{"title":"Support Vector Machine Re-scoring of Hidden Markov Models","authors":"Alba Sloin, D. Burshtein","doi":"10.1109/EEEI.2006.321107","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321107","url":null,"abstract":"We present a method that uses a set of maximum-likelihood (ML) trained discrete HMM models as a baseline system, and an SVM training scheme to re-score the results of the baseline HMMs. It turns out that the re-scoring model can be represented as an un-normalized HMM. We refer to these models as pseudo-HMMs. The pseudo-HMMs are in fact a generalization of standard HMMs, and by proper discriminative training they can result in performance improvement compared to standard HMMs. We consider two SVM training algorithms. The first corresponds to the one against all method. The second corresponds to the one class transformation training method. The one class training algorithm can be extended to an iterative algorithm, similar to segmental K-means. In this case the final output of the algorithm is a single set of pseudo-HMMs. Although they are not normalized, this set of pseudo-HMMs can be used in the standard recognition procedure (the Viterbi recognizer), as if they were plain HMMs. We use an isolated noisy digit recognition task to demonstrate that SVM re-scoring of HMMs typically reduces the error rate significantly compared to standard ML training.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123516469","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}
{"title":"Cover's test of rationality revisited: Computability aspects of hypothesis testing","authors":"Amir Leshem","doi":"10.1109/EEEI.2006.321057","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321057","url":null,"abstract":"In this paper we discuss computability aspects of hypothesis testing. We describe two main results. First we determine the type of sets that admit a weak decision procedure. Surprisingly some non-computable sets admit a computable weak decision procedure. This strengthens results of Cover and Putnam. We then apply the notion of weak decision procedure to the testing of the physical Church-Turing thesis. While our first theorem states that there are non-computable sets that admit weak decision procedures, we are able to show that no weak decision procedure can help us to decide that a physical device is capable of computing non Turing computable functions or that a physical constant encodes the bits of a non-computable real. This has strong implications on the validity of physical theories entailing the failure of the physical Church-Turing thesis.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121998518","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}
{"title":"One-Time Signatures Revisited: Practical Fast Signatures Using Fractal Merkle Tree Traversal","authors":"D. Naor, Amir Shenhav, A. Wool","doi":"10.1109/EEEI.2006.321066","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321066","url":null,"abstract":"One-time signatures have been known for more than two decades, and have been studied mainly due to their theoretical value. Recent works motivated us to examine the practical use of one-time signatures in high-performance applications. In this paper we describe FMTseq - a signature scheme that merges recent improvements in hash tree traversal into Merkle's onetime signature scheme. Implementation results show that the scheme provides a signature speed of up to 35 times faster than a 2048-bit RSA signature scheme, for about one million signatures, and a signature size of only a few kilobytes. We provide an analysis of practical parameter selection for the scheme, and improvements that can be applied in more specific scenarios.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130134843","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}
{"title":"Algebraic Representation of DC-DC Converters and Symbolic Method of their Analysis","authors":"Y. Berkovich, A. Shenkman, Ioinovici, B. Axelrod","doi":"10.1109/EEEI.2006.321080","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321080","url":null,"abstract":"The algebraic representation of idealized DC-DC PWM converters covering various structures and their connection is given. The algebra is constructed on the basis of hyper-complex numbers - hypernions and allows the analysis of the structures of the converters by using of a symbolical method. The possibility of the calculation of a discontinuous current and soft switching is also considered. Such an abstract description of various schemas allows reducing their variety to several basic structures and formalizes a problem of the synthesis of the converters with the desired parameters. The numerical solution of the algebraic expressions is easily carried out with the help of the standard mathematical programs MATHCAD or MATLAB.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130979895","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}
{"title":"Investigation of Time Domain Design of Digital Controllers for PWM Converters","authors":"M. Peretz, S. Ben-Yaakov","doi":"10.1109/EEEI.2006.321088","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321088","url":null,"abstract":"A design method for digital controller of PWM DC-DC converters was developed, tested by simulations and verified experimentally. The proposed approach is based on the fact that the closed loop response of a digitally controlled system is largely determined by the first few samples of the compensator. This concept is used to fit a digital PID template to the desired response. The proposed controller design method is carried out in the time domain and thus, bypasses errors related to continuous o discrete domain transformation and discretization. A digital PID controller for a Buck type converter was implemented experimentally on a TMS320LF2407 DSP core. The measured closed loop attributes were 3.5KHz bandwidth and phase margin of 40°. Good agreement was found between the design goals and the experimentally determined response.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128762266","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}
{"title":"Universal Scanning of Mixing Random Fields and the Performance of the Peano-Hilbert Scan","authors":"A. Cohen, N. Merhav, T. Weissman","doi":"10.1109/EEEI.2006.321083","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321083","url":null,"abstract":"We investigate the problem of scanning and prediction (\"scandiction\", for short) of multidimensional data arrays. This problem arises in several aspects of image and video processing, such as predictive coding, where an image is compressed by coding the prediction error sequence resulting from scandicting it. Specifically, given a strongly mixing random field, we show that there exists a scandiction scheme which is independent of the field's distribution, yet almost surely asymptotically achieves the same performance as if this distribution was known. We then discuss the scenario where the Peano-Hilbert scanning order is used, accompanied by an optimal predictor, and derive a bound on the excess loss compared to optimal finite state scandiction, which is valid for any individual image and any bounded loss function.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121839708","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}
{"title":"Minimum BER criteria for DFE in Optical Communication System","authors":"G. Katz, D. Sadot","doi":"10.1109/EEEI.2006.321137","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321137","url":null,"abstract":"As the link distance or the data rate increase intersymbol interference (ISI) caused by chromatic dispersion increases and ISI mitigation becomes an important issue. The demand for reducing cost per gigabytes is one of the main reasons for using electronic equalizers which are very cost-effective. The common and useful optimization criterion to estimate the DFE coefficients is the minimum mean square error (MMSE). However, the MMSE solution does not achieve optimal performances, and the minimum BER (MBER) is the ultimate performances criterion. Here, we derive an analytical expression for the probability of error with DFE and examine the optimal optimization criterion, MBER for DFE in optical communication system. A BER performance comparison analysis shows that the MBER optimization criterion introduces improvements over the conventional minimum MSE optimization criterion.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131561741","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}
{"title":"Dirty Paper Coding with a Finite Input Alphabet","authors":"T. Gariby, U. Erez, S. Shamai","doi":"10.1109/EEEI.2006.321110","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321110","url":null,"abstract":"We study a dirty paper channel model where the input is constrained to belong to a PAM constellation. In particular, we provide lower bounds on the capacity as well as explicit coding schemes for the binary-input dirty-paper channel. We examine the case of causal as well as non-causal side information.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133168524","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}
{"title":"A First Approach Towards a High-Speed PCI-based Data Acquisition Card for Industrial Applications","authors":"Pablo Costi-Kowolik, Roberto Gutiérrez-Mazón","doi":"10.1109/EEEI.2006.321142","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321142","url":null,"abstract":"The Peripheral Component Interconnect (PCI) bus is the fastest communication protocol available for off-the-shelf Personal Computers (PCs). It is a suitable base for designing cost-effective Data Acquisition Cards (DAQs). These can achieve a high data throughput, making them ideal for implementation into different industrial environments. In this document, a PCI prototype card is presented, which is the first step for the development of a high-performance low-cost DAQ card. Moreover, with this card, the empirical PCI DMA transfer rate of two host PCs was measured, offering real empirical results of the PCI bus performance, instead of the well known theoretical 132 Mbytes/s (33 MHz, 32-bit). This theoretical value is not a reliable reference for designing a PCI DAQ card as, in fact, this data transfer bandwidth cannot be achieved.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114442892","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}
{"title":"An effective technique for simultaneous interconnect channel delay and noise reduction in nanometer VLSI design","authors":"K. Moiseev, S. Wimer, A. Kolodny","doi":"10.1109/EEEI.2006.321062","DOIUrl":"https://doi.org/10.1109/EEEI.2006.321062","url":null,"abstract":"Capacitive coupling is the primary source of noise in nanometer technology digital CMOS VLSI circuits. It becomes worse with technology scaling. The interconnect capacitive crosstalk noise can be characterized by two parameters: peak noise voltage, and delay uncertainty. Delay uncertainty optimization can be seen as a subset of interconnect delay optimization. This paper addresses the problem of ordering and sizing parallel wires in a single metal layer within an interconnect channel of a given width, such that cross-capacitances are optimally shared for simultaneous noise and delay minimization. Using an Elmore delay model including cross capacitances for a bundle of wires and well-known crosstalk models, we show that \"symmetric hill\" wire ordering according to the strength of signal drivers, which is known to optimize channel timing characteristics, can be used also for minimizing channel noise metrics. Examples using state-of-the-art circuits in 65-nanometer technology are analyzed and discussed.","PeriodicalId":142814,"journal":{"name":"2006 IEEE 24th Convention of Electrical & Electronics Engineers in Israel","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122059571","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}