An instrumental variable based subspace tracking algorithm based on subspace averaging

In this paper an instrumental variable based subspace tracking algorithm is proposed. The basic idea of the algorithm is to reduce the amount of computations using a certain perturbation/approximation strategy. The complexity is reduced to O(mn²), which should be compared to O(ml²) for the SVD, where m, l≫n in general (m denotes the number of sensors, l denotes the number of instruments, and n denotes the number of signals). The proposed algorithm turns out to be related to Karasalo´s subspace averaging approach (1986). In a series of simulations we demonstrate that the detection, stationary estimation, and tracking performance of the proposed algorithm is essentially equivalent to that achieved by the truncated SVD