Summary: | 碩士 === 逢甲大學 === 資訊工程所 === 91 === The fast Walsh Hadamard transform is a simple and important algorithm in digital signal processing, especially, in digital image compression. The fast Walsh Hadamard transform is a block recursive algorithm, which can be easily expressed as a tensor product formula. Also, the tensor product notation can be used to express the topology of interconnection networks. In this thesis, we will emerge the specification of tensor product formulation of block recursive algorithms and interconnection networks to design fast Walsh Hadamard transform algorithms on various interconnection networks, including hypercube networks, omega networks, and baseline networks. The design process starts from a tensor product formula of the fast Walsh Hadamard transform. Given a specific interconnection network, we will manipulate the tensor product formula of the fast Walsh Hadamard transform to an equivalent formula which fits the form of the tensor product formula of that interconnection network. The resulting algorithms are suitable for designing VLSI circuits of DSP algorithms.
|