Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513905
B. Talbot, L. Talbot
We present a memory-binding density transformation as a means of improving the performance of entropy coders acting on memory sources.
我们提出了一种内存绑定密度变换，作为一种提高熵编码器作用于内存源的性能的方法。
{"title":"Source coding with a reversible memory-binding probability density transformation","authors":"B. Talbot, L. Talbot","doi":"10.1109/WITS.1994.513905","DOIUrl":"https://doi.org/10.1109/WITS.1994.513905","url":null,"abstract":"We present a memory-binding density transformation as a means of improving the performance of entropy coders acting on memory sources.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126655802","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513920
Z. Li, Shu Zhao Zou
Summary form only given. Consider a linear model, y/sub i/=x'/sub i//spl beta/+e/sub i/, i=1,2,..., x'/sub i/s are p(/spl ges/1) dimension known vectors and /spl beta/(/spl isin/R/spl deg/) is an unknown parametric vector and e/sub i/ are assumed i.i.d.r.v.'s from a common unknown density function f(x) with med (e/sub i/)=0. Based on LAD (least absolute deviations) estimator /spl beta//spl tilde/ of /spl beta/, we propose a nonparametric method to estimate unknown f(x). A kernel estimator f/spl tilde//sub n/(x) is obtained. Large sample properties of f/spl tilde//sub n/(x) are studied. Some computational examples are also given.
{"title":"Nonparametric kernel estimation for error density","authors":"Z. Li, Shu Zhao Zou","doi":"10.1109/WITS.1994.513920","DOIUrl":"https://doi.org/10.1109/WITS.1994.513920","url":null,"abstract":"Summary form only given. Consider a linear model, y/sub i/=x'/sub i//spl beta/+e/sub i/, i=1,2,..., x'/sub i/s are p(/spl ges/1) dimension known vectors and /spl beta/(/spl isin/R/spl deg/) is an unknown parametric vector and e/sub i/ are assumed i.i.d.r.v.'s from a common unknown density function f(x) with med (e/sub i/)=0. Based on LAD (least absolute deviations) estimator /spl beta//spl tilde/ of /spl beta/, we propose a nonparametric method to estimate unknown f(x). A kernel estimator f/spl tilde//sub n/(x) is obtained. Large sample properties of f/spl tilde//sub n/(x) are studied. Some computational examples are also given.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122261214","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513889
J. Craig
In many data communication receivers up/down counters are used as a critical part of the processing to determine whether the symbol timing and/or carrier phase tracking phase-locked loops are in-lock or out-of-lock, and it is necessary to calculate the various probabilities for true and false indications of in-lock or out-of-lock. A random walk along a line (which is viewed as a Markov chain) is an exact model of an up/down counter. The random walk has N states, and in this application one end is a partially reflecting barrier, and the other end is an absorbing barrier or sink. Previously published analyses have focused on finding the average time to make a declaration and its variance. The author concentrates on finding the probabilities of making a true or a false declaration within a certain number of symbol intervals or within a certain length of time.
{"title":"Markov chains and random walks in data communication receivers","authors":"J. Craig","doi":"10.1109/WITS.1994.513889","DOIUrl":"https://doi.org/10.1109/WITS.1994.513889","url":null,"abstract":"In many data communication receivers up/down counters are used as a critical part of the processing to determine whether the symbol timing and/or carrier phase tracking phase-locked loops are in-lock or out-of-lock, and it is necessary to calculate the various probabilities for true and false indications of in-lock or out-of-lock. A random walk along a line (which is viewed as a Markov chain) is an exact model of an up/down counter. The random walk has N states, and in this application one end is a partially reflecting barrier, and the other end is an absorbing barrier or sink. Previously published analyses have focused on finding the average time to make a declaration and its variance. The author concentrates on finding the probabilities of making a true or a false declaration within a certain number of symbol intervals or within a certain length of time.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122928575","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513886
G. Davis, S. Mallat
To compute the optimal expansion of signals in redundant dictionary of waveforms is an NP complete problem. We introduce a greedy-algorithm, called matching pursuit, that performs a sub-optimal expansion. This algorithm can be interpreted as a shape-gain multistage vector quantization. The waveforms are chosen iteratively in order to best match the signal structures. Matching pursuits are general procedures used to compute adaptive signal representations. Applications to speech and image processing with dictionaries of Gabor functions are shown, in particular for the noise removal.
{"title":"Wavelet vector quantization with matching pursuit","authors":"G. Davis, S. Mallat","doi":"10.1109/WITS.1994.513886","DOIUrl":"https://doi.org/10.1109/WITS.1994.513886","url":null,"abstract":"To compute the optimal expansion of signals in redundant dictionary of waveforms is an NP complete problem. We introduce a greedy-algorithm, called matching pursuit, that performs a sub-optimal expansion. This algorithm can be interpreted as a shape-gain multistage vector quantization. The waveforms are chosen iteratively in order to best match the signal structures. Matching pursuits are general procedures used to compute adaptive signal representations. Applications to speech and image processing with dictionaries of Gabor functions are shown, in particular for the noise removal.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121512530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513900
M. Pawlak, U. Stadtmuller
The problem of recovering bandlimited signals from noisy data is considered. Whittaker-Shannon cardinal expansions based estimates involving sampling windows and truncation of higher frequencies are introduced. Weak and strong pointwise convergence properties of the proposed estimates are derived.
{"title":"Nonparametric estimation of a class of smooth functions","authors":"M. Pawlak, U. Stadtmuller","doi":"10.1109/WITS.1994.513900","DOIUrl":"https://doi.org/10.1109/WITS.1994.513900","url":null,"abstract":"The problem of recovering bandlimited signals from noisy data is considered. Whittaker-Shannon cardinal expansions based estimates involving sampling windows and truncation of higher frequencies are introduced. Weak and strong pointwise convergence properties of the proposed estimates are derived.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124369341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513899
L. Nelson, H. Poor
This work describes an EM-based approach to multi-user detection that treats the signals of interfering users as hidden data. We consider a new algorithm based on the space-alternating generalized EM (SAGE) algorithm appropriate for estimation of discrete random parameters, and we use it to derive rapidly-convergent nearly optimum multi-user receivers.
{"title":"EM and SAGE algorithms for multi-user detection","authors":"L. Nelson, H. Poor","doi":"10.1109/WITS.1994.513899","DOIUrl":"https://doi.org/10.1109/WITS.1994.513899","url":null,"abstract":"This work describes an EM-based approach to multi-user detection that treats the signals of interfering users as hidden data. We consider a new algorithm based on the space-alternating generalized EM (SAGE) algorithm appropriate for estimation of discrete random parameters, and we use it to derive rapidly-convergent nearly optimum multi-user receivers.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122104764","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513906
Z. Tian
Projection pursuit autoregression (MPPAR) and projection pursuit moving average (MPPMA) with multivariate polynomials as ridge functions in both cases are proposed in this paper. The L/sub 2/-convergence of the methods is proved. This paper also proposes two new algorithms for MPPAR and MPPMA. By using the methods, we establish the mathematical models about the Wolfer sunspot data and Canadian lynx data.
{"title":"Projection pursuit autoregression and projection pursuit moving average","authors":"Z. Tian","doi":"10.1109/WITS.1994.513906","DOIUrl":"https://doi.org/10.1109/WITS.1994.513906","url":null,"abstract":"Projection pursuit autoregression (MPPAR) and projection pursuit moving average (MPPMA) with multivariate polynomials as ridge functions in both cases are proposed in this paper. The L/sub 2/-convergence of the methods is proved. This paper also proposes two new algorithms for MPPAR and MPPMA. By using the methods, we establish the mathematical models about the Wolfer sunspot data and Canadian lynx data.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123089322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513892
R. L. Fry
A maximum entropy formulation leads to a neural network which is factorable in both form and function into individual neurons corresponding to the Hopfield neural model. A maximized mutual information criterion dictates the optimal learning methodology using locally available information.
{"title":"Maximized mutual information using macrocanonical probability distributions","authors":"R. L. Fry","doi":"10.1109/WITS.1994.513892","DOIUrl":"https://doi.org/10.1109/WITS.1994.513892","url":null,"abstract":"A maximum entropy formulation leads to a neural network which is factorable in both form and function into individual neurons corresponding to the Hopfield neural model. A maximized mutual information criterion dictates the optimal learning methodology using locally available information.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114471946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513860
A. Nobel
We establish the Bayes risk consistency of an unsupervised greedy-growing algorithm that produces tree-structured classifiers from labeled training vectors. The algorithm employs a composite splitting criterion equal to a weighted sum of Bayes risk and Euclidean distortion.
{"title":"Greedy growing of tree-structured classification rules using a composite splitting criterion","authors":"A. Nobel","doi":"10.1109/WITS.1994.513860","DOIUrl":"https://doi.org/10.1109/WITS.1994.513860","url":null,"abstract":"We establish the Bayes risk consistency of an unsupervised greedy-growing algorithm that produces tree-structured classifiers from labeled training vectors. The algorithm employs a composite splitting criterion equal to a weighted sum of Bayes risk and Euclidean distortion.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129264810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-10-27DOI: 10.1109/WITS.1994.513916
J. Kogan
Sequential properties of the Viterbi algorithm are studied based on a renewal sequence of the most informative stopping times which can be explicitly found during the Viterbi recognition of the most likeliest hidden Markovian state-sequence.
{"title":"The most informative stopping times for Viterbi algorithm: sequential properties","authors":"J. Kogan","doi":"10.1109/WITS.1994.513916","DOIUrl":"https://doi.org/10.1109/WITS.1994.513916","url":null,"abstract":"Sequential properties of the Viterbi algorithm are studied based on a renewal sequence of the most informative stopping times which can be explicitly found during the Viterbi recognition of the most likeliest hidden Markovian state-sequence.","PeriodicalId":423518,"journal":{"name":"Proceedings of 1994 Workshop on Information Theory and Statistics","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129278426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：