond to the most significant bit of the octet in the RTP packet (i.e. assuming the G.711 samples are handled as octets on the host machine, the sign bit shall be the most significant bit of ....

Jayant, N. and P. Noll, Digital Coding of Waveforms---Principles and Applications to Speech and Video. Englewood Cli#s, New Jersey: Prentice-Hall, 1984.

....practical interest. In particular, the study of such coding schemes within the individual sequence framework which we consider here is especially relevant for many image compression situations in which there is no natural statistical model for the data. The reader is referred, e.g. to [7, 9] and the references therein for a comprehensive account of the theory and practice of DPCM based coding schemes. Following is a laconic description intended primarily to introduce notation for later use. A DPCM delayless source code of order s is an element of F , which is fully characterized ....

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewod Cliffs, NJ: Prentice-Hall, 1984.

....is inspired by the derivations in Section 2, where bit rate e#cient coding is based on encoder decoder pairs that have an internal state that is kept synchronized. This intuition leads to the design of a novel predictive coding scheme, similar to standard Di#erential Pulse Code Modulation (DPCM) [3], where the predictor incorporates all available knowledge of the dynamics of the system. In our case, since the goal is for both encoder and decoder to track the state of the system, our predictor is constructed using the reconstructed state data to estimate the expected state of the system at ....

N. S. Jayant and P. Noll. Digital Coding of Waveforms: Principles and Applications to Speech and Video. Prentice-Hall, Englewood Cli#s, New Jersey, 1984.

....optimality. c(x) USQ USQ c ] x) A b A e Fig. 3. Block diagram of the asymptotically optimal scalable coder V. SIMULATION RESULTS A. law corapanding for a raeraoryless Laplace source The performance of the AOS scheme is compared experimentally to the CS scheme for law companding [9]. The CS and the AOS schemes are implemented as described in Section III and Section IV, respectively. The input is a memoryless Laplace source with zero mean and variance rr 2 = 100. Consideration of this process is motivated by the observation that speech, image and video signals possess ....

N.S. Jayant and P. Noll, Digital Coding of Waveforms: principles and applications to speech and video, ch. 4, pp. 115-220. Prentice-Hall, 1984.

....bitstreams. 1 Introduction The traditional problem in video coding for the past several decades has been that of compression: describe the signal with as few bits as possible. The signal is treated as a single waveform, with compression employing techniques such as transform and or prediction [8, 10] and resulting in a lossy representation. In several cases, however, it is beneficial to segment the original signal into multiple parts, and handle each one independently. Such an approach was originally applied to speech using the so called sub band coding approach [10] which partitions the ....

....and or prediction [8, 10] and resulting in a lossy representation. In several cases, however, it is beneficial to segment the original signal into multiple parts, and handle each one independently. Such an approach was originally applied to speech using the so called sub band coding approach [10], which partitions the signal into multiple Presented in part at the IEEE Int l Conf. on Image Processsing, Austin, Texas, November 1994. frequency bands. The primary motivation is that, since the human aural system perceives the various frequency bands in different ways, one could apply ....

N. S. Jayant and P. Noll. Digital Coding of Waveforms: Principles and Applications to Speech and Video. Prentice Hall, Englewood Cliffs, New Jersey, 1984.

.... vector quantization where (#a 1 , 2 , M ) are jointly determined by the vector (a 1 , a 2 , aM ) For a random source A and a quantizer Q, let A : Q(A) If the number of bins is fixed, say N , to minimize the expected mean square error E (AQ (A) Lloyd Max algorithm [24, 15] gives the optimal solution where smaller bins are allocated to more popular values. However, if we fix the expected mean square error, say D, and would like to minimize the entropy of A, surprisingly, the uniform quantizer outperforms the quantizer obtained by Lloyd Max algorithm; in fact it ....

Nuggehally S. Jayant and Peter Noll. Digital coding of waveforms: principles and applications to speech and video. Englewood Cli#s, New Jersey, 1984.

....density S ee (e ) coe x . With these assumptions, one considers the possibility of improvement of the noise level by pre filtering the input process before quantization and post filtering it after the quantization 16 with the inverse of the original filter (see Fig. 8) It is known [28] that the best prefilter F (e ) is given by jF (e )j (6:1) and that the phase of F (e ) is arbitrary. This is commonly referred as half whitening since the power spectral density of the output of F (e ) which is flatter than S xx (e ) but not completely flat. The ....

N.S. Jayant and P. Noll, Digital coding of waveforms : principles and applications to speech and video, Englewood Cliffs, NJ: Prentice-Hall, 1984.

....pdf. The sources are: Gaussian or normal, having pdf 1 x 2. f= X) x e , Laplacian or two sided exponential, having pdf 1 e vlXl; and gamma or broad tailed, having pdf f= X) e , where a = x 2. These sources, often used in modeling signals encountered in speech and image coding practice [5][10] can be regarded as instances of the generalized Gaussian density as defined in [11] The pdf s of the three sources are shown in Figure 1.2. Note that the gamma pdf grows arbitrarily large near the origin. 1.3 Scalar Quantization A K level scalar quantizer Q is a nonlinear, noninvertible ....

....uniform quantization [23] 25] consideration was re stricted to the case fi = 0. Uniform quantizers of this type are often termed midriser, as their characteristic staircase functions rise at the origin. At the other extreme are uniform quantizers with fi = A 2, which are often termed midtread[5]. Clearly, for the symmetric pdf sources of interest, a midriser uniform quantizer can never have an entropy of less than one bit per sample, since the maximum possible probability of a quantization region in that case is 1 2. 2.3 MSE Suboptimality of Uniform Quantization for Laplacian Sources ....

[Article contains additional citation context not shown here]

Jayant, N.S., and Noll, P., Digital Coding of Waveforms -- Principles and Applications to Speech and Video, Prentice-Hall, Englewood Cliffs, New Jersey, 1984.

....compacting orthogonal block transform, where the basis vectors are allowed to extend beyond the block boundaries. For an orthogonal block transform, diagonalization of the covariance matrix oc curs as a by product of maximizing energy compaction subject to the constraint of constant total energy [62]. Therefore, to the extent that an orthogonal subband decomposition achieves energy compaction, pixels are uncorrelated across subbands. Approximate uncorrelatedness within subbands can be easily understood in the frequency domain. The spectra of natural images are often modeled as smooth, except ....

....not imply approximate independence of the subband pixels, as previously noted in Chapter 1. Subband coding of images Subband and subband pyramid decomposition are often used in lossy compression. Subband based lossy compression is termed subband coding. Subband coding was first applied to speech [62] in the seventies and extended to images in the mid eighties [143] Much of the underlying motivation and 17 theory carries over directly from an older technique called transform coding [56] As mentioned above, a block transform is a special case of subband decomposition. The basic idea behind ....

[Article contains additional citation context not shown here]

Nuggehally S. Jayant and Peter Noll. Digital Coding of Waveforms: Principles and Appli- cations to Speech and Video. Prentice-Hall, Englewood Cliffs, N J, 1984.

....3 4 5 6 7 2 3 4 5 6 7 8 8 2 27 24 19 3 4 5 8 9 7 6 1 j k 22 Fig. 2. Example of index assignment for N 1 = N 2 = 8 and N = 34. MDSQ systems, whose outputs are transmitted over the two channels. Another way to remove the correlations in the source is to use linear prediction [6]. Before describing the multiple description system that uses linear prediction, we describe a single channel system used often in source coding to decorrelate the source before quantizing. The block diagram of such a system is shown in Figure 3 [6] This is the di erential pulse code modulation ....

....in the source is to use linear prediction [6] Before describing the multiple description system that uses linear prediction, we describe a single channel system used often in source coding to decorrelate the source before quantizing. The block diagram of such a system is shown in Figure 3 [6]. This is the di erential pulse code modulation (DPCM) system, which consists of a linear predictor and a quantizer. We assume that fX n : n an integerg is a discrete time real, zero mean stationary, and ergodic random process with autocorrelation RXX (k) E[X n X n k ] We also de ne RXX (0) ....

N. S. Jayant and P. Noll, Digital coding of waveforms: Principles and applications to speech and video. Englewood Clis, New Jersey: Prentice Hall Inc, 1984, chap. 4-6, pp. 115-350.

....ensembles for such systems, which we now explore. In the process, we obtain additional insights into the design, performance evaluation, and implementation of QIM embedding methods, particularly those of low complexity. A convenient structure to consider is that of so called dithered quantizers [29], 30] which have the property that the quantization cells and reconstruction points of any given quantizer in the ensemble are shifted versions of the quantization cells and reconstruction points of any other quantizer in the ensemble. In nonwatermarking contexts, the shifts typically correspond ....

.... squared minimum distance (26) 10 A uniform distribution for the dither sequence implies that the quantization error is statistically independent of the host signal and leads to fewer false contours, both of which are generally desirable properties from a perceptual viewpoint [29]. CHEN AND WORNELL: QUANTIZATION INDEX MODULATION 1429 Fig. 4. Dither modulation with uniform quantization step sizes. a quantity that can be used to characterize the achievable performance of QIM realizations more generally, as we will develop. B. Spread Transform Dither Modulation One special ....

[Article contains additional citation context not shown here]

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewood Cliffs, NJ: PrenticeHall, 1984.

....the event that the received sequence containing the bits B i representing V i FAILS, or if B i is spread over two received sequences, that one or both of these FAILS, and where NF i is the NO FAIL event, i.e. the complement of F i . By the usual argument of combined source channel coding (cf. [5], pp. 179 181) E[ V i V i ) 2 F i ] D s,i D c,i , 3) where D s,i = D s,i (R s,i ) is the quantizer distortion described To be presented at ICASSP, May 2001 earlier and D c,i is the channel distortion defined by D c,i = j,j =1 N s,i (y i,j y i,j ) 2 P i (j) P i (w i,j w ....

N.S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video, Englewood Cliffs, NJ: Prentice-Hall, 1984.

.... Of course, Shannon showed that no source channel coding system could have better performance, e.g. less distortion in the reconstructed data samples, than the best tandem system [1] However, over the years there has been considerable interest in joint source channel coding, for example [2] [16], with the motivation of attaining comparable performance with less complexity or delay than tandem source channel coding. On the other hand, little quantitative evidence for this claim has appeared in the literature. In this paper, we seek such evidence by quantitatively analyzing the ....

N. S. Jayant & P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Audio. Englewood Cliffs, NJ: Prentice-Hall, 1984. 20

....error, 30 DPerc =f#coding activity; masking phenomenon#; 24) where coding activty is influenced directly by the quantization error. For a sake of easyness, we consider now the first part of this problem: the quantization error. This rate distortion problem has been studied in detail [Jay84, Ram94, Chi97b, Lin98, Chi97a, Lin96a] but the generally used relations are not useful in this work, since they are obtained after a very complex computing process. We observe that (see Fig. 20) for each frame type, the quantization error captured by the MSE metric is linear with relation to ....

Jayant N. S. and Noll P. Digital coding of waveforms: principles and applications to speech and video. Prentice-Hall, 1984.

....and video coding, where a number of standards are based on them. In speech coding they form the basis of ITU G.721, 722, 723, and 726, and in video coding they form the basis of the interframe coding schemes standardized in the MPEG and H.26X series. Comprehensive discussions may be found in books [265], 374] 196] 424] 50] 458] and survey papers [264] 198] Though decorrelation was an early motivation for predictive quantization, the most common view at present is that the primary role of the predictor is to reduce the variance of the variable to be scalar quantized. This view stems ....

.... similar in form to that of the source that its operational distortion rate function is smaller than that of the original source by, approximately, the ratio of the variance of the source to that of the prediction error, a quantity that is often called a prediction gain [350] 396] 482] 397] [265]. Analyses of this form usually claim that under high resolution conditions the distribution of the prediction error approaches that of the error when predictions are based on past source samples rather than past reproductions. However, it is not clear that the accuracy of this approximation ....

[Article contains additional citation context not shown here]

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video, PrenticeHall, Englewood Cli#s, NJ, 1984.

....error, 30 DPerc = f(coding activity; masking phenomenon) 24) where coding activty is influenced directly by the quantization error. For a sake of easyness, we consider now the first part of this problem: the quantization error. This rate distortion problem has been studied in detail [Jay84, Ram94, Chi97b, Lin98, Chi97a, Lin96a] but the generally used relations are not useful in this work, since they are obtained after a very complex computing process. We observe that (see Fig. 20) for each frame type, the quantization error captured by the MSE metric is linear with relation to ....

Jayant N. S. and Noll P. Digital coding of waveforms: principles and applications to speech and video. Prentice-Hall, 1984.

....suboptimal) index assignments with low implementation complexities. Various families of recursively defined index assignments have been extensively studied in the past, including the well known Natural Binary Code (NBC) Folded Binary Code (FBC) Two s Complement Code (TCC) and Gray Code (GC) [29]. Huang [30] 32] computed distortion formulas for the Natural Binary Code and the Gray Code for uniform scalar quantizers and uniform scalar sources. He asserted that the Natural Binary Code was optimal among all possible index assignments for the uniform source [31] This was proven by Crimmins ....

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewood Cliffs, NJ: PrenticeHall, 1984.

....descriptions are received at the receiver. Our study shows that compression errors are highly nonlinear and complex and cannot be compensated by a linear process. We then describe in detail our proposed artificial neural network (ANN) architecture. A. Quantization Errors in MDC Since coding gain [12] is proportional to autocorrelation ae of a video source and ae is reduced after pixel based interleaving, we expect the PSNR of an MDC system to be smaller than that of the original system at the same bit rate. This phenomenon is illustrated in Table I that compares ae and PSNR of a horizontally ....

N. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video, Englewood Cliffs, Prentice-Hall, 1984.

....To speed searches, the last occupied slot is tracked through the member last variable. When a session member is about to be deleted, the array is cleansed of any pointers to it. 10.5 Audio Encoding The most common audio encoding is law companding. The law transfer characteristic is given by [12] c(x) xm log(1 x=xm ) 1 where c(x) is the coded value corresponding to input x, with the maximum absolute value of x given by xm . has a value of 255. The companding gain is given by = log(1 ) This characteristic is approximated by a piecewise linear function to ease translation ....

N. S. Jayant and P. Noll, Digital Coding of Waveforms---Principles and Applications to Speech and Video. Englewood Cliffs, New Jersey: Prentice-Hall, 1984.

No context found.

N. S. Jayant and P. Noll, "Digital Coding of Waveforms: Principles and Applications to Speech and Video", Prentice Hall, 1984.

....compression tools fail in the case of digital audio data. A preprocessing stage, which eliminates the statistical dependencies within the signal, leads to an almost uncorrelated source which is easier to code. Such decorrelation can either be achieved by linear prediction or by linear transforms [1]. The samples of the difference signal or the transform coefficients have to be quantized which causes inevitable errors in the output signal. Therefore, lossless coding has to be considered as a combination of conventional lossy coding and an additional transmission of the coding error. Fig. 1 ....

....disc [5] The categories are based on the according SQAM sections. Maximum adaptive block length: M = 4096. LTAC version 1.61. 4 Lossy but subjectively transparent audio coding Speech and Audio Coding First proposals to reduce wideband audio coding rates have followed those for speech coding [1]. Speech and audio coding are similar in that in both cases quality is based on the properties of human auditory perception. However, speech can be coded very efficiently because a speech production model is available, whereas nothing similar exists for audio signals (see Figure 3) Fig. 3. ....

[Article contains additional citation context not shown here]

N. S. Jayant and P. Noll, "Digital Coding of Waveforms: Principles and Applications to Speech and Video," Prentice Hall, 1984.

....will be of increasing importance to provide economically acceptable overall bit rates depending on network characteristics, potential applications, and coding goals such as high quality and low complexity. First proposals to reduce wideband audio coding rates have followed those for speech coding [1]. Differences between audio and speech signals are manifold, CD format. Other sampling rates are 32 kHz , 48 kHz, and 96 kHz. Amplitude resolution goes up to 32 b sample. however: audio coding implies higher sampling rates, better amplitude resolution, higher dynamic range, larger variations in ....

....classify the quality of coders on an N point quality scale. The final result of such tests is an averaged judgement called the mean opinion score (MOS) Two 5 point adjectival grading scales are in use, one for signal quality, and the other one for signal impairment, and an associated numbering [1]. The 5 point ITU R impairment scale of Table 2 is extremely useful if coders with only small impairments have to be graded. The advantage of MOS values is that different impairment factors can be assessed simultaneously and that even small impairments can be graded. On the negative side, ....

N. S. Jayant and P. Noll, "Digital Coding of Waveforms: Principles and Applications to Speech and Video," Prentice Hall, 1984.

No context found.

N. S. Jayant and P. Noll. Digital Coding of Waveforms: Principles and Applications to Speech and Video. Prentice Hall, 1984.

No context found.

Jayant, N. S. and P. Noll: 1984, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Prentice Hall.

No context found.

N. S. Jayant and P. Noll. Digital Coding of Waveforms: Principles and Applications to Speech and Video. Prentice Hall, 1984.

No context found.

N. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewood Cliffs, NJ: Prentice-Hall, 1984.

No context found.

N. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewood Cliffs, NJ: Prentice-Hall, 1984.

No context found.

N. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Upper Saddle River, NJ: Prentice-Hall, 1984.

No context found.

N. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewood Cliffs, NJ: Prentice-Hall, 1984.

No context found.

N. S. Jayant and P. Noll. Digital Coding of Waveforms: Principles and Applications to Speech and Video. Prentice Hall, 1984.

No context found.

Jayant, N. S. and P. Noll: 1984, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Prentice Hall.

No context found.

N. S. Jayant and P. Noll. Digital Coding of Waveforms: Principles and Applications to Speech and Video (Prentice-Hall, 1984)

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Upper Saddle River, NJ: Prentice-Hall, 1984. 171

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video, Prentice Hall, Englewood Cliffs, NJ, 1984.

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewood Cliffs, NJ: PrenticeHall, 1984.

No context found.

Jayant, N. and P. Noll, Digital Coding of Waveforms--Principles and Applications to Speech and Video Englewood Cliffs, New Jersey: Prentice-Hall, 1984.

No context found.

N. Jayant and P. Noll, "Digital coding of waveform principles and applications to speech and audio", Englewood Cliffs, NJ, Prentice-Hall, 1984.

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms-- Principles and Applications to Speech and Video Englewood Cliffs, New Jersey: Prentice-Hall, 1984.

No context found.

N.S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video, Englewood Cliffs, NJ: Prentice-Hall, 1984.

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewood Cliffs, New Jersey: Prentice Hall, 1984.

No context found.

N. S. Jayant and P. Noll, Digital Coding for Waveforms: Principles and Applications to Speech and Video. Prentice-Hall, 1984.

No context found.

N. S. Jayant and P. Noll. Digital Coding of Waveforms: Principles and Applications to Speech and Video. Prentice Hall, Englewood Cliffs, NJ, 1984.

No context found.

Jayant, N.S. and P. Noll. Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewood Cliffs, NJ: Prentice-Hall, 1984.

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video, Prentice Hall, Englewood Cliffs, NJ, 1984.

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video, Prentice Hall, Englewood Cli#s, NJ, 1984.

No context found.

N. S. Jayant, P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video, Prentice Hall, Englewood Cli#s, NJ, 1984. 23

No context found.

Jayant, N.S. and Noll, P. "Digital Coding of Waveforms: Principles and Applications to speech and video". Englewood Cliffs, NJ, Prentice-Hall, 1984.

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video. Englewood Cliffs, NJ: Prentice-Hall, 1984.

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms : Principles and Applications to Speech and Video, Prentice-Hall, Englewood Cliffs, NJ, 1984.

No context found.

N. S. Jayant and P. Noll, Digital Coding of Waveforms: Principles and Applications to Speech and Video, Prentice-Hall, 1984.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC