Publications

Refine Results

(Filters Applied) Clear All

Channel engineering of SOI MOSFETs for RF applications

October 5, 2009
  |
Conference Paper
Author:
Chang-Lee Chen
…
Published in:
2009 IEEE Int. SOI Conf. Proc., 5 October 2009.
Topic:
semiconductor (CMOS)
R&D area:

Summary

Channel engineering of SOI MOSFETs is explored by altering ion implantation without adding any new fabrication steps to the standard CMOS process. The effects of implantation on characteristics important for RF applications, such as transconductance, output resistance, breakdown voltage, are compared. Data show that the best overall RF MOSFET has no body and drain-extension implants.
READ LESS

Summary

Channel engineering of SOI MOSFETs is explored by altering ion implantation without adding any new fabrication steps to the standard CMOS process. The effects of implantation on characteristics important for RF applications, such as transconductance, output resistance, breakdown voltage, are compared. Data show that the best overall RF MOSFET has...

READ MORE
Channel engineering of SOI MOSFETs for RF applications

Development of dual polarization aviation weather products for the FAA

October 5, 2009
  |
Conference Paper
Author:
David J. Smalley
…
Published in:
34th Conf. on Radar Meteorology, 5-9 October 2009.
Topic:
aviation weather
R&D area:
R&D group:

Summary

Weather radar products from the United States' NEXRAD network are used as key components in FAA weather systems such as CIWS, ITWS, and WARP. The key products, High Resolution VIL (HRVIL) and High Resolution Enhanced Echo Tops (HREET), provide primary information about precipitation location and intensity. The NEXRAD network will become dual polarization capable beginning in late 2010 adding the ability to classify hydrometeors. This new aspect from radar remote sensing of the weather offers opportunity to provide new aviation weather products and augment existing ones. This paper will detail the dual polarization product development program at MIT Lincoln Laboratory (LL) in support of FAA system needs. Current development efforts focus on four products. Two new products will provide volumetric analysis seeking aviation hazards (icing and hail). Two existing products, HRVIL and HREET, will be invigorated by dual polarization data to yield improved data quality and mitigation of partial beam blockage. The LL program has partnered with NCAR and NSSL subject matter experts to bring their most advanced research results into these new and improved products. The LL program also has partnered with Valparaiso University for them to provide dual polarization and local sonde sounding data especially during suspected icing conditions. The new Icing Hazards Level (IHL) product is expected to provide the most benefit to the FAA. Its development also poses the greatest challenge both in scope and in the ability of S-band radar to sense the phenomena of interest. Icing phenomena include supercooled drops/droplets and ice crystals and the associated aviation hazard could be aloft or near/at the surface. Graupel is an indication that supercooled water has accreted to ice particles. The initial NEXRAD hydrometeor classifier will not have an explicit supercooled water class or the benefit of supporting data. It will have ice crystal and graupel classes. The LL approach will utilize at least some additional data (vertical thermodynamic profiles). Techniques applied to the development of IHL will likely have applicability to the other products as well. Aspects of the IHL development will also be discussed in the paper.
READ LESS

Summary

Weather radar products from the United States' NEXRAD network are used as key components in FAA weather systems such as CIWS, ITWS, and WARP. The key products, High Resolution VIL (HRVIL) and High Resolution Enhanced Echo Tops (HREET), provide primary information about precipitation location and intensity. The NEXRAD network will...

READ MORE
Development of dual polarization aviation weather products for the FAA

A log-frequency approach to the identification of the Wiener-Hammerstein model

October 1, 2009
  |
Journal Article
Author:
Joel I. Goodman
…
Published in:
IEEE Sig. Proc. Lett., Vol. 16, No. 10, October 2009, pp. 889-892.
Topic:
signal processing
R&D area:
R&D group:

Summary

In this paper we present a simple closed-form solution to the Wiener-Hammerstein (W-H) identification problem. The identification process occurs in the log-frequency domain where magnitudes and phases are separable. We show that the theoretically optimal W-H identification is unique up to an amplitude, phase and delay ambiguity, and that the nonlinearity enables the separate identification of the individual linear time invariant (LTI) components in a W-H architecture.
READ LESS

Summary

In this paper we present a simple closed-form solution to the Wiener-Hammerstein (W-H) identification problem. The identification process occurs in the log-frequency domain where magnitudes and phases are separable. We show that the theoretically optimal W-H identification is unique up to an amplitude, phase and delay ambiguity, and that the...

READ MORE
A log-frequency approach to the identification of the Wiener-Hammerstein model

Wafer-scale 3D integration of InGaAs image sensors with Si readout circuits

September 28, 2009
  |
Conference Paper
Author:
Chang-Lee Chen
…
Published in:
3DIC 2009, IEEE Int. Conf. on 3D System Integration, 28-30 September 2009.
Topic:
integrated circuit
R&D area:
R&D group:

Summary

In this work, we modified our wafer-scale 3D integration technique, originally developed for Si, to hybridize InP-based image sensor arrays with Si readout circuits. InGaAs image arrays based on the InGaAs layer grown on InP substrates were fabricated in the same processing line as silicon-on-insulator (SOI) readout circuits. The finished 150-mm-diameter InP wafer was then directly bonded to the SOI wafer and interconnected to the Si readout circuits by 3D vias. A 1024 x 1024 diode array with 8-um pixel size is demonstrated. This work shows the wafer-scale 3D integration of a compound semiconductor with Si.
READ LESS

Summary

In this work, we modified our wafer-scale 3D integration technique, originally developed for Si, to hybridize InP-based image sensor arrays with Si readout circuits. InGaAs image arrays based on the InGaAs layer grown on InP substrates were fabricated in the same processing line as silicon-on-insulator (SOI) readout circuits. The finished...

READ MORE
Wafer-scale 3D integration of InGaAs image sensors with Si readout circuits

Unmanned aircraft collision avoidance using partially observable Markov decision processes

September 22, 2009
  |
Project Report
Author:
Mykel J. Kochenderfer
…
Published in:
MIT Lincoln Laboratory Report ATC-356
Topic:
collision avoidance
R&D area:

Summary

Before unmanned aircraft can fly safely in civil airspace, robust airborne collision avoidance systems must be developed. Instead of hand-crafting a collision avoidance algorithm for every combination of sensor and aircraft configuration, this project investigates the automatic generation of collision avoidance logic given models of aircraft dynamics, sensor performance, and intruder behavior. By formulating the problem of collision avoidance as a partially-observable Markov decision process (POMDP), a generic POMDP solver can be used to generate avoidance strategies that optimize a cost function that balances flight-plan deviation with collision. Experimental results demonstrate the suitability of such an approach using three different sensor modalities and two aircraft performance models.
READ LESS

Summary

Before unmanned aircraft can fly safely in civil airspace, robust airborne collision avoidance systems must be developed. Instead of hand-crafting a collision avoidance algorithm for every combination of sensor and aircraft configuration, this project investigates the automatic generation of collision avoidance logic given models of aircraft dynamics, sensor performance, and...

READ MORE
Unmanned aircraft collision avoidance using partially observable Markov decision processes

Redeployment of the New York TDWR - technical analysis of candidate sites and alternative wind shear sensors

September 14, 2009
  |
Project Report
Author:
Suilou Huang
…
Published in:
MIT Lincoln Laboratory Report ATC-351
Topic:
wind shear
R&D area:
R&D group:

Summary

The John F. Kennedy International Airport (JFK) and LaGuardia Airport (LGA) are protected from wind shear exposure by the New York Terminal Doppler Weather Radar (TDWR), which is currently located at Floyd Bennet Field, New York. Because of a September 1999 agreement between the Department of the Interior and the Department of Transportation, this location is required to be vacated no later than January 2023. Therefore, a study based on model simulations of wind shear detection probability was conducted to support future siting selection and alternative technologies. A total of 18 candidate sites were selected for analysis, including leaving the radar where it is. (The FAA will explore the feasibility of the latter alternative; it is included in this study only for technical analysis.) The 18 sites are: Six candidate sites that were identified in the initial New York TDWR site-survey studies in the 1990s (one of which is the current TDWR site), a site on Staten Island, two Manhattan skyscrapers, the current location of the WCBS Doppler weather radar in Twombly Landing, New Jersey, and eight local airports including JFK and LGA themselves. Results clearly show that for a single TDWR system, all six previously surveyed sites are suitable for future housing of the TDWR. Unfortunately, land acquisition of these sites will be at least as challenging as it was in the 1990s due to further urban development and likely negative reaction from neighboring residents. Evaluation results of the on-airport siting of the TDWR (either at JFK or at LGA) indicate that this option is feasible if data from the Newark TDWR are simultaneously used. This on-airport option would require software modification such as integration of data from the two radar systems an dimplementation of "overhead" feature detection. The radars on the Manhattan skyscrapers are not an acceptable alternative due to severe ground clutter. The Staten Island site and most other candidate airports are also not acceptable due to distance and/or beam blockage. The existing Airport Surveillance Radar (ASR-9) Weather Systems Processor (WSP) at JFK and the Bookhaven (OKX) Weather Surveillance Radar 1988-Doppler (WSR-88D, commonly known as NEXRAD) on Long Island cannot provide sufficient wind shear protection mainly due to limited wind shear detection capability and/or distance.
READ LESS

Summary

The John F. Kennedy International Airport (JFK) and LaGuardia Airport (LGA) are protected from wind shear exposure by the New York Terminal Doppler Weather Radar (TDWR), which is currently located at Floyd Bennet Field, New York. Because of a September 1999 agreement between the Department of the Interior and the...

READ MORE
Redeployment of the New York TDWR - technical analysis of candidate sites and alternative wind shear sensors

2-D processing of speech for multi-pitch analysis.

September 6, 2009
  |
Conference Paper
Author:
Tianyu Tom Wang
…
Published in:
INTERSPEECH 2009, 6-10 September 2009.
Topic:
signal processing
R&D area:
R&D group:

Summary

This paper introduces a two-dimensional (2-D) processing approach for the analysis of multi-pitch speech sounds. Our framework invokes the short-space 2-D Fourier transform magnitude of a narrowband spectrogram, mapping harmonically related signal components to multiple concentrated entities in a new 2-D space. First, localized time-frequency regions of the spectrogram are analyzed to extract pitch candidates. These candidates are then combined across multiple regions for obtaining separate pitch estimates of each speech-signal component at a single point in time. We refer to this as multi-region analysis (MRA). By explicitly accounting for pitch dynamics within localized time segments, this separability is distinct from that which can be obtained using short-time autocorrelation methods typically employed in state-of-the-art multi-pitch tracking algorithms. We illustrate the feasibility of MRA for multi-pitch estimation on mixtures of synthetic and real speech.
READ LESS

Summary

This paper introduces a two-dimensional (2-D) processing approach for the analysis of multi-pitch speech sounds. Our framework invokes the short-space 2-D Fourier transform magnitude of a narrowband spectrogram, mapping harmonically related signal components to multiple concentrated entities in a new 2-D space. First, localized time-frequency regions of the spectrogram are...

READ MORE
2-D processing of speech for multi-pitch analysis.

A comparison of query-by-example methods for spoken term detection

September 6, 2009
  |
Conference Paper
Author:
Wade Shen
…
Published in:
INTERSPEECH 2009, 6-10 September 2009.
Topic:
speech recognition
R&D area:
R&D group:

Summary

In this paper we examine an alternative interface for phonetic search, namely query-by-example, that avoids OOV issues associated with both standard word-based and phonetic search methods. We develop three methods that compare query lattices derived from example audio against a standard ngrambased phonetic index and we analyze factors affecting the performance of these systems. We show that the best systems under this paradigm are able to achieve 77% precision when retrieving utterances from conversational telephone speech and returning 10 results from a single query (performance that is better than a similar dictionary-based approach) suggesting significant utility for applications requiring high precision. We also show that these systems can be further improved using relevance feedback: By incorporating four additional queries the precision of the best system can be improved by 13.7% relative. Our systems perform well despite high phone recognition error rates (> 40%) and make use of no pronunciation or letter-to-sound resources.
READ LESS

Summary

In this paper we examine an alternative interface for phonetic search, namely query-by-example, that avoids OOV issues associated with both standard word-based and phonetic search methods. We develop three methods that compare query lattices derived from example audio against a standard ngrambased phonetic index and we analyze factors affecting the...

READ MORE
A comparison of query-by-example methods for spoken term detection

A framework for discriminative SVM/GMM systems for language recognition

September 6, 2009
  |
Conference Paper
Author:
William M. Campbell
…
Published in:
INTERSPEECH 2009, 6-10 September 2009.
Topic:
speaker recognition
R&D area:
R&D group:

Summary

Language recognition with support vector machines and shifted-delta cepstral features has been an excellent performer in NIST-sponsored language evaluation for many years. A novel improvement of this method has been the introduction of hybrid SVM/GMM systems. These systems use GMM supervectors as an SVM expansion for classification. In prior work, methods for scoring SVM/GMM systems have been introduced based upon either standard SVM scoring or GMM scoring with a pushed model. Although prior work showed experimentally that GMM scoring yielded better results, no framework was available to explain the connection between SVM scoring and GMM scoring. In this paper, we show that there are interesting connections between SVM scoring and GMM scoring. We provide a framework both theoretically and experimentally that connects the two scoring techniques. This connection should provide the basis for further research in SVM discriminative training for GMM models.
READ LESS

Summary

Language recognition with support vector machines and shifted-delta cepstral features has been an excellent performer in NIST-sponsored language evaluation for many years. A novel improvement of this method has been the introduction of hybrid SVM/GMM systems. These systems use GMM supervectors as an SVM expansion for classification. In prior work...

READ MORE
A framework for discriminative SVM/GMM systems for language recognition

Discriminative N-gram selection for dialect recognition

September 6, 2009
  |
Conference Paper
Author:
Frederick S. Richardson
…
Published in:
INTERSPEECH 2009, 6-10 September 2009.
Topic:
language recognition
R&D area:
R&D group:

Summary

Dialect recognition is a challenging and multifaceted problem. Distinguishing between dialects can rely upon many tiers of interpretation of speech data - e.g., prosodic, phonetic, spectral, and word. High-accuracy automatic methods for dialect recognition typically rely upon either phonetic or spectral characteristics of the input. A challenge with spectral system, such as those based on shifted-delta cepstral coefficients, is that they achieve good performance but do not provide insight into distinctive dialect features. In this work, a novel method based upon discriminative training and phone N- grams is proposed. This approach achieves excellent classification performance, fuses well with other systems, and has interpretable dialect characteristics in the phonetic tier. The method is demonstrated on data from the LDC and prior NIST language recognition evaluations. The method is also combined with spectral methods to demonstrate state-of-the-art performance in dialect recognition.
READ LESS

Summary

Dialect recognition is a challenging and multifaceted problem. Distinguishing between dialects can rely upon many tiers of interpretation of speech data - e.g., prosodic, phonetic, spectral, and word. High-accuracy automatic methods for dialect recognition typically rely upon either phonetic or spectral characteristics of the input. A challenge with spectral system...

READ MORE
Discriminative N-gram selection for dialect recognition
1051-1060 of 2301