Publications

Refine Results

(Filters Applied) Clear All

Experimental evaluation of features for robust speaker identification

Published in:
IEEE Trans. Speech Audio Process., Vol. 2, No. 4, October 1994, pp. 639-643.

Summary

This correspondence presents an experimental evaluation of different features and channel compensation techniques for robust speaker identification. The goal is to keep all processing and classification steps constant and to vary only the features and compensations used to allow a controlled comparison. A general, maximum-likelihood classifier based on Gaussian mixture densities is used as the classifier, and experiments are conducted on the King speech database, a conversational, telephone-speech database. The features examined are mel-frequency and linear-frequency filterbank cepstral coefficients, linear prediction ceptral coefficients. The channel compensation techniques examined are cepstral mean removal, RASTA processing, and a quadratic trend removal technique. It is shown for this database that performance difference between the basic features is small, and the major gains are due to the channel compensation techniques. The best "across-the-divide" recognition accuracy of 92% is obtained for both high-order LPC features and band-limited filterbank features.
READ LESS

Summary

This correspondence presents an experimental evaluation of different features and channel compensation techniques for robust speaker identification. The goal is to keep all processing and classification steps constant and to vary only the features and compensations used to allow a controlled comparison. A general, maximum-likelihood classifier based on Gaussian mixture...

READ MORE

TDWR scan strategy implementation

Author:
Published in:
MIT Lincoln Laboratory Report ATC-222

Summary

The Terminal Doppler Weather Radars (TDWRs) installed at major airports around the country are intended to enhance the safety of air travel by the detection and timely warning of hazardous wind shear conditions in the airport terminal area. To meet these objectives, scan strategies to efficiently cover the protected airspace were developed after extensive testing at several sites with different meteorological environments. Since the topology and geometry differ at each TDWR location, special considerations were necessary to define the specific scan sequences for each site. This report describes the criteria used to establish these scan sequences, including the determination of the lowest practicable elevation angle for each site - the "surface scan," which is used to detect microburst surface outflows, and other special scans such as the "MTS scan," which is used to illuminate the remote Moving Target Simulator (MTS).
READ LESS

Summary

The Terminal Doppler Weather Radars (TDWRs) installed at major airports around the country are intended to enhance the safety of air travel by the detection and timely warning of hazardous wind shear conditions in the airport terminal area. To meet these objectives, scan strategies to efficiently cover the protected airspace...

READ MORE

Air traffic control development at Lincoln Laboratory

Published in:
Lincoln Laboratory Journal, Vol. 7, No. 2, Fall 1994, pp. 147-148.

Summary

Lincoln Laboratory-developed air traffic control technologies, which were described in the Fall 1989 issue of this journal, are now in operational use. These technologies include the Mode-Select beacon system, the Traffic Alert and Collision Avoidance System, the Precision Runway Monitor system, the Terminal Doppler Weather Radar, and the Moving Target Detector signal processor used in the current generation of Airport Surveillance Radars. Our newest efforts focus on utilization of the Global Positioning System for both navigation and surveillance, and on the development of automation aids for air traffic control and management.
READ LESS

Summary

Lincoln Laboratory-developed air traffic control technologies, which were described in the Fall 1989 issue of this journal, are now in operational use. These technologies include the Mode-Select beacon system, the Traffic Alert and Collision Avoidance System, the Precision Runway Monitor system, the Terminal Doppler Weather Radar, and the Moving Target...

READ MORE

Automated microburst wind-shear prediction

Published in:
Lincoln Laboratory Journal, Vol. 7, No. 2, Fall 1994, pp. 399-426.

Summary

We have developed an algorithm that automatically and reliably predicts microburst wind shear. The algorithm, developed as part of the FAA Integrated Terminal Weather System (ITWS), can provide warnings several minutes in advance of hazardous low-altitude wind-shear conditions. Our approach to the algorithm emphasizes fundamental principles of thunderstorm evolution and downdraft development and incorporates heuristic and statistical methods as needed for refinement. In the algorithm, machine-intelligent image processing and data-fusion techniques are applied to Doppler radar data to detect those regions of growing thunderstorms and intensifying downdrafts which lead to microbursts. The algorithm then uses measurements of the ambient temperature/humidity structure in the atmosphere to aid in predicting a microburst's peak outflow strength. The algorithm has been tested in real time as part of the ITWS operational test and evaluation at Memphis, Tennessee, and Orlando, Florida, in 1994.
READ LESS

Summary

We have developed an algorithm that automatically and reliably predicts microburst wind shear. The algorithm, developed as part of the FAA Integrated Terminal Weather System (ITWS), can provide warnings several minutes in advance of hazardous low-altitude wind-shear conditions. Our approach to the algorithm emphasizes fundamental principles of thunderstorm evolution and...

READ MORE

Evaluation of runway-assignment and aircraft-sequencing algorithms in terminal area automation

Published in:
Lincoln Laboratory Journal, Vol. 7, No. 2, Fall 1994, pp. 215-238.

Summary

The Federal Aviation Administration has responded to the steady growth of air traffic and the ensuing increase in delays at airports by initiating new programs for increasing the efficiency of existing air traffic control facilities. The Terminal Air Traffic Control Automation (TATCA) program is intended to increase efficiency by providing controllers with planning aids and advisories to help them in vectoring, sequencing, and spacing traffic arriving at busy airports. Two important algorithms in this system allocate arrivals to multiple runways and set up the best sequences for landing aircraft. This article evaluates the potential for such algorithms to achieve higher throughput with less delay. The results show that, at airports with multiple active runways, the introduction of algorithms for systematic allocation of runways increases throughput considerably. These algorithms are in fact more effective than algorithms that aim at generating optimal landing sequences based on aircraft weight-class inputs. This result is fortuitous because algorithms for optimal sequencing are significantly more difficult to implement in practice than are algorithms for runway allocation. This study also provides a scientific basis for estimating future benefits of terminal automation by using traffic models patterned on actual recorded traffic-flow data, and by proposing a unified method for assessing performance.
READ LESS

Summary

The Federal Aviation Administration has responded to the steady growth of air traffic and the ensuing increase in delays at airports by initiating new programs for increasing the efficiency of existing air traffic control facilities. The Terminal Air Traffic Control Automation (TATCA) program is intended to increase efficiency by providing...

READ MORE

Performance of the runway-status light system at Logan Airport

Published in:
Lincoln Laboratory Journal, Vol. 7, No. 2, Fall 1994, pp. 187-214.

Summary

"Runway incursions are a persistent problem in airport ground-movement operations. Numerous critical conflicts and several fatal accidents have occurred as a result of unauthorized or otherwise inappropriate entry of aircraft or surface vehicles onto an active runway. Many of these conflicts developed quickly, leaving little time for effective intervention by either the controller or the pilots involved. A reliable system of automatic runway-status lights would be an effective way to prevent such time-critical incursions. The runway-status light system (RSLS) at Boston's Logan International Airport is an off-line proof-of-concept technology-demonstration system designed to show that automatically operated runway-status lights can promptly and reliably transmit runway-status information to pilots and surface-vehicle operators, thereby preventing unsafe runway entry or unsafe takeoff. The demonstration system does not include actual lights on the airport surface but has relied instead on an illuminated airport model board, which has allowed system development to proceed in a realistic operating environment of live airport traffic without interfering with airport operations. The results of an initial proof-of-concept assessment indicate that the system performs well, even though it is an early prototype. Missed-detection and false-alarm rates are low, and interference with normal airport operations promises to be negligible. The demonstration has shown the technical feasibility of a system of automatic runway-status lights.
"
READ LESS

Summary

"Runway incursions are a persistent problem in airport ground-movement operations. Numerous critical conflicts and several fatal accidents have occurred as a result of unauthorized or otherwise inappropriate entry of aircraft or surface vehicles onto an active runway. Many of these conflicts developed quickly, leaving little time for effective intervention by...

READ MORE

Supporting the deployment of the Terminal Doppler Weather Radar (TDWR)

Published in:
Lincoln Laboratory Journal, Vol. 7, No. 2, Fall 1994, pp. 379-398.

Summary

The Terminal Doppler Weather Radar (TDWR) program was initiated in the mid-1 980s to develop a reliable automated Doppler-radar-based system for detecting weather hazards in the airport terminal area and for providing warnings that will help pilots avoid these hazards when landing and departing. This article describes refinements made to the TDWR system since 1988, based on subsequent Lincoln Laboratory testing in Kansas City, Missouri, and Orlando, Florida. During that time, Lincoln Laboratory developed new capabilities for the system such as the integration of warnings from TDWR and the Low Level Wind Shear Alert System (LLWAS). Extensive testing with the Lincoln Laboratory TDWR testbed system has reconfirmed the safety benefits of TDWR.
READ LESS

Summary

The Terminal Doppler Weather Radar (TDWR) program was initiated in the mid-1 980s to develop a reliable automated Doppler-radar-based system for detecting weather hazards in the airport terminal area and for providing warnings that will help pilots avoid these hazards when landing and departing. This article describes refinements made to...

READ MORE

The Integrated Terminal Weather System (ITWS)

Published in:
Lincoln Laboratory Journal, Vol. 7, No. 2, Fall 1994, pp. 449-474.

Summary

The Integrated Terminal Weather System (ITWS) is one of two major development projects sponsored by the FMs Aviation Weather Development Program. Focused on the environment within the airport terminal area, ITWS integrates data from FAA and National Weather Service (NWS) sensors and systems to provide a suite of weather informational products for improving air terminal planning, capacity, and safety. This article provides an overview of the ITWS project, presenting the system concept, some of the design and engineering challenges, and plans for development that will lead to operational systems in the field.
READ LESS

Summary

The Integrated Terminal Weather System (ITWS) is one of two major development projects sponsored by the FMs Aviation Weather Development Program. Focused on the environment within the airport terminal area, ITWS integrates data from FAA and National Weather Service (NWS) sensors and systems to provide a suite of weather informational...

READ MORE

Implementation of the 1992 Terminal Area-Local Analysis and Prediction System (T-LAPS)

Published in:
MIT Lincoln Laboratory Report ATC-219

Summary

The Integrated Terminal Weather System (ITWS) development program was initiated by the Federal Aviation Administration (FAA) to produce a fully automated, integrated terminal weather information system to improve the safety, efficiency and capacity of terminal area aviation operations. The ITWS will acquire data from FAA and National Weather Service (NWS) sensors as well as from aircraft in flight in the terminal area. The ITWS will provide air traffic personnel with products that are immediately usable without further metorological interpretation. These products include current terminal area weather and short-term (0-30 minute) predictions of significant weather phenomena. The Terminal area-Local Analysis and Prediction System (T-LAPS) is being evaluated as a possible provider of the Terminal Winds Product for the ITWS. T-LAPS is a direct descendant of the Local Analysis and Prediction System (LAPS) developed at the National Oceanic and Atmospheric Administraiton's (NOAA's) Forecast Systems Laboratory (FSL). T-LAPS takes meteorological data from a wide variety of data sources as input and provides a gridded, three-dimensional (3-D) analysis of the state of the local atmosphere in the terminal area as output. For the 1992 system, the output was a gridded 3-D analysis of the horizontal winds. This information is intended to be used by the Terminal Air Traffic Control Automation (TATCA) program to estimate the effects of winds on aircraft in the terminal area. The 1993 and 1994 T-LAPS systems will incorporate more sophisticated wind analysis algorithms. The T-LAPS '92 demonstration at the Lincoln Laboratory Terminal Doppler Weather Radar (TDWR) FL-2CC field site in Kissimmee, Florida, during August and September was quite successful. The primary area of coverage was a 120 km by 120 km box centered on the Orlando International Airport. The T-LAPS system was able to utilize radar information from both the TDWR testbed and the operational NEXRAD/WSR-88D radar in Melbourne, Florida. This report documents the implementation of the T-LAPS system that was run during the 1992 summer demonstration and discusses the design and some implementation details of the system.
READ LESS

Summary

The Integrated Terminal Weather System (ITWS) development program was initiated by the Federal Aviation Administration (FAA) to produce a fully automated, integrated terminal weather information system to improve the safety, efficiency and capacity of terminal area aviation operations. The ITWS will acquire data from FAA and National Weather Service (NWS)...

READ MORE

Large population speaker recognition using wideband and telephone speech

Published in:
Proc. SPIE, Vol. 2277, Automatic Systems for the Identification and Inspection of Humans, 28-29 July 1994, pp. 111-120.

Summary

The two largest factors affecting automatic speaker identification performance are the size of the population to be distinguished among and the degradations introduced by noisy communication channels (e.g. telephone transmission). To experimentally examine these two factors, this paper presents text-independent speaker identification results for varying speaker population sizes up to 630 speakers for both clean, wideband speech and telephone speech. A system based on Gaussian mixture speaker models is used for speaker identification and experiments are conducted on the TIMIT and NTIMIT databases. The aims of this study are to (1) establish how well text-independent speaker identification can perform under near ideal conditions for very large populations (using the TIMIT database), (2) gauge the performance loss incurred by transmitting the speech over the telephone network (using the NTIMIT database), and (3) examine the validity of current models of telephone degradations commonly used in developing compensation techniques (using the NTIMIT calibration signals). This is believed to be the first speaker identification experiments on the complete 630 speaker TIMIT and NTIMIT databases and the largest text-independent speaker identification task reported to date. Identification accuracies of 99.5% and 60.7% are achieved on the TIMIT and NTIMIT databases, respectively.
READ LESS

Summary

The two largest factors affecting automatic speaker identification performance are the size of the population to be distinguished among and the degradations introduced by noisy communication channels (e.g. telephone transmission). To experimentally examine these two factors, this paper presents text-independent speaker identification results for varying speaker population sizes up to...

READ MORE