Free space optical (FSO) networks are emerging as a viable, cost effective technology for rapidly deployable broadband communication infrastructure. The main drawback of this type of networks is their dynamic performance, especially under adverse weather conditions and high nodes mobility. Topology control is used as the means to achieve survivable optical wireless networking under hostile conditions, based on dynamic and autonomous topology reconfiguration. The topology control process involves tracking and acquisition of nodes, assessment of link-state information, collection and distribution of topology data, and the algorithmic solution of an optimal topology. Design, analysis and comparison of algorithms and heuristics for configuring optimized topologies in dynamic environments are presented. Heuristics were developed for ring networks (2 optical transceivers per node) as well as for 3-degree networks (3 optical transceivers per node). This paper focuses on the design of efficient and scalable algorithms for physical layer topology optimization. That is, algorithms to select the topology configuration which optimizes a given physical layer objective. Performance and scalability results are shown for the various heuristics used, in different scenarios and for different network sizes.