AES 123rd Convention - Where Audio Comes Alive
Home Visitors Exhibitors Press Students Authors
registration
detailed calendar
paper sessions
workshops
broadcast sessions
tutorials
master classes
live sound seminars
symposium
students/career
historical
special events
technical tours
exhibitor seminars
training sessions
standards
technical committees
awards ceremony

AES New York 2007
Paper Session P10

P10 - Automotive Audio and Amplifiers

Saturday, October 6, 12:30 pm — 3:30 pm
Chair: Richard Stroud, Stroud Audio - Kokomo, IN, USA

P10-1 Improved Stereo Imaging in AutomobilesMichael Smithers, Dolby Laboratories - Sydney, NSW, Australia
A significant challenge in the automobile listening environment is the predominance of off-axis listening positions. This leads to audible artifacts including comb filtering and indeterminate stereo imaging; both in traditional stereo and more recent multichannel loudspeaker configurations. This paper discusses the problem of off-axis listening as well as methods to improve stereo imaging in a symmetric manner using all-pass FIR and IIR filters. This paper also discusses a more efficient IIR filter design that achieves similar performance to previous filter designs. Use of these filters results in stable, virtual sources in front of off-axis listeners.
Convention Paper 7223 (Purchase now)

P10-2 A Listening Test System for Automotive Audio—Part 3: Comparison of Attribute Ratings Made in a Vehicle with Those Made Using an Auralization SystemPatrick Hegarty, Sylvain Choisel, Søren Bech, Bang & Olufsen a/s - Struer, Denmark
A system has been developed to allow listening tests of car audio sound systems to be conducted over headphones. The system employs dynamic binaural technology to capture and reproduce elements of an in-car soundfield. An experiment, a follow-up to a previous work, to validate the system is described. Seven trained listeners were asked to rate a range of stimuli in a car as well as over headphones for 15 elicited attributes. Analysis of variance was used to compare ratings from the two hardware setups. Results show the ratings for spatial attributes to be preserved while differences exist for some timbral and temporal attributes.
Convention Paper 7224 (Purchase now)

P10-3 A Listening Test System for Automotive Audio - Part 4: Comparison of Attribute Ratings Made by Expert and Non-Expert ListenersSylvain Choisel, Patrick Hegarty, Bang & Olufsen a/s - Struer, Denmark; Flemming Christensen, Benjamin Pedersen, Wolfgang Ellermeier, Jody Ghani, Wookeun Song, Aalborg University - Aalborg, Denmark
A series of experiments was conducted in order to validate an experimental procedure to perform listening tests on car audio systems in a simulation of the car environment in a laboratory, using binaural synthesis with head-tracking. Seven experts and 40 non-expert listeners rated a range of stimuli for 15 sound-quality attributes developed by the experts. This paper presents a comparison between the attribute ratings from the two groups of participants. Overall preference of the non-experts was also measured using direct ratings as well as indirect scaling based on paired comparisons. The results of both methods are compared.
Convention Paper 7225 (Purchase now)

P10-4 The Application of Direct Digital Feedback for Amplifier System ControlCraig Bell, David Jones, Robert Watts, Zetex Semiconductors - Oldham, UK
An effective feedback topology is clearly a beneficial requirement for a well performing digital amplifier. The ability to cancel corrupting influences such as power supply ripple and unmatched components is necessary for good sonic performance. Additional benefits derive from the fact that the feedback information is processed in the digital domain. Current delivered into the loudspeaker load can be inferred. The amplifier acts as a voltage source, the value of which is derived from the recorded source material. The current delivered into the loudspeaker is also clearly influenced by the load impedance, which varies with frequency and other factors. This paper describes the ability of the system to measure current and derive loudspeaker impedance and the actual delivered power and goes on to illustrate the applications in real systems.
Convention Paper 7226 (Purchase now)

P10-5 Generation of Variable Frequency Digital PWMPallab Midya, Freescale Semiconductor Inc. - Lake Zurich, IL, USA
Digital audio amplifiers convert digital PCM to digital PWM to be amplified by a power stage. This paper introduces a method to generate a quantized duty ratio digital PWM with a switching frequency over a 20 percent range to mitigate EMI issues. The method is able to compensate for the variation in switching frequency such that the SNR in the audio band is comparable to fixed frequency PWM. To obtain good rejection of the noise introduced by the variation of the PWM frequency higher order noise shapers are used. This paper describes in detail the algorithm for a fourth order noise shaper. Using this method dynamic range in excess of 120 dB unweighted over a 20 kHz bandwidth is achieved.
Convention Paper 7227 (Purchase now)

P10-6 Recursive Natural Sampling for Digital PWMPallab Midya, Bill Roeckner, Theresa Paulo, Freescale Semiconductor Inc. - Lake Zurich, IL, USA
This paper presents a highly accurate and computationally efficient method for digital-domain computation of naturally sampled digital pulse width modulation (PWM) signals. This method is used in a switching digital audio amplifier. The method is scalable for performance versus calculation complexity. Using a second order version of the algorithm with no iteration, intermodulation linearity of better than 113 dB is obtained with a full scale input at 19 kHz and 20 kHz. Matlab simulation and measured results from a digital amplifier implemented with this algorithm are presented. Overall system performance is not limited by the accuracy of the natural sampling method.
Convention Paper 7228 (Purchase now)


Last Updated: 20070820, mei


.