29 research outputs found
Using hermite bases in studying capacity-achieving distributions over awgn channels
No full textThis paper studies classes of generic deterministic, discrete time, memoryless, and nonlinear additive white Gaussian noise (AWGN) channels. Subject to multiple types of constraints such as the even-moment and compact-support constraints or a mixture, the optimal input is proved to be discrete with finite number of mass points in the vast majority of the cases. Only under the even-moment constraint and for special cases that emulate the average power constrained linear channel, capacity is found to be achieved by an absolutely continuous input. The results are extended to channels where the distortion is generally piecewise nonlinear where the discrete nature of the optimal input is conserved. These results are reached through the development of methodology and tools that are based on standard decompositions in a Hilbert space with the Hermite polynomials as a basis, and it is showcased how these bases are natural candidates for general information-theoretic studies of the capacity of channels affected by AWGN. Intermediately, novel results regarding the output rate of decay of Gaussian channels are derived. Namely, the output probability distribution of any channel subjected to additive Gaussian noise decays necessarily slower than the Gaussian itself. Finally, numerical computations are provided for some sample cases, optimal inputs are determined, and capacity curves are drawn. These results put into question the accuracy of adopting the widely used expression [[1]\over[2]]\log(1+[\ssr SNR]) for computing capacities of Gaussian deterministic channels. © 1963-2012 IEEE.ABBE EA, 2009, P IEEE INT S INF THE, P1644; Abou-Faycal IC, 2001, IEEE T INFORM THEORY, V47, P1290, DOI 10.1109-18.923716; Ash R., 1965, INFORM THEORY; Chung Kai Lai, 2001, COURSE PROBABILITY T; Courant R., 1953, METHOD MATH PHYS, VI; Desurvire E, 2009, CLASSICAL AND QUANTUM INFORMATION THEORY: AN INTRODUCTION FOR THE TELECOM SCIENTIST, P1, DOI 10.1017-CBO9780511803758; Fedoryuk M. V., 2001, HERMITE POLYNOMIALS; Felix V., 1990, P ANN CONV EXH, P183; Gallager R. G., 1968, INFORM THEORY RELIAB; HIRT W, 1988, IEEE T INFORM THEORY, V34, P380; Huber P. J., 1981, ROBUST STAT; Katz M, 2004, IEEE T INFORM THEORY, V50, P2257, DOI 10.1109-TIT.2004.834745; Loeve M., 1955, PROBABILITY THEORY; Lomnitz Y, 2011, IEEE T INFORM THEORY, V57, P7333, DOI 10.1109-TIT.2011.2169130; Luenberger D. G., 1969, OPTIMIZATION VECTOR; Luo C., 2006, THESIS MIT; Ma M. G. J., 2003, CONT MATH, V336, P137; Mitra PP, 2001, NATURE, V411, P1027, DOI 10.1038-35082518; Moran P. A. P., 1968, INTRO PROBABILITY TH; Munkris J. R., 2000, TOPOLOGY; SHANNON CE, 1948, ATandT TECH J, V27, P623; Shiryaev A. N., 1996, PROBABILITY; Silverman H., 1975, COMPLEX VARIABLES; SMITH JG, 1971, INFORM CONTROL, V18, P203, DOI 10.1016-S0019-9958(71)90346-9; Stein E M, 2003, COMPLEX ANAL; Tchamkerten A, 2004, IEEE T INFORM THEORY, V50, P2773, DOI 10.1109-TIT.2004.836662; WALTER GG, 1965, T AM MATH SOC, V116, P492, DOI 10.2307-1994130; Wiener N., 1958, FOURIER INTEGRAL CER; Wittaker E. T., 1962, COURSE MODERN ANAL; Zamir R, 2004, IEEE T INFORM THEORY, V50, P1362, DOI 10.1109-TIT.2004.82815311
Pitch Modification of Speech Signal Using Source Filter Model by Linear Prediction for Prosodic Transformations
No full text
Binary Adaptive Coded Pilot Symbol Assisted Modulation Over Rayleigh Fading Channels Without Feedback
No full textPilot symbol assisted modulation (PSAM) is a standard approach for transceiver design for time-varying channels, with channel estimates obtained from pilot symbols being employed for coherent demodulation of the data symbols. In this paper, we show that PSAM schemes can be improved by adapting the coded modulation strategy at the sender to the quality of the channel measurement at the receiver, without requiring any channel feedback from the receiver. We consider performance in terms of achievable rate for binary signaling schemes. The transmitter employs interleaved codes, with data symbols coded according to their distance from the nearest pilot symbols. Symbols far away from pilot symbols encounter poorer channel measurements at the receiver anal are therefore coded with lower rate codes, while symbols dose to pilot symbols benefit from recent channel measurements and are coded with higher rate codes. The performance benefits from this approach are quantified in the context of binary signaling over time-varying Rayleigh fading channels described by a Gauss-Markov model. The spacing of the pilot symbols is optimized to maximize the mutual information between input and output in this setting. Causal and noncausal channel estimators of varying complexity and delay are considered. It is shown that, by appropriate optimization for the spacing between consecutive pilot symbols, the adaptive coding techniques proposed can improve achievable rate, without any feedback from the receiver to the sender. Moreover, channel estimation based on the two closest pilot symbols is generally close to optimal. © 2005 IEEE.Abou-Faycal IC, 2001, IEEE T INFORM THEORY, V47, P1290, DOI 10.1109-18.923716; Adireddy S, 2002, IEEE T INFORM THEORY, V48, P2338, DOI 10.1109-TIT.2002.800466; ADIREDDY S, 2001, P 2000 INT C AC SPEE; ADIREDDY S, 2000, P INT C ASSP IST TUR, V5, P2541; ADIREDDY S, UNPUB IEEE T INF THE; ADIREDDY S, 2001, INT S INF THEOR JUN; ADIREDDY S, 2001, P C INF SCI SYST MAY; ADIREDDY S, 2000, P C INF SCI SYST MAR; Alouini MS, 1999, IEEE T VEH TECHNOL, V48, P1047, DOI 10.1109-25.775355; MCGEEHAN JP, 1984, IEEE T COMMUN, V32, P81, DOI 10.1109-TCOM.1984.1095960; BDEIR A, 2004, COMM THEOR S IEEE IN, V2, P224; Bello P.A., 1963, IEEE Transactions on Communication Systems, VCS-11, DOI 10.1109-TCOM.1963.1088747; Biglieri E, 1998, IEEE T INFORM THEORY, V44, P2619, DOI 10.1109-18.720551; Caire G, 1999, IEEE T INFORM THEORY, V45, P2007, DOI 10.1109-18.782125; CAVERS JK, 1991, IEEE T VEH TECHNOL, V40, P686, DOI 10.1109-25.108378; CHAN TH, 2003, 41 ANN ALL C COMM CO; COX DC, 1975, IEEE T COMMUN, V23, P1271, DOI 10.1109-TCOM.1975.1092716; FORNEY B, 2002, CODES GRAPHS SYSTEMS; GALLAGER R, 1987, POWER LIMITED CHANNE; Goeckel DL, 1999, IEEE T COMMUN, V47, P844, DOI 10.1109-26.771341; Goldsmith AJ, 1997, IEEE T COMMUN, V45, P1218, DOI 10.1109-26.634685; Goldsmith AJ, 1997, IEEE T INFORM THEORY, V43, P1986, DOI 10.1109-18.641562; Hassibi B, 2003, IEEE T INFORM THEORY, V49, P951, DOI 10.1109-TIT.2003.809594; Hassibi B., 2000, LINEAR ESTIMATION; Ho P, 1996, IEEE T COMMUN, V44, P337, DOI 10.1109-26.486328; HUANG J, 2003, 37 ANN C INF SCI SYS; KHAIRY MM, 1999, P IEEE S COMPUTERS C, V1, P105; KIM YS, 1997, P IEEE INT C COMM IC, P1518; KLEIN TE, 2001, THESIS MIT; Lapidoth A, 1998, IEEE T INFORM THEORY, V44, P2148, DOI 10.1109-18.720535; Lapidoth A., 1999, Proceedings of the 1999 IEEE Information Theory and Communications Workshop (Cat. No. 99EX253), DOI 10.1109-ITCOM.1999.781400; LIM CH, 1998, IEE ELECT LETT, V34, P940; MA X, 2002, P INT C ASSP MAY; MA X, 2001, ICC 2001 HELS FINL J, P1866; Ma XL, 2003, IEEE T SIGNAL PROCES, V51, P1351, DOI 10.1109-TSP.2003.810304; MEDARD M, 2000, P INT S INF THEOR, P413; Medard M., 2000, IEEE T INFORM THEORY, V46, P935; Negi R, 1998, IEEE T CONSUM ELECTR, V44, P1122, DOI 10.1109-30.713244; Ohno S, 2004, IEEE T INFORM THEORY, V50, P2138, DOI 10.1109-TIT.2004.833365; Papoulis A, 2002, PROBABILITY RANDOM V; QIU X, 1999, IEEE T VEH TECHNOL, V48, P1237; RICHTERS JS, 1964, 464 MIT RLE; SAMPEI S, 1994, IEICE T COMMUN, VE77B, P1096; SAMPEI S, 1993, IEEE T VEH TECHNOL, V42, P137, DOI 10.1109-25.211451; Tang XY, 1999, IEEE T COMMUN, V47, P1856; Torrance J. M., 1995, P RAD REC ASS SYST C, P36; Torrance JM, 1999, IEEE T VEH TECHNOL, V48, P1527, DOI 10.1109-25.790528; Torrance JM, 1999, IEEE T VEH TECHNOL, V48, P1237, DOI 10.1109-25.775372; TORRANCE JM, 1997, ELECTRON LETT, V32, P1218; VERDU S, 1990, IEEE T INFORM THEORY, V36, P1019, DOI 10.1109-18.57201; Viswanathan H, 1999, IEEE T INFORM THEORY, V45, P761, DOI 10.1109-18.749027; WEBB WT, 1995, IEEE T COMMUN, V43, P2223, DOI 10.1109-26.392965; WELBURN L, 1999, IEEE T COMMUN, V47, P1527; Wolfowitz J., 1978, CODING THEOREMS INFO; Wong C. H., 2000, P 51 IEEE VTS VEHICU, V3, P2044; Wong CH, 2000, IEEE T COMMUN, V48, P367, DOI 10.1109-26.837037; XU Y, 1999, MILITARY COMM C P, V1, P8628252
Redesigning deep neural networks: Bridging game theory and statistical physics
No full textInternational audienc
A Micro-Moment System for Domestic Energy Efficiency Analysis
No full textDomestic user behavior is a crucial factor guiding overall power consumption, necessitating the development of systems that analyze and help shape energy-efficient behavior. Therefore, the most important step in the process is the collection and understanding of highly detailed domestic consumption data. This article presents an appliance-based energy data collection and analysis system for energy efficiency applications. It leverages the concept of micro-moments, which are short-timed and energy-based events that form the overall energy behavior of the end user. The system comprises sensing modules for recording energy consumption, occupancy, temperature, humidity, and luminosity storing recordings on a database server. Sensing parameters were tested in terms of connection stability and measurement accuracy. A four-week contextual appliance-level dataset has been collected from research cubicles. Collected data were also classified into corresponding micro-moments with a variety of classifiers including ensemble decision trees and deep learning, achieving high stability and accuracy of 99%. Further, the micro-moment usage efficiency is calculated to quantify the efficiency of usage at the appliance level. 2021 IEEE.Manuscript received November 4, 2019; revised March 23, 2020; accepted April 22, 2020. Date of publication June 9, 2020; date of current version March 9, 2021. This article was made possible by National Priorities Research Program (NPRP) grant No. 10-0130-170288 from the Qatar National Research Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of the authors. (Corresponding author: Abdullah Alsalemi.) Abdullah Alsalemi, Yassine Himeur, and Faycal Bensaali are with the Department of Electrical Engineering, Qatar University, 2713 Doha, Qatar (e-mail: [email protected]; [email protected]; [email protected]).Scopu
Delay-efficient GOP size control algorithm in Wyner-Ziv Video Coding
No full textIn Wyner-Ziv Video Coding (WZVC) (a special form of distributed video coding) the heavy work of motion estimation is shifted to the decoder. Unlike conventional video systems, such a scheme allowed for new emerging technologies that require low-complexity encoders. A key factor to the performance of WZVC is the quality of the Side Information (SI) present at the decoder which heavily depends on the key frame separation. For this reason, the Group Of Pictures (GOP) size plays a significant role in the coding efficiency of the system. In this paper, we focus on the control of GOP size in transform-domain WZVC with feedback channel. We present a novel algorithm that performs online selection of the GOP size relying on past system behavior. Our proposed algorithm incurs minimal delays to the system making it suitable for real-time applications. ©2009 IEEE.Aaron A., 2006, P IEEE INT C IM PROC; Aaron A., 2004, P SPIE INT C VIS COM; Ascenso J., 2003, P IEEE INT C IM PROC; Girod B, 2005, P IEEE, V93, P71, DOI 10.1109-JPROC.2004.839619; ITU-T I. JTC1, 109181 ISOIEC ITUT; Puri R., 2002, P ALL C COMM CONTR C; Slepian J. D., 1973, IEEE T INFORM THEORY, VIT-19, P471, DOI DOI 10.1109-TIT.1973.1055037; WYNER AD, 1976, IEEE T INFORM THEORY, V22, P1, DOI 10.1109-TIT.1976.1055508; Yaacoub C., 2009, INT J DIGITAL MULTIM, V20090
