Chirp parameter estimation from tensor decomposition

The non-stationary properties of chirp signals restrict the application of the parameter estimation algorithms of single-frequency signals in the scene of chirp signals, simultaneously the traditional chirp parameter estimation algorithms also have limitations, to overcome some of the limitations this paper proposes a new algorithm which can estimate chirp parameter from tensor decomposition. The new algorithm uses the received discrete data aligning according to a certain form to build multidimensional data structures and then applies the tensor decomposition in chip parameter estimation using the shift invariance of subspace, the algorithm provides a new way of thinking in the field of chirp parameter estimation and it can apply in the scene of multiple chirp signals, the simulations prove the effectiveness of the algorithm.