Publications
Estimation of handset nonlinearity with application to speaker recognition
September 1, 2000
Journal Article
Published in:
IEEE Trans. Speech Audio Process., Vol. 8, No. 5, September 2000, pp. 567-584.
Topic:
R&D area:
Summary
A method is described for estimating telephone handset nonlinearity by matching the spectral magnitude of the distorted signal to the output of a nonlinear channel model, driven by an undistorted reference. This "magnitude-only" representation allows the model to directly match unwanted speech formants that arise over nonlinear channels and that are a potential source of degradation in speaker and speech recognition algorithms. As such, the method is particularly suited to algorithms that use only spectral magnitude information. The distortion model consists of a memoryless nonlinearity sandwiched between two finite-length linear filters. Nonlinearities considered include arbitrary finite-order polynomials and parametric sigmoidal functionals derived from a carbon-button handset model. Minimization of a mean-squared spectral magnitude distance with respect to model parameters relies on iterative estimation via a gradient descent technique. Initial work has demonstrated the importance of addressing handset nonlinearity, in addition to linear distortion, in speaker recognition over telephone channels. A nonlinear handset "mapping" applied to training or testing data to reduce mismatch between different types of handset microphone outputs, improves speaker verification performance relative to linear compensation only. Finally, a method is proposed to merge the mapper strategy with a method of likelihood score normalization (hnorm) for further mismatch reduction and speaker verification performance improvement.
Summary
A method is described for estimating telephone handset nonlinearity by matching the spectral magnitude of the distorted signal to the output of a nonlinear channel model, driven by an undistorted reference. This "magnitude-only" representation allows the model to directly match unwanted speech formants that arise over nonlinear channels and that...
READ MORE
Improving RUC-1 wind estimates by incorporating near-real-time aircraft reports
August 1, 2000
Journal Article
Published in:
Weather For., Vol. 15, No. 4, August 2000, pp. 447-460.
Topic:
R&D area:
R&D group:
Summary
A verification study of wind accuracy is presented for wind nowcasts generated by augmenting Rapid Update Cycle (RUC) wind forecasts with near-real-time aircraft reports using the Integrated Terminal Weather System (ITWS) gridded winds algorithm. Aircraft wind reports collected between the end of the RUC data collection interval and the time each RUC forecasts is valid are available for use in augmenting the RUC wind forecast to form a wind nowcast. The 60-km resolution, hourly RUC-1 wind forecasts are used. ITWS-based nowcast wind errors and RUC forecast wind errors are examined statistically over a 1-yr dataset. The addition of the recent aircraft reports significantly reduces the rms vector error and the 90th percentile vector error. Also reduced is the number of hours of sustained large errors and the correlation among errors. The errors increase with increasing wind speed, in part due to an underestimation of wind speed that increases with increasing wind speed. The errors in the augmented wind fields decrease with increasing numbers of Aircraft Communications Addressing and Reporting System reports. Different types of weather are also seen to influence wind field accuracy.
Summary
A verification study of wind accuracy is presented for wind nowcasts generated by augmenting Rapid Update Cycle (RUC) wind forecasts with near-real-time aircraft reports using the Integrated Terminal Weather System (ITWS) gridded winds algorithm. Aircraft wind reports collected between the end of the RUC data collection interval and the time...
READ MORE
MEMs microswitch arrays for reconfigurable distributed microwave components
July 16, 2000
Conference Paper
Published in:
IEEE Antennas and Propagation Society Int. Symp. Digest, Vol. 2, 16-21 July 2000, pp. 587-591.
Topic:
Summary
A revolutionary device technology and circuit concept is introduced for a new class of reconfigurable microwave circuits and antennas. The underlying mechanism is a compact MEMs cantilever microswitch that is arrayed in two-dimensions. The switches have the ability to be individually actuated. By constructing distributed circuit components from an array, the individual addressability of the microswitch provides the means to reconfigure the circuit trace and, thus, provides the ability to either fine-tune or completely reconfigure the circuit element's behavior. Device performance can be reconfigured over a decade in bandwidth in the nominal frequency range of 1 to 100 GHz. In addition, other circuit-element attributes can be reconfigured such as instantaneous bandwidth, impedance, and polarization (for antennas). This will enable the development of next-generation communication, radar and surveillance systems with agiIity to reconfigure operation for diverse operating bands, modes, power levels, and waveforms.
Summary
A revolutionary device technology and circuit concept is introduced for a new class of reconfigurable microwave circuits and antennas. The underlying mechanism is a compact MEMs cantilever microswitch that is arrayed in two-dimensions. The switches have the ability to be individually actuated. By constructing distributed circuit components from an array...
READ MORE
Wind prediction accuracy for air traffic management decision support tools
June 13, 2000
Conference Paper
Published in:
Proc. 3rd Int. Air Traffic Management R&R Seminar, 13-16 June 2000, pp. 1-9.
Summary
Air traffic automation depends on accurate trajectory predictions. Flight tests show that wind errors are a large source of error. Wind-field accuracy is sufficient on average, but large errors occasionally exist that cause significant errors in trajectory-prediction. A year long study was conducted to better understand the wind-prediction errors, to establish metrics for quantifying large errors, and to validate two approaches to improve wind prediction accuracy. Three methods are discussed for quantifying large errors: percentage of point errors that exceed 10 m/s, probability distribution of point errors, and the number of hourly time periods with a high number of large errors. The baseline wind-prediction system evaluated for this study is the Rapid Update Cycle (RUC). Two approaches to improving the original RUC wind predictions are examined. The first approach is to enhance RUC in terms of increased model resolution, enhancement of the model physics, and increased observational input data. The second method is to augment the RUC output, in near-real time, through an optimal-interpolation scheme that incorporates the latest aircraft reports received since the last RUC update. Both approaches are shown to greatly reduce the occurrence of large wind errors.
Summary
Air traffic automation depends on accurate trajectory predictions. Flight tests show that wind errors are a large source of error. Wind-field accuracy is sufficient on average, but large errors occasionally exist that cause significant errors in trajectory-prediction. A year long study was conducted to better understand the wind-prediction errors, to...
READ MORE
Speaker recognition using G.729 speech codec parameters
June 5, 2000
Conference Paper
Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, ICASSP, Vol. II, 5-9 June 2000, pp. 1089-1092.
Topic:
R&D area:
Summary
Experiments in Gaussian-mixture-model speaker recognition from mel-filter bank energies (MFBs) of the G.729 codec all-pole spectral envelope, showed significant performance loss relative to the standard mel-cepstral coefficients of G.729 synthesized (coded) speech. In this paper, we investigate two approaches to recover speaker recognition performance from G.729 parameters, rather than deriving cepstra from MFBs of an all-pole spectrum. Specifically, the G.729 LSFs are converted to "direct" cepstral coefficients for which there exists a one-to-one correspondence with the LSFs. The G.729 residual is also considered; in particular, appending G.729 pitch as a single parameter to the direct cepstral coefficients gives further performance gain. The second nonparametric approach uses the original MFB paradigm, but adds harmonic striations to the G.729 all-pole spectral envelope. Although obtaining considerable performance gains with these methods, we have yet to match the performance of G.729 synthesized speech, motivating the need for representing additional fine structure of the G.729 residual.
Summary
Experiments in Gaussian-mixture-model speaker recognition from mel-filter bank energies (MFBs) of the G.729 codec all-pole spectral envelope, showed significant performance loss relative to the standard mel-cepstral coefficients of G.729 synthesized (coded) speech. In this paper, we investigate two approaches to recover speaker recognition performance from G.729 parameters, rather than deriving...
READ MORE
Weather radar development and application programs
June 1, 2000
Journal Article
Published in:
Lincoln Laboratory Journal, Vol. 12, No. 2, 2000, pp. 367-382.
Topic:
R&D area:
R&D group:
Summary
Weather phenomena such as microburst wind shear and severe thunderstorms are major concerns to the aviation industry. A number of significant airplane accidents have resulted from wind-shear encounters during takeoff and landing, and thunderstorms are a major contributor to airplane delay. Providing fully automated and timely warnings of these phenomena by radar is challenging because it requires rapid and accurate analysis of the three-dimensional storm structure in the presence of intense ground-clutter returns. For the last two decades, Lincoln Laboratory has been tackling this challenge by applying advanced radar signal- and image-processing techniques to weather radar data. The resulting technology is being deployed in radar-based weather information systems at major airports throughout the United States. We first discuss the salient meteorological factors that contribute to the formation of microburst wind shear, then we provide some general background on the use of pulse-Doppler radar for weather detection. We describe two specific Lincoln Laboratory programs that have generated deployed systems: the Terminal Doppler Weather Radar (TDWR) and the ASR-9 Weather Systems Processor (WSP). The article concludes with a discussion of future detection strategies that emphasizes the fusion of weather radar data by the Integrated Terminal Weather System (ITWS).
Summary
Weather phenomena such as microburst wind shear and severe thunderstorms are major concerns to the aviation industry. A number of significant airplane accidents have resulted from wind-shear encounters during takeoff and landing, and thunderstorms are a major contributor to airplane delay. Providing fully automated and timely warnings of these phenomena...
READ MORE
The NIST Speaker Recognition Evaluation - overview, methodology, systems, results, perspective
June 1, 2000
Journal Article
Published in:
Speech Commun., Vol. 31, Nos. 2-3, June 2000, pp. 225-254.
Topic:
R&D area:
Summary
This paper, based on three presentations made in 1998 at the RLA2C Workshop in Avignon, discusses the evaluation of speaker recognition systems from several perspectives. A general discussion of the speaker recognition task and the challenges and issues involved in its evaluation is offered. The NIST evaluations in this area and specifically the 1998 evaluation, its objectives, protocols and test data, are described. The algorithms used by the systems that were developed for this evaluation are summarized, compared and contrasted. Overall performance results of this evaluation are presented by means of detection error trade-off (DET) curves. These show the performance trade-off of missed detections and false alarms for each system and the effects on performance of training condition, test segment duration, the speakers' sex and the match or mismatch of training and test handsets. Several factors that were found to have an impact on performance, including pitch frequency, handset type and noise, are discussed and DET curves showing their effects are presented. The paper concludes with some perspective on the history of this technology and where it may be going.
Summary
This paper, based on three presentations made in 1998 at the RLA2C Workshop in Avignon, discusses the evaluation of speaker recognition systems from several perspectives. A general discussion of the speaker recognition task and the challenges and issues involved in its evaluation is offered. The NIST evaluations in this area...
READ MORE
The design and validation of the ITWS synthetic sensor data generator
April 12, 2000
Project Report
Published in:
MIT Lincoln Laboratory Report ATC-289
Summary
The Integrated Terminal Weather System (ITWS) is an aviation safety and air traffic management decision support system that acquires data from various FAA and NWS sensors and generates a number of products for dissemination to FAA facilities managing air traffic in the terminal area. The development and demonstrations of ITWS have been conducted over a multi-year period at several major airports (Memphis, TN, Orlando, FL, Dallas, TX, and New York, NY). Although there are many meteorological events observed at these four airports, the experimental test data sets obtained will not fully suffice for ITWS qualification testing because of limitations in the severity of the weather events and because of the sensor configurations available at these locations. This report describes the design and validation of the Synthetic Data Generator (SDG), which is a tool to provide a production ITWS system with meteorologically consistent scenarios and full ITWS sensor configurations that will create maximal computational loads that can be expected when the system is deployed. Also, the SDG will be a tool for ongoing ITWS maintenance and support. As such, the SDG will complement the extensive experimental data sets collected at the four ITWS demonstration sites. The SDG is designed to specify parameters for a collection of meteorological models describing the various weather phenomena, their motion, appearance, and growth/decay. The software creates several three-dimensional (3D) grids of reflectivity and velocity at each time-step. Finally, the SDG generates sensor (i.e., TDWR, NEXRAD, ASR-9) data by applying the model for each specific sensor's measurements to the 3D grids. The validation of the meteorological model and the sensor model data have been accomplished using a display tool and by assessing results numerically.
Summary
The Integrated Terminal Weather System (ITWS) is an aviation safety and air traffic management decision support system that acquires data from various FAA and NWS sensors and generates a number of products for dissemination to FAA facilities managing air traffic in the terminal area. The development and demonstrations of ITWS...
READ MORE
Spectral beam combining of a broad-stripe diode laser array in an external cavity
March 15, 2000
Journal Article
Published in:
Opt. Lett., Vol. 25, No. 6, 15 March 2000, pp. 405-407.
Summary
The outputs from an 11-element, linear diode laser array with broad stripes have been beam combined into a single beam with a beam quality of ~20X diffraction limited in the plane of the junction. This beam combining was achieved by use of a common external cavity containing a grating, which simultaneously forces each array element to operate at a different, but controlled, wavelength and forces the beams from all the elements to overlap and propagate in the same direction. The power in the combined beam was 50% of the output from the bare laser array.
Summary
The outputs from an 11-element, linear diode laser array with broad stripes have been beam combined into a single beam with a beam quality of ~20X diffraction limited in the plane of the junction. This beam combining was achieved by use of a common external cavity containing a grating, which...
READ MORE
An enhanced bandwidth design technique for electromagnetically coupled microstrip antennas
February 1, 2000
Journal Article
Published in:
IEEE Trans. Antennas Propag., Vol. 48, No. 2, February 2000, pp. 161-164.
Topic:
Summary
This paper describes a method of enhancing the bandwidth of two different electromagnetically coupled microstrip antennas by utilization of a tuning stub. An approximate theory and equations are developed to demonstrate the potential bandwidth improvement and required stub impedance characteristics. A novel dual-stub design is presented that achieves better characteristics than a conventional quarter wavelength open-end stub. As examples, the bandwidth (VSWR < 2) of a conventional proximity-coupled microstrip antenna is increased from 4.8 to 8.4% and the bandwidth of a stacked aperture-coupled microstrip antenna is increased from 27.5 to 34.5% using this technique.
Summary
This paper describes a method of enhancing the bandwidth of two different electromagnetically coupled microstrip antennas by utilization of a tuning stub. An approximate theory and equations are developed to demonstrate the potential bandwidth improvement and required stub impedance characteristics. A novel dual-stub design is presented that achieves better characteristics...
READ MORE