Complex systems in operation are constrained by legacy; in other words, the existing properties, structure, components, protocols, software, people and etc. that are inherited over time. This inheritance heavily influences the type and timing of feasible and available changes to the system, the effectiveness and desirability of their impact while accounting for uncertainty, and the future constraints imposed as a result. This thesis introduces the Strategic Evolution of Systems, a novel framework for evolving complex systems that directly addresses legacy challenges during system operation within the context of space communication networks. The framework -perspective, position, plan and pattern -is based on Mintzberg's "emergent" interpretation of strategy.This thesis also presents several unique ideas including the concept of option lock-out, or the tendency to lose access to potentially desirable regions of the architectural space when exercising a transition; an energy analogy to model static architecture value; an entropy-based formulation to evaluate the desirability, or dynamic multidimensional value, of an architecture by considering the structural and temporal space of possible transitions; and the application of the entropy-based formulation to define the overall desirability of an architecture as its position, or current situation (favorable or unfavorable) relative to accessible alternatives, in order to identify the most advantageous immediate transition.A key contribution of this thesis is a method to value legacy in a physical non-market traded system, including a demonstration of its application to a system in which benefits and costs are nonmonetary in nature. Other important contributions include a change exposure tool, referred to as a Strategic Advantage Map, to visualize the near-and long-term impact of immediate transitions relative to legacy. Here, an architecture's position relative to the legacy system can be thought of as the region of entropy space it occupies (evaluated over time and uncertainty). The more dominating this region of position entropy is, the more desirable the architecture. For monetary-based systems, a second change exposure tool includes an "Iceberg Exposure,"which maps the exposure of net present value for each accessible transition option relative to a neutral no-gain-no-loss line, resulting in a graph resembling an iceberg. The visualization tools allow decision makers to quickly evaluate the impact (risk/opportunity) of change, based on their concept of desirability.Case studies include a historical look at the NASA Deep Space Network for insight into legacy and complex system evolution, a demonstration of the Strategic Evolution of Systems framework for a global commercial satellite communication system, and an illustration of the method extended to non-monetary systems for the deployment of communication assets to support manned exploration 3 of Earth's moon. The satellite system case study introduces an extended market model that evaluates the attainable...