Enabling Practical Uses of Arithmetic Transforms – A Comprehensive Analysis
نویسندگان
چکیده
Arithmetic Transform (AT) has been known under various names, including Inverse Integer Reed-Muller (IRM) transform. We outline the developments that place AT in the center of very practical arithmetic circuit applications, and explain the conditions for making the AT practical. We show that AT allows the only known scheme of treating all the approximation and imprecision sources of arithmetic error in the optimization, through an efficient static method. Finally, the role of AT in the stochastic computing is highlighted.
منابع مشابه
Efficient erasure decoding of Reed-Solomon codes
We present a practical algorithm to decode erasures of Reed-Solomon codes over the q elements binary field in O(q log 2 q) time where the constant implied by the O-notation is very small. Asymptotically fast algorithms based on fast polynomial arithmetic were already known, but even if their complexity is similar, they are mostly impractical. By comparison our algorithm uses only a few Walsh tr...
متن کاملNumerical stability of fast trigonometric and orthogonal wavelet transforms
Fast trigonometric transforms and periodic orthogonal wavelet transforms are essential tools for numerous practical applications. It is very important that fast algorithms work stable in a floating point arithmetic. This survey paper presents recent results on the worst case analysis of roundoff errors occurring in floating point computation of fast Fourier transforms, fast cosine transforms, a...
متن کاملInformation Hiding in Software with Mixed Boolean-Arithmetic Transforms
As increasingly powerful software analysis and attack tools arise, we need increasingly potent software protections. We generate an unlimited supply of obscuring transforms via mixed-mode computation over Boolean-arithmetic (mba) algebras corresponding to real-world functions and data. Such transforms resist reverse engineering with existing advanced tools and create np-hard problems for the at...
متن کاملAn FPGA implementation of Walsh-Hadamard transforms for signal processing
This paper describes two approaches suitable for an FPGA implementation of Walsh-Hadamard transforms. These transforms are important in many signal-processing applications including speech compression, filtering and coding. Two novel architectures for the Fast Hadamard Transforms using both systolic architecture and distributed arithmetic techniques are presented. The first approach uses the Ba...
متن کاملContext-Based Arithmetic Coding for the DCT: Achieving high compression rates with block transforms and simple context modeling
Recent image compression schemes have focused primarily on wavelet transforms, culminating in the JPEG-2000 standard. Block based DCT compression, on which the older JPEG standard is based, has been largely neglected because wavelet based coding methods appear to offer better image quality. This paper presents a simple compression algorithm that uses arithmetic coding on the bit-planes of the D...
متن کامل