Publications

Refine Results

(Filters Applied) Clear All

Link prediction methods for generating speaker content graphs

Published in:
ICASSP 2013, Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, 25-31 May 2013.

Summary

In a speaker content graph, vertices represent speech signals and edges represent speaker similarity. Link prediction methods calculate which potential edges are most likely to connect vertices from the same speaker; those edges are included in the generated speaker content graph. Since a variety of speaker recognition tasks can be performed on a content graph, we provide a set of metrics for evaluating the graph's quality independently of any recognition task. We then describe novel global and incremental algorithms for constructing accurate speaker content graphs that outperform the existing k nearest neighbors link prediction method. We evaluate those algorithms on a NIST speaker recognition corpus.
READ LESS

Summary

In a speaker content graph, vertices represent speech signals and edges represent speaker similarity. Link prediction methods calculate which potential edges are most likely to connect vertices from the same speaker; those edges are included in the generated speaker content graph. Since a variety of speaker recognition tasks can be...

READ MORE

Large-scale community detection on speaker content graphs

Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, ICASSP, 25-31 May 2013.

Summary

We consider the use of community detection algorithms to perform speaker clustering on content graphs built from large audio corpora. We survey the application of agglomerative hierarchical clustering, modularity optimization methods, and spectral clustering as well as two random walk algorithms: Markov clustering and Infomap. Our results on graphs built from the NIST 2005+2006 and 2008+2010 Speaker Recognition Evaluations (SREs) provide insight into both the structure of the speakers present in the data and the intricacies of the clustering methods. In particular, we introduce an additional parameter to Infomap that improves its clustering performance on all graphs. Lastly, we also develop an automatic technique to purify the neighbors of each node by pruning away unnecessary edges.
READ LESS

Summary

We consider the use of community detection algorithms to perform speaker clustering on content graphs built from large audio corpora. We survey the application of agglomerative hierarchical clustering, modularity optimization methods, and spectral clustering as well as two random walk algorithms: Markov clustering and Infomap. Our results on graphs built...

READ MORE

Sparse volterra systems: theory and practice

Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, ICASSP, 25-31 May 2013.

Summary

Nonlinear effects limit analog circuit performance, causing both in-band and out-of-band distortion. The classical Volterra series provides an accurate model of many nonlinear systems, but the number of parameters grows extremely quickly as the memory depth and polynomial order are increased. Recently, concepts from compressed sensing have been applied to nonlinear system modeling in order to address this issue. This work investigates the theory and practice of applying compressed sensing techniques to nonlinear system identification under the constraints of typical radio frequency (RF) laboratories. The main theoretical result shows that these techniques are capable of identifying sparse Memory Polynomials using only single-tone training signals rather than pseudorandom noise. Empirical results using laboratory measurements of an RF receiver show that sparse Generalized Memory Polynomials can also be recovered from two-tone signals.
READ LESS

Summary

Nonlinear effects limit analog circuit performance, causing both in-band and out-of-band distortion. The classical Volterra series provides an accurate model of many nonlinear systems, but the number of parameters grows extremely quickly as the memory depth and polynomial order are increased. Recently, concepts from compressed sensing have been applied to...

READ MORE

An Expectation Maximization Approach to Detecting Compromised Remote Access Accounts(267.16 KB)

Published in:
Proceedings of FLAIRS 2013, St. Pete Beach, Fla.

Summary

Just as credit-card companies are able to detect aberrant transactions on a customer’s credit card, it would be useful to have methods that could automatically detect when a user’s login credentials for Virtual Private Network (VPN) access have been compromised. We present here a novel method for detecting that a VPN account has been compromised, in a manner that bootstraps a model of the second unauthorized user.
READ LESS

Summary

Just as credit-card companies are able to detect aberrant transactions on a customer’s credit card, it would be useful to have methods that could automatically detect when a user’s login credentials for Virtual Private Network (VPN) access have been compromised. We present here a novel method for detecting that a...

READ MORE

Optimized airborne collision avoidance in mixed equipage environments

Published in:
MIT Lincoln Laboratory Report ATC-408

Summary

Developing robust collision avoidance logic that reliably prevents collision without excessive alerting is challenging due to sensor error and uncertainty in the future paths of the aircraft. Over the past few years, research has focused on the use of a computational method known as dynamic programming for producing an optimized decision logic for airborne collision avoidance. This report focuses on recent research on coordination, interoperability, and multiple-threat encounters. The methodology presented in this report results in logic that is safer and performs better than legacy TCAS. Modeling and simulation indicate that the proposed methodology can bring significant benefit to the current airspace and can support the need for safe, non-disruptive collision protection as the airspace continues to evolve.
READ LESS

Summary

Developing robust collision avoidance logic that reliably prevents collision without excessive alerting is challenging due to sensor error and uncertainty in the future paths of the aircraft. Over the past few years, research has focused on the use of a computational method known as dynamic programming for producing an optimized...

READ MORE

Airport surface traffic management decision support - perspectives based on tower flight data manager prototype

Summary

This report describes accomplishments and insights gathererd during the development of decision support tools as part of the Terminal Flight Data Manager (TFDM) program. This work was performed by MIT Lincoln Laboratory and sponsored by the Federal Aviation Administration (FAA). The TFDM program integrated flight data, aircraft surveillance, information on weather and traffic flow constraints, and other data required to optimize airport conguration and arrival/departure management functions. The prototype has been evaluated in both human-in-the-loop simulations, and during operational tests at Dallas/Fort Worth (DFW) International Airport. In parallel, the Laboratory estimated future national operational benefits for TFDM decision support functions, using analysis and performance data gathered from major airports in the US. This analysis indicated that the greatest potential operational benefits would come from decision support tools that facilitate: i) managing runway queues and sequences, ii) tactical management of flight routes and times, impacted by weather and traffic constraints, and iii) managing airport configuration changes. Evaluation of TFDM prototype decision support functions in each of these areas provided valuable insights relative to the maturity of current capabilities and research needed to close performance gaps.
READ LESS

Summary

This report describes accomplishments and insights gathererd during the development of decision support tools as part of the Terminal Flight Data Manager (TFDM) program. This work was performed by MIT Lincoln Laboratory and sponsored by the Federal Aviation Administration (FAA). The TFDM program integrated flight data, aircraft surveillance, information on...

READ MORE

P-sync: a photonically enabled architecture for efficient non-local data access

Summary

Communication in multi- and many-core processors has long been a bottleneck to performance due to the high cost of long-distance electrical transmission. This difficulty has been partially remedied by architectural constructs such as caches and novel interconnect topologies, albeit at a steep cost in terms of complexity. Unfortunately, even these measures are rendered ineffective by certain kinds of communication, most notably scatter and gather operations that exhibit highly non-local data access patterns. Much work has gone into examining how the increased bandwidth density afforded by chip-scale silicon photonic interconnect technologies affects computing, but photonics have additional properties that can be leveraged to greatly accelerate performance and energy efficiency under such difficult loads. This paper describes a novel synchronized global photonic bus and system architecture called P-sync that uses photonics' distance independence to greatly improve performance on many important applications previously limited by electronic interconnect. The architecture is evaluated in the context of a non-local yet common application: the distributed Fast Fourier Transform. We show that it is possible to achieve high efficiency by tightly balancing computation and communication latency in P-sync and achieve upwards of a 6x performance increase on gather patterns, even when bandwidth is equalized.
READ LESS

Summary

Communication in multi- and many-core processors has long been a bottleneck to performance due to the high cost of long-distance electrical transmission. This difficulty has been partially remedied by architectural constructs such as caches and novel interconnect topologies, albeit at a steep cost in terms of complexity. Unfortunately, even these...

READ MORE

Interdependence of the electricity generation system and the natural gas system and implications for energy security

Published in:
MIT Lincoln Laboratory Report TR-1173

Summary

Concern about energy security on domestic Department of Defense installations has led to the possibility of using natural gas-fired electricity generators to provide power in the event of electric grid failures. The natural gas system in the United States is partly dependent on electricity for its ability to deliver natural gas from the well-head to the consumer, but it also uses natural gas from the system itself to fuel some of the drilling rigs, processing units, and pipeline compressors. The vulnerability of the system to a disruption in the national electricity supply network varies depending on the cause and breadth of the disruption and where in the country one is located relative to that disruption, as the interconnected nature of transmission pipelines, the penetration of electric motor-driven compressors and other equipment, and the availability of nearby gas production, import terminals, or storage varies. In general, the gas supply system is reliable for short-term, limited-area disruptions in the electricity supply, and firm delivery contracts for natural gas increase the likelihood of continued operation, but for disruptions that cover large sections of the electric grid encompassing areas from extraction wells to customers and which last longer than available gas in storage or transmission pipeline constraints from elsewhere, contractual force majeure limits will come into play rendering the firm delivery contracts void; operation of gas-fueled power generation systems that are not dual-fuel capable for longer than weeks to a few months (depending on time of year) will be unlikely. Several weather-related outages in recent years have provided limited case studies showing the system's resilience, but no long-term, widespread electricity grid failures have occurred.
READ LESS

Summary

Concern about energy security on domestic Department of Defense installations has led to the possibility of using natural gas-fired electricity generators to provide power in the event of electric grid failures. The natural gas system in the United States is partly dependent on electricity for its ability to deliver natural...

READ MORE

High-voltage GaN-on-silicon Schottky diodes

Published in:
CS ManTech 2013, 13-16 May 2013.
Topic:

Summary

M/A-COM Technology Solutions has continuing joint development efforts sponsored by the Department of Energy with MIT main campus and MIT Lincoln Laboratory to develop GaN-on-silicon two and three-terminal high-voltage/high current switching devices. The initial developmental goals were for a Schottky diode that has a reverse breakdown blocking voltage of >600 volts and is capable of handling 10 amperes of forward current. A comparison of the M/A-COM Technology Solutions lateral GaN Schottky diode on-resistance as a function of reverse breakdown voltage for a number of both lateral and vertical GaN Schottky diode geometries taken from the literature is presented. The substrates employed for all of these data points are either sapphire, SiC, silicon, and even one study which utilized single crystal GaN. Also included in this plot are theoretical limits for the basic materials typically used in GaN Schottky diode construction. It can be seen that the reverse breakdown results of approximately 1500 volts for M/A/-COM Technology Solutions lateral anode connected field GaN Schottky diodes on silicon substrates compare extremely favorably with the reported performance of the state-of-the-art devices, regardless of substrate material or design geometry.
READ LESS

Summary

M/A-COM Technology Solutions has continuing joint development efforts sponsored by the Department of Energy with MIT main campus and MIT Lincoln Laboratory to develop GaN-on-silicon two and three-terminal high-voltage/high current switching devices. The initial developmental goals were for a Schottky diode that has a reverse breakdown blocking voltage of >600...

READ MORE

Single-mode tapered quantum cascade lasers

Published in:
Appl. Phys. Lett., Vol. 102, No. 18, 6 May 2013.

Summary

We demonstrate tapered quantum cascade lasers monolithically integrated with a distributed Bragg reflector acting as both a wavelength-selective back mirror and a transverse mode filter. Each of the 14 devices operates at a different wavelength between 9.2 and 9.7 um, where nine devices feature single-mode operation at peak powers between 0.3 and 1.6W at room temperature. High output power and excellent beam quality with peak brightness values up to 1.6MW cm^-2 sr^-1 render these two-terminal devices highly suitable for stand-off spectroscopy applications.
READ LESS

Summary

We demonstrate tapered quantum cascade lasers monolithically integrated with a distributed Bragg reflector acting as both a wavelength-selective back mirror and a transverse mode filter. Each of the 14 devices operates at a different wavelength between 9.2 and 9.7 um, where nine devices feature single-mode operation at peak powers between...

READ MORE