Background
Type: Article

Optimal multiband pulse shaping and improved power spectrum estimation using stopbands information

Journal: IEEE Signal Processing Letters (10709908)Year: 2013Volume: 20Issue: Pages: 1006 - 1009
DOI:10.1109/LSP.2013.2278622Language: English

Abstract

In this letter, we derive formulas for optimal discrete-time pulse shaping for communications in multiband channels. Furthermore, we propose a technique for extending a short signal segment for the purpose of power spectral estimation using the available information about the signal stopbands. The later technique can be also used for minimizing the energy leakage into stopbands of a signal, e.g., as in windowing for orthogonal frequency division modulation (OFDM). Simulation results show that when the pulse duration is long enough, the stopband energy can be practically reduced to zero using the proposed technique. The proposed power spectral estimation technique can improve the resolution and reduce the energy leakage from signal passbands into adjacent stopbands. Moreover, it can reduce the out-of-band power leakage significantly when used for windowing in OFDM. © 1994-2012 IEEE.


Author Keywords

Bandstop filter designDMTOFDMpower spectral estimationpulse shapingVDSL2windowingYule-Walker AR method

Other Keywords

OptimizationOrthogonal frequency division multiplexingPower spectrumPulse shapingPulse shaping circuitsSpectrum analysisAR methodDMTPower spectral estimationVDSL2windowingInformation use