Subband-based blind signal separationfor noisy speech recognition

Park, Hyung‐Min; Jung, Ho‐Young; Lee, Te-Won; Lee, Sang Yup

doi:10.1049/el:19991358

Cited by 50 publications

(20 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In another FD approach applied to speech recognition [26], the energy of the observables in the FD is considered, but the unmixing matrix varies for different frequency bins. As a result, conventional approaches suffer from permutation problems and gain issues as discussed in [24].…”

Section: B Fd-ica Unitmentioning

confidence: 99%

Two-Stage Approach for Detection and Reduction of Motion Artifacts in Photoplethysmographic Data

Krishnan

Natarajan

Warren

2010

IEEE Trans. Biomed. Eng.

235

127

View full text Add to dashboard Cite

Abstract-Corruption of photopleythysmograms (PPGs) by motion artifacts has been a serious obstacle to the reliable use of pulse oximeters for real-time, continuous state-of-health monitoring. In this paper, we propose an automated, two-stage PPG data processing method to minimize the effects of motion artifacts. The technique is based on our prior work related to motion artifact detection (stage 1) [R. Krishnan, B. Natarajan, and S. Warren, "Analysis and detection of motion artifacts in photoplethysmographic data using higher order statistics," in Proc.

show abstract

Section: B Fd-ica Unitmentioning

confidence: 99%

Two-Stage Approach for Detection and Reduction of Motion Artifacts in Photoplethysmographic Data

Krishnan

Natarajan

Warren

2010

IEEE Trans. Biomed. Eng.

235

127

View full text Add to dashboard Cite

show abstract

“…Yen and Zhao, 14 Park et al , 15 and Erten and Salam 16 have focused on signal separation for speech recognition, but their methods could find use in HF applications. Yen and Zhao proposed a method for adaptive de-correlation filtering (ADF).…”

Section: Ham and Faourmentioning

confidence: 99%

Improved techniques for monitoring the HF spectrum

Giesbrecht

Clarke²,

Abbott

2004

SPIE Proceedings

View full text Add to dashboard Cite

A critical review of contemporary papers on modulation recognition, signal separation, and Single Station Location (SSL) is described in the context of High-Frequency (HF) radio-communications. High-frequency communications is undergoing resurgence despite advances in long-range satellite communication systems. Defense agencies are using the HF spectrum for backup communications as well as for spectrum surveillance applications. Spectrum management organizations are monitoring the HF spectrum to control and enforce licensing. This type of activity usually requires a system that is able to determine the location of a source of transmissions, separate valid signals from interferers and noise, and characterize signals-of-interest (SOI). The immediate aim is to show that commercial-of-the-shelf (COTS) equipment can be used to locate HF transmission sources, enhance SOIs and reject interference, and recognize signal types. The described work on single-station-location (SSL), signal separation, and modulation recognition is contributing to these goals. This paper describes the overall objectives and some of the disadvantages and benefits of various schemes for single-station-location (SSL), signal separation, and modulation recognition. It also proposes new approaches that may relieve shortcomings of existing methods -including selection of benchmarks or modulations for various transmission scenarios and propagation modes, and use of multiple digital receivers or compression techniques to improve modulation recognition, signal separation, and location of HF emitters.

show abstract

“…BSS is the separation of a set of unknown mixed signals (Cardoso, 1998;Shalvi and Weinstein, 1990). It has been extensively applied to different signal sources, such as observed speech signals (Park et al, 1999), overlapped images (Cichocki et al, 2002), and noise-corrupted signals (Tu et al, 2001). ICA is one technique to perform BSS, which is used to find unknown latent components of observed multivariate data.…”

Section: Introductionmentioning

confidence: 99%