VLSI Realization Of CDF9/7 Wavelet Transform In JPEG2000

As a noval theory, Wavelet transform is a very important production in the phylogeny of mathematics, and has strongly impacted on both mathmatics and engineering applications. The discrete wavelet transform (DWT) has been adoppted as the core transforming algorithm in JPEG2000, a latest static image compression standard.The theory fundamental of wavelet transform was firstly described in details, and the multi-resolution analysis (MRA), Mallat algorithm and Lifting Scheme were also introduced respectively. Based on the the algorithms and the lifting schemes of Le Gell5/3 and CDF9/7, which were adoppted by the JPEG2000 standard, the debug and the soft simulation of CDF9/7 were performed . The implementation of two-in and two-out 2-D lifting-based 2-D discrete wavelet transform (2D-DWT) , was then proposed ,based on W-scan and parallel architecture. The memory requirements and power consumptions of the DWT circuits can be reduced effectively because of the implementation of proposed 2-D DWT, based on embedded symmetric extension and parallel architecture, which require only six lines single port RAM instead of dural port RAM in the buffer, leading to the significantly reduction of the size of the embedded storage and the amount of memory accesses.On the basis of the hardware architecture, the thesis describes the whole design in Verilog HDL at the full synthesizable RTL level with synchronous design technology to improve the stability of the system. The simulation and implementation of DWT were performed on Xilinx's ISE6.0, and the experiment results show that the design can quickly and precisely complete DWT, and can meet the real-time requirements for application.