Parallel Form of the Pipelined Intermediate Architecture for Two-dimensional Discrete Wavelet Transform

Abstract A lifting-based VLSI architecture for twodimensional discrete wavelet transform (2-D DWT) for 5/3 and 9/7 algorithms, called, pipelined intermediate architecture was proposed by Ibrahim et al., which aim at reducing power consumption of the overlapped external memory access without using the expensive line-buffer. In this paper, we explore parallelism in order to best meet real-time applications of 2-D DWT with demanding requirements in terms of speed, throughput, and power consumption. Therefore, 2-parallel and 3-parallel form of the single pipelined intermediate architecture are proposed. The 2-parallel and 3-parallel pipelined intermediate architectures achieve speedup factors of 2 and 3, respectively, as compared with single pipelined intermediate architecture proposed by Ibrahim et al.