A conceptual approach for military data fusion

Military operations become more and more complex. Information overload, and uncompleted and uncertain data made military information system design a challenge that would confer or withhold information superiority. The latter could represent military superiority especially in complex warfare such as urban operations or counterterrorism. This paper proposes a process for military information system design. This approach is based on a merge between the existing fusion frameworks on the one hand, and on the author’s experience in data fusion, on the other hand. Specifically, the design is based on system objectives and accuracy requirements, constrained by a determined sequence of events, in this case detection, identification, tracking and estimate of future states (DITE). Then, a five-dimension structure is proposed for data modeling and management: space (x,y,z), time (t) and possible worlds (w). On this basis, data exploitation and resource management could be performed with regard to network inferences, data perfection conditions and general data fusion approaches. A learning process could enhance the system by an intelligent component that would perform optimization.