Space-time coding and diversity techniques have been an active research area for wireless communication since long, and have been seen to significantly improve performances of communication systems. Although similar approaches for radars have not been attempted yet, it is rational to expect comparable performance gains for the multi-aperture radar provided we can come up with the optimal transmit solutions for these space-time systems. To make the most of a multi-aperture transmitter, it needs to be considered and utilized as a non-separable sensor in space and time, i.e. each aperture propagates a coherent but uncorrelated signal in time and frequency.; A methodology to create such non-separable space-time signals for various illumination optimization criteria is developed, and the performance validated through various tests and simulations. By expanding the space-time signal in terms of a set of orthonormal basis functions, the problem is reduced to that of finding the optimal set of weights for these functions, i.e. a vector. In combination with the vector-matrix models developed, the structure enables the use of linear algebraic techniques like eigen analysis for determining the optimal solutions.; Out of the many optimization criteria developed, the most applicable one is found to be the mini-max or maxi-min. However solutions to this criterion are not easy to find; an iterative procedure having a sound algebraic basis is presented as a solution. The efficacy of the method has been demonstrated with relevant results, and its dependence studied on a number of parameters. Particularly interesting is the form of the resulting signal, which is examined both at the transmitter and also at the targets; some very important conclusions are drawn based on the observations. |