Grant: Discovery Grants Program - Individual
Concordia University
$19,000.00
Automatic generation of signal processing algorithms and architectures using nature-inspired paradigms
EF 2007-08
Résumé
The purpose of the proposed research program is to study, design and implement an intelligent algorithm- and architecture-generating system for a slew of targeted signal processing tasks with an embedded search engine working based on the principles of evolutionary computation and other nature-inspired optimization paradigms. The evolutionary design of algorithms and dedicated architectures will be carried out by focusing on discrete-time digital systems consisting of an array of interconnected modules. The modules will be designed internally, and their interconnection is optimized externally, to perform various dedicated signal processing tasks including, but not limited to, digital filtering, digital watermarking and signal compression. The final design candidates are obtained as the solutions to an optimization problem. The optimization is based on the principles of genetic algorithms, genetic programming, coevolutionary algorithms, and other nature-inspired meta-heuristics. These methods, because of their effectiveness in solving difficult combinatorial optimization problems, have the potential of generating innovative digital signal processing systems from scratch, only based on the prescribed specifications. A distinguishing characteristic of the proposed system is that it will be designed to possess the capability to reuse, accumulate and extract knowledge about the design space. We will also study and propose the formal description methods for encoding the target signal processing system as a data structure. This is a key element in the success of this research program and plays a specially significant role when we intend to harvest potentially new theoretical knowledge from what the system has discovered as the candidate solutions to our problem. Alternatively, a human expert may also analyze the final design candidates generated by the evolutionary process and reveal the theoretical underpinnings of the successful solutions. This process is significantly enhanced by the effectiveness of the formal description method. The formal description of the system will be treated in a higher level at the initial phase of the research. It will be enahanced to give rise to a system specification for implementation using field-programmable gate arrays.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
327679-2006
Name nom
Samadi, Saed
Department département
Electrical and Computer Engineering
Organization id
70
Institution établissement
Concordia University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2007
Competition year année de concours
2006
Award amount
19000
Program id
RGPIN
Program name 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
334
Committee name en
Commun.,Computers & Components
Committee name fr
Communic.,ordin. et composants
Area of application code
1206
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Engineering
Area of application fr
Génie
Research subject code
2514
Research subject group en
Electrical and electronic engineering
Research subject group fr
Génie électrique et électronique
Research subject en
Digital signal processing
Research subject fr
Traitement des signaux numériques
Partie
2
Nb partie
5
Application title
Automatic generation of signal processing algorithms and architectures using nature-inspired paradigms
Keyword
NA
Application summary
The purpose of the proposed research program is to study, design and implement an intelligent algorithm- and architecture-generating system for a slew of targeted signal processing tasks with an embedded search engine working based on the principles of evolutionary computation and other nature-inspired optimization paradigms. The evolutionary design of algorithms and dedicated architectures will be carried out by focusing on discrete-time digital systems consisting of an array of interconnected modules. The modules will be designed internally, and their interconnection is optimized externally, to perform various dedicated signal processing tasks including, but not limited to, digital filtering, digital watermarking and signal compression. The final design candidates are obtained as the solutions to an optimization problem. The optimization is based on the principles of genetic algorithms, genetic programming, coevolutionary algorithms, and other nature-inspired meta-heuristics. These methods, because of their effectiveness in solving difficult combinatorial optimization problems, have the potential of generating innovative digital signal processing systems from scratch, only based on the prescribed specifications. A distinguishing characteristic of the proposed system is that it will be designed to possess the capability to reuse, accumulate and extract knowledge about the design space. We will also study and propose the formal description methods for encoding the target signal processing system as a data structure. This is a key element in the success of this research program and plays a specially significant role when we intend to harvest potentially new theoretical knowledge from what the system has discovered as the candidate solutions to our problem. Alternatively, a human expert may also analyze the final design candidates generated by the evolutionary process and reveal the theoretical underpinnings of the successful solutions. This process is significantly enhanced by the effectiveness of the formal description method. The formal description of the system will be treated in a higher level at the initial phase of the research. It will be enahanced to give rise to a system specification for implementation using field-programmable gate arrays.