Chapter Preview
TopBackground
As previously described, this subject merges two areas of study. A brief background of both areas is presented here to provide the necessary comprehension.
Key Terms in this Chapter
Evolutionary Electronics (EEL): Also known as Hardware Evolution, Evolvable Hardware, EvolWare, or bio-inspired electronics, is a research area which covers all the applications involving the use of Evolutionary Computation in electronic systems’ design.
Evolutionary Algorithms (EAs): Algorithms which are based on evolutionary computation principles.
Genetic Algorithm (GA): An evolutionary algorithm that works with encoded population-based search imitating nature through the following mechanisms: selection, reproduction and mutation. The encoding consists of mapping each possible solution belongs to the search space to a symbol string, usually bits. Each encoded string is called chromosome. A population of chromosomes which is a sample of the search space is the base for the heuristic search.
Field Programmable Gate Array (FPGA): A programmable chip that contains logic gates as well as that can be arranged in different ways. To receive different configurations, the FPGA is connected through an interface to a computer which contains vendor specific software that offers to the developer two kinds of specification languages: one graphical and the other textual. VHDL (IEEE Standardized), Verilog, and AHDL are examples of the textual description languages.
Evolutionary Computation (EC): A research area that consists of computational optimization search methods based on heuristics which imitates nature to find optimized solutions.
Field Programmable Analog Array (FPAA): A programmable chip that contains operational amplifiers and passive components that can be arranged in different ways. To receive different configurations, the FPAA is connected through an interface to a computer which contains vendor specific software that offers to the developer a specification language usually graphical.
Genetic Programming (GP): A variation of genetic algorithm in which the members of the population are parse trees. GP usually is used to evolve symbolic information (examples are: computer code, symbolic regression, and automatic electrical circuit design).