Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Grants Program - Individual

University of Calgary

$46,000.00

Non linear filters and hidden Markov models

EF 2009-10

Résumé
The central theme of the research is the optimal estimation of signals observed in noise. The most successful algorithm for this is the Kalman filter. Extensions of the Kalman filter when the system jumps around a set of nominal operating modes have been developed, but they have been ad hoc procedures. Indeed, the IMM technology appears to be the current standard in the United States. The research will develop an exact filter obtained in previous work which provides improvements on the IMM, investigate Viterbi modifications and derive related smoothers. The results potentially have applications in many areas of communications, tracking and signal detection. We shall also develop estimation results for point processes. With the development of optical communication technology this modeling is of increasing interest. Some applications arise quite naturally from observed data that is modeled by counting processes, for example, the arrival of photon packets, or even the arrival of email packets. Further, several counting processes admit a marked point process representation, where each count event is coincident with another event of interest. An example is an email process, where its point process could be the times of arrival and the mark process the number of bytes in each email event. Immediate communication applications here include observing email processes to estimate traffic character or parameter values. Hidden Markov Models have been successfully applied in the sequencing of genes, and more recently proteins. We shall investigate further applications of our results in this area.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Alberta, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
7964-2009
Name nom
Elliott, Robert
Department département
Haskayne School of Business
Organization id
11
Institution établissement
University of Calgary
Province en
Alberta
Province fr
Alberta
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2009
Competition year année de concours
2009
Award amount
46000
Program id
RGPIN
Program nane en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
1079
Committee name en
Electrical and Computer Engineering
Committee name fr
Génie électrique et de l'informatique
Area of application code
513
Area of application group en
Manufacturing processes and products
Area of application group fr
Procédés et produits industriels
Area of application en
Electrical and electronic machinery and equipment (including computer hardware)
Area of application fr
Machinerie et équipement électriques et électroniques (y compris matériel informatique)
Research subject code
2507
Research subject group en
Electrical and electronic engineering
Research subject group fr
Génie électrique et électronique
Research subject en
Communications systems
Research subject fr
Systèmes de télécommunications
Partie
1
Nb partie
5
Application title
Non linear filters and hidden Markov models
Keyword
NA
Application summary
The central theme of the research is the optimal estimation of signals observed in noise. The most successful algorithm for this is the Kalman filter. Extensions of the Kalman filter when the system jumps around a set of nominal operating modes have been developed, but they have been ad hoc procedures. Indeed, the IMM technology appears to be the current standard in the United States. The research will develop an exact filter obtained in previous work which provides improvements on the IMM, investigate Viterbi modifications and derive related smoothers. The results potentially have applications in many areas of communications, tracking and signal detection. We shall also develop estimation results for point processes. With the development of optical communication technology this modeling is of increasing interest. Some applications arise quite naturally from observed data that is modeled by counting processes, for example, the arrival of photon packets, or even the arrival of email packets. Further, several counting processes admit a marked point process representation, where each count event is coincident with another event of interest. An example is an email process, where its point process could be the times of arrival and the mark process the number of bytes in each email event. Immediate communication applications here include observing email processes to estimate traffic character or parameter values. Hidden Markov Models have been successfully applied in the sequencing of genes, and more recently proteins. We shall investigate further applications of our results in this area.