Abstract: Session ITT-10

ITT-10.1	Adaptive Space-Frequency Rake Receivers for WCDMA Christopher Brunner, Martin Haardt (Siemens AG, Mobile Networks, OEN MN P 36), Josef A Nossek (Institute for Network Theory and Circuit Design, Munich Univ. of Technology) Adaptive space-frequency rake receivers use maximum ratio combining and multi-user interference suppression to obtain a considerable increase in performance in DS-CDMA systems such as WCDMA. To this end, the signal-plus-interference-and-noise and the interference-plus-noise space-time covariance matrices are estimated. The computational complexity is reduced significantly by transforming the covariance matrices into the space-frequency domain and by omitting noisy space-frequency bins. The optimum weight vector for symbol decisions is the "largest" generalized eigenvector of the resulting matrix pencil. By iteratively updating the optimum weight vector slot by slot, real-time applicability becomes feasable while the fast fading is still tracked. The performance and the computational complexity depend on the number of space-frequency bins, antenna elements, and iterations. Therefore, the performance can easily be scaled with respect to the available computational power.
ITT-10.2	A Practical Method to Reduce a Number of Reference Signals for the ANC System Masaichi Akiho (Alpine Electronics, Inc.), Miki Haseyama, Hideo Kitajima (Hokkaido University) In this paper, we propose a practical method to reduce a number of reference signals for the active noise cancellation (ANC) system and the filter characteristics to generate the reduced number of reference signals, which maintain the original value of the coherence function. This method finds the number of independent noise sources and provides the filter characteristics based on SVD (singular value decomposition) of the power spectrum matrix of the reference signals. Then, we also use the multiple coherence function analysis to select dominant components in the reference signals. The method contributes greatly in reducing the number of reference signals for the ANC system that uses the large number of reference signals. We also discuss the characteristics of the filters that synthesis the new set of reference signals. And an experimental test is performed to confirm the theory.
ITT-10.3	DSP-BASED SOLUTION FOR AMBIENT NOISE REDUCTION IN MOBILE PHONES Jamil CHAOUI, Sebastien De Gregorio, Guillaume Gallissian, Yves Masse (Texas Instruments France) Using a mobile handset in noisy environments makes the far-end speech difficult to be perceived from the near-end user perspective. As he is surrounded by a loud background noise, handset user needs to be highly concentrated to understand the far-end speaker conversation. Ambient noise reduction algorithms provide an efficient solution to build a silent zone between the handset loudspeaker and the user ear, thus improving speech understanding for the near-end speaker. This paper presents a full description of an innovative ambient noise reduction system developed on a TI TMS320C54x DSP. This contribution is combining both theoretical and experimental considerations, raising potential issues that may be encountered when implementing these applications on a real system. It will be shown that the advanced architecture of the TI TMS320C54x DSP makes the ambient noise reduction application possible to be executed in the same CPU performing in the same time wireless digital cellular baseband processing.
ITT-10.4	Cancellation of Siren Noise from Two Way Voice Communications Inside Emergency Vehicles Robert S Sherratt (The University of Reading), David M Townsend (Fulcrum Systems Ltd), Chris G Guy (The University of Reading) Sirens' used by police, fire and paramedic vehicles have been designed so that they can be heard over large distances, but unfortunately the siren noise enters the vehicle and corrupts intelligibility of voice communications from the emergency vehicle to the control room. Often the siren needs to be turned off to enable the control room to hear what is being said. This paper discusses a siren noise filter system that is capable of removing the siren noise picked up by the two-way radio microphone inside the vehicle. The removal of the siren noise improves the response time for emergency vehicles and thus save lives. To date, the system has been trialed within a fire tender in a non-emergency situation, with good results. A demonstration of the siren filter to various sirens can be heard by accessing http://www.elec.reading.ac.uk/rss.html and following the link to 'siren cancellation'.
ITT-10.5	System Identification Using Orthogonal Functions and Application to Acoustic Echo Cancellation Idil Haskan, Aysin Ertuzun (Bogazici University Istanbul/Turkey) In this paper, a new Laguerre domain adaptive filter algorithm which will be refered to as Laguerre domain adaptive filter II (LDAF II) has been proposed. The performance of the adaptive filtering algorithms is simulated for acoustic echo cancellation application. The performances of the algorithm are specified using various quantities. All the results of the work for different performance quantities, are presented with several graphics and they are compared with Legendre functions based adaptive filter (LFB ADF) [1] and LMS adaptive filter (LMS ADF).
ITT-10.6	Adaptive Acoustic Echo Cancellation Based on FIR and IIR Filter Banks Thorsten Ansahl, Imre Varga, Ingrid Kremmer, Wen Xu (Siemens AG, Mobile Phones Dept.) In this paper we investigate various subband AEC systems in real handsfree situation, including FIR and IIR analysis and synthesis QMF filterbanks in polyphase structure. The adaptation in the subbands is performed by the Affine Projection Algorithm in comparison to the NLMS algorithm. The IIR filters are superior to FIR filters in the sense that they lead to low signal delay and sharp frequency separation. Furthermore the computational complexity is greatly reduced by the use of IIR filters. The results show that splitting the signal into more subbands has advantages. Both wideband and narrowband speech signals have been evaluated.
ITT-10.7	A Robust Speech Detection Algorithm for Speech Activated Hands-Free Applications Duanpei Wu, M. Tanaka, R. Chen, L. Olorenshaw, M. Amador, X. Menendez-Pidal (Spoken Language Technology, Sony US Research Laboratories) This paper describes a novel noise robust speech detection algorithm that can operate reliably in severe car noisy conditions. High performance has been obtained with the following techniques: (1) noise suppression based on principal component analysis for pre-processing, (2) robust endpoint detection using dynamic parameters [1] and (3) speech verification using periodicity of voiced signals with harmonic enhancement. Noise suppression improves the SNR as compared with nonlinear spectrum subtraction by about 20 dB. This makes the endpoint detection operate reliably in SNRs down to -10 dB. In car environments, road bump noises are problematic for speech detectors causing mis-detection errors. Speech verification helps to remove these errors. This technology is being used in Sony car navigation products.

< ITT-9

Last Update:  February 4, 1999         Ingo Höntsch