VLSI Digital Signal Processing Systems: Design and Implementation

10.1    INTRODUCTION

Any required digital filter spectrum can be realized using finite impulse response (FIR) or infinite impulse response (IIR) digital filters. IIR digital filters are often preferred since a specified spectrum can be implemented using a much lower order IIR filter than FIR. FIR and IIR digital filters are ideal for frequency shaping in deterministic environments, but are not useful in time-varying systems. Adaptive filters are used in these cases for applications such as noise cancellation, echo cancellation, beamforming, system identification, etc. Unlike the fixed coefficients FIR or IIR digital filters, the coefficients of the adaptive digital filters are adapted at each iteration until they converge.

Exploiting concurrency in the form of either pipelining or parallel processing is straightforward for nonrecursive computations. However, recursive and adaptive digital filters cannot be easily pipelined or processed in parallel due to the presence of feedback loops in these filters. This chapter presents approaches for pipelining and parallel processing in recursive digital filters using the look-ahead computation and incremental block processing techniques and the relaxed look-ahead transformations for pipelining of LMS and lattice adaptive filters.

This chapter is organized as follows. Pipelining in IIR digital filters using interleaving of filtering operations for multiple channels and look-ahead computation for single channel are introduced in Section 10.2 ...