On receiver structures for channels having memory

Chang, Ronald Y.; Hancock, John C.

doi:10.1109/tit.1966.1053923

Cited by 121 publications

(51 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…8 The definition is rigorously extended to cases with i = −∞ and/or j = ∞, by assuming P X 0 |z j i to be a regular version of the conditional distribution (cf., e.g., [17]) of X0 given Z Lemma 4 and Claim 2 parallel similar results in sequential decision theory [31] (e.g., in the data compression case, the limiting values of the block and conditional entropies coincide, defining the entropy rate). Their proofs are also standard, but are given in Appendix B for completeness.…”

Section: Universal Optimality: the Stochastic Settingmentioning

confidence: 99%

“…The number in square brackets is the value of k employed, which was obtained using the compressibility heuristic described in Section 6-B. For each combination we also show the residual error rate of the optimal Bayesian distribution-dependent scheme tailored for the specific corresponding value of the pair (p, δ), as implemented by the forward-backward recursions [8,1].…”

Section: -A Binary Symmetric Markov Source Corrupted By a Bscmentioning

confidence: 99%

“…[18] for the many applications); from DNA sequence analysis and processing [45,49,48] to enhancement of facsimile and other binary images; from blind equalization problems to joint source-channel decoding when a discrete source is sent unencoded through a noisy channel [8,21].…”

mentioning

confidence: 99%

“…Indeed, for the case where the states evolve according to a known Markov process and the channel (from state to observation) is known, the above optimum Bayesian scheme can be implemented with reasonable complexity via forward-backward dynamic programming [8,1]. It should be mentioned, however, that even for the simplest among cases where the underlying signal has memory, namely the case of a binary Markov chain observed through a Binary Symmetric Channel (BSC), the bit-error rate of the optimal denoiser is not explicitly known for all values of the transition probability and the channel error rate; only the asymptotic behavior of the bit error rate, as the transition probabilities become small, [28,46] and conditions for the optimality of "singlet decoding" (cf.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Universal discrete denoising: known channel

Weissman

Ordentlich

Seroussi

et al. 2005

IEEE Trans. Inform. Theory

174

View full text Add to dashboard Cite

context models, denoising, discrete filtering, discrete memoryless channels, individual sequences, noisy channels, universal algorithms A discrete denoising algorithm estimates the input sequence to a discrete memoryless channel (DMC) based on the observation of the entire output sequence. For the case in which the DMC is known and the quality of the reconstruction is evaluated with a given single-letter fidelity criterion, we propose a discrete denoising algorithm that does not assume knowledge of statistical properties of the input sequence. Yet, the algorithm is universal in the sense of asymptotically performing as well as the optimum denoiser that knows the input sequence distribution, which is only assumed to be stationary and ergodic. Moreover, the algorithm is universal also in a semi-stochastic setting, in which the input is an individual sequence, and the randomness is due solely to the channel noise. The proposed denoising algorithm is practical, requiring a linear number of regis ter-level operations and sub-linear working storage size relative to the input data length. Abstract A discrete denoising algorithm estimates the input sequence to a discrete memoryless channel (DMC) based on the observation of the entire output sequence. For the case in which the DMC is known and the quality of the reconstruction is evaluated with a given single-letter fidelity criterion, we propose a discrete denoising algorithm that does not assume knowledge of statistical properties of the input sequence. Yet, the algorithm is universal in the sense of asymptotically performing as well as the optimum denoiser that knows the input sequence distribution, which is only assumed to be stationary and ergodic. Moreover, the algorithm is universal also in a semi-stochastic setting, in which the input is an individual sequence, and the randomness is due solely to the channel noise. The proposed denoising algorithm is practical, requiring a linear number of register-level operations and sub-linear working storage size relative to the input data length.

show abstract

Section: Universal Optimality: the Stochastic Settingmentioning

confidence: 99%

Section: -A Binary Symmetric Markov Source Corrupted By a Bscmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Universal discrete denoising: known channel

Weissman

Ordentlich

Seroussi

et al. 2005

IEEE Trans. Inform. Theory

174

View full text Add to dashboard Cite

show abstract

“…The development of an efficient optimization technique [92] 1.0 estimate the model parameters so as 1.0 'match' the observed signal patterns has culminated in the theory of HMM-based signal representation. T!·3 success of this hidden Markov modeling technique has been proven by its application in ecology (e.g., [93]), text analysis (e.g., [94]), coding theory (e.g., [95]) and speech recognition (e.g., [96]). …”

Section: Characterization Of Hidden Markov Modelmentioning

confidence: 99%

Auditory distortion measures for speech coder evaluation

Kabal¹,

De²

1994

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

Adaptive Equalizers

Weso��owski

2003

Wiley Encyclopedia of Telecommunications

View full text Add to dashboard Cite

The paper considers the problem of adaptive equalization. After introduction of the system model and explanation of the phenomenon of intersymbol interference observed on many communication channels, the equalizer structures and algorithms are classified. Then, linear equalizers are considered, concentrating on Zero‐Forcing (ZF), Mean Square Error (MSE) and Least Squares (LS) adaptation algorithms. Subsequently, the choice of a reference signal and fast linear equalizers using periodic test signals are described. Next, the symbol‐spaced equalizer with the fractionally‐spaced equalizer is compared, showing performance superiority of the latter. The operation of the decision feedback equalizer (DFE) is explained in the next section. Subsequently, more complicated than a DFE, non‐linear equalization structures are described such as those based on the Maximum a Posteriori Probability (MAP) symbol‐by‐symbol detection and the Maximum Likelihood (ML) sequence estimation. Equalizers for trellis‐coded signals are also considered. The last part of the paper is devoted to blind adaptive equalizers and their algorithms. The Bussgang algorithms, the algorithms using the second‐order statistics of the received signal and the probabilistic algorithms are shortly characterized.

show abstract

On receiver structures for channels having memory

Cited by 121 publications

References 4 publications

Universal discrete denoising: known channel

Universal discrete denoising: known channel

Auditory distortion measures for speech coder evaluation

Adaptive Equalizers

Contact Info

Product

Resources

About