In this paper we consider OFDM systems with non-contiguous subcarriers where distributed spectrum fragments are applicable for transmission. A new model is proposed for pilot optimization with respect to location and power. The least-square approach is applied for channel estimation at the receiver and the optimization objective is to minimize the mean square error of the channel estimation. This optimization problem is then formulated as minimizing the spread of the eigenvalues of a positive semidefinite matrix, which is a typical semidefinite programming problem, and hence it can be solved with conventional interior-point method. Simulation results show the efficiency of the proposed algorithm.
Read full abstract