A Fundamental Basis for Power-Reduction in VLSI Circuits - Circuits and Systems, 1996., ISCAS '96, 'Connecting the World'., 1996 IEEE International Symposium
نویسنده
چکیده
Presented in this paper is a fundamental mathematical basis for power-reduction in VLSI systems. This basis is employed to 1.) derive lower bounds on the power dissipation in digital systems and 2.) unify existing power-reduction techniques under a common framework. The proposed basis is derived from information-theoretic arguments. In particular, a digital signal processing algorithm is viewed as a process of information transfer with an inherent information transfer rate requirement of R bits/sec. Different architectures implementing a given algorithm are equivalent to different communication networks each with a certain capacity C (also in bits/sec). The absoluteJower bound on the power dissipation for any given architecture is then obtained by minimizing the signal power such that its channel capacity C is equal to the desired information transfer rate R. Numerical calculations for a simple static CMOS circuit and fundamental basis for the power-reduction capabilities of parallel processing and pipelining are presented.
منابع مشابه
A NEW SYSTOLIC ARCHITECTURE FOR FAST DCT COMPUTATION - Circuits and Systems, 1996., ISCAS '96, 'Connecting the World'., 1996 IEEE International Symposium
This paper presents a fast algorithm along with its systolic array implementation for computing the 1-D Npoint discrete cosine transform (DCT), where N is a power of two. The architecture requires log,N multipliers and can evaluate one complete N-point DCT every N clock cycles. It possesses the features of regularity and modularity, and is thus well suited to VLSI implementation. As compared to...
متن کاملSYSTOLIC ARRAY IMPLEMENTATION OF EUCLID'S ALGORITHM FOR INVERSION AND DIVISION IN GF(2/sup m/) - Circuits and Systems, 1996., ISCAS '96, 'Connecting the World'., 1996 IEEE International Symposium
This paper presents a new systolic VLSI architecture for computing inverses and divisions in finite fields GF(2") based on a variant of Euclid's algorithm. It is highly regular, modular, and thus well suited to VLSI implementation. It has O(m2) area complexity and can produce one result per clock cycle with a latency of 8m-2 clock cycles. As compared to existing related systolic architectures w...
متن کاملCOMPUTING THE GENERIC INDEX OF THE CIRCUIT EQUATIONS OF LINEAR ACTIVE NETWORKS - Circuits and Systems, 1996., ISCAS '96, 'Connecting the World'., 1996 IEEE International Symposium
In many problems, such as the analysis of electrical circuits, one has to deal with Differential-Algebraic Equations (DAE’s). Knowing their generic index is of interest for practical and numerical as well as for theoretical reasons. In this paper we give an algorithm that determines the generic index of the Circuit Equations of linear active electrical networks in polynomial time from tlie netw...
متن کاملDesign of Cascade-Form IIR Filters with Powers-of-Two Coefficients Using Mixed Integer Linear Progra - Circuits and Systems, 1996., ISCAS '96, 'Connecting the World'., 1996 IEEE International Symposium
We propose an optimal design method for digital IIR filters with powers-of-two coefficients. This method is based upon the formulation of a linear optimization problem that minimizes filter’s complexity for given specifications. It is shown that by taking the logarithm to the transfer function of cascade-form IIR filters, the design problem becomes linear and can be solved by mixed integer line...
متن کاملA COST-EFFECTIVE VLSI ARCHITECTURE FOR HIGH-THROUGHPUT SEQUENTIAL DECODER - Circuits and Systems, 1996., ISCAS '96, 'Connecting the World'., 1996 IEEE International Symposium
This paper presents a new VLSI solution for high-speed digital communications based on long-constraint convolntional codes. The proposed VLSI architecture implements a modified sorter-based sequential decoding algorithm with an achievable maximum decoding rate of 25 Mbits/s. Unlike i ts previous version based on shiftable content addressable memory (SCAM) only, path recording is now implemented...
متن کامل