PIPELINING OF LATTICE IIR DIGITAL-FILTERS

In recent years, rapid advances in VLSI technology have had much impac t on modern signal processing. Some of the desirable properties for VL SI realization are regularity, local connection and pipelinability. La ttice digital filters, which have many applications in signal modeling , spectrum estimation, and adaptive filtering, exhibit good finite wor d-length behavior, but cannot be pipelined to finer levels (such as bi t or multi-bit levels) due to the presence of feedback loops. Although lattice filters can be pipelined by the cut-set localization procedur e, it should be noted that the maximum sample rate cannot be increased by this technique. In this paper, based upon the properties of the Sc hur algorithm, a pipelining method in lattice digital filters is intro duced, by which the sample rate can be increased at any desired level. By constraining the poles to be located at equal angular and radial s pacing, the denominator of the transfer function is forced to be in sc attered look-ahead form. It is shown that this transfer function satis fies the pipelining property of lattice filters. Furthermore, based up on state variable description, new methods for scaling and output roun doff noise calculations are introduced; these can be easily applied to lattice IIR digital filters or lattice IIR digital filters connected with other type of filters. The relationship between pipelining stages and the output roundoff noise is analyzed using first and second-orde r IIR filters. The use of pipelined lattice IIR digital filters in low -power applications is also demonstrated.