Set-membership affine projection channel estimation for wireless sensor networks

A set-membership affine projection algorithm is applied to estimate the communication channel between the wireless sensor nodes in a general form, where the channel is modeled as a complex matrix in the presence of additive white gaussian noise. An efficient hybrid model for the affine projection algorithm is briefly introduced and the problem of matrix invertibility in some cases of the affine projection algorithm is resolved by a new method which does not use any matrix inversion. Simulations show good performance of our proposed algorithm in terms of convergence speed and demonstrate reduced complexity.