A Non-Cryogenic Tunable Diode Laser Monitor for On-Road Vehicle Emissions

Not Available This proposal addresses the problem of placement or rearrangement of antennas and ground stations in a satellite support network that will provide good network performance at low cost. The problem is motivated by the economic concerns associated with consolidation of satellite support network resources. The solution to this problem is extremely complex and requires juggling many different constraints and performance requirements. Herein is proposed the development of a set of genetic search algorithms for automatic near-optimal solution of this problem. Genetic search methods are a type of directed trial-and-error, and therefore, as they are not gradient-based, are ideal for optimization problems with complex or non-smooth constraints or performance measures. Development of these algorithms must be preceded by an in-depth analysis of the problem, during which factors and constraints relevant to antenna placement can be identified. Performance criteria for evaluation of the antenna/ground station configurations produced by the algorithms will also be defined. The algorithms developed will then be implemented and tested on a representative set of placement scenarios.