A pipelined adaptive lattice filter architecture

The stochastic gradient adaptive lattice filter is pipelined by the application of the relaxed look-ahead. This form of look-ahead maintains the functional behavior instead of the input-output mapping. The sum and product relaxations are employed to pipeline the lattice filter. The hardware complexity of the proposed pipelined filter is of the same order as the sequential filter. Thus, the new architecture is attractive from an implementation point of view. Convergence analysis is carried out to illustrate the trade-off offered by relaxed look-ahead. Linear prediction of a speech signal is used as a simulation example