A general class of stepped frequency trains

For uniform frequency stepped pulse trains, there can be undesirable peaks of the autocorrelation function, known as "grating lobes". In this paper, we address this issue, using an approach, which allows us to suppress grating lobes below a desired threshold level in the case of appropriately chosen stepped frequency waveforms, i.e., sequences of narrowband pulses that span the desired bandwidth. We will introduce a multiparametric generalization of a stepped frequency train, and by exploiting a factorization of the autocorrelation function, will be able to effect a useful tradeoff between competing properties of the factors by careful choices of relevant parameters. We provide a detailed description of the method applied to a representative data set.