-bit Serial Binary Addition with Linear Threshold Networks Sorin Cotofana and Stamatis Vassiliadis
نویسنده
چکیده
In this paper we investigate-bit serial addition in the context of feed-forward linear threshold gate based networks. We show that two n-bit operands can be added in 2d p n e overall delay with a feed-forward network constructed with d p n e + 1 linear threshold gates and 1 2 (5d p n e 2 + 9d p n e) + 2 latches. The maximum weight value is 2 d p n e and the maximum fan-in is 3d p n e+1. We also investigate the implications our scheme have to the performance and the cost under small weights and small fan-in requirements. We deduce that, if the weight values are to be limited by a constant W, two n-bit operands can be added in logW] + n log W] overall delay with a feed-forward network that has the implementation cost logW] + 1, in terms of linear threshold gates, 1 2 (55logW] 2 + 99logW]) + 2 in terms of latches, and a maximum fan-in of 33logW] + 1. We also prove that, if the fan-in values are to be limited by a constant F + 1, two n-bit operands can be added in F 3 + n F 3 ] overall delay with a feed-forward network that has the implementation cost F 3 + 1, in terms of linear threshold gates, 1 2 (5 F 3 2 + 9 F 3) + 2 in terms of latches, and a maximumweight value of 2 F 3 ]. An asymptotic bound of O(n logn) is derived for the addition overall delay in the case that the weight values have to be linearly bounded, i.e. in the order of O(n). The implementation cost in this case is in the order of O(log n), in terms of linear threshold gates, and in the order of O(log 2 n), in terms of latches. The maximum fan-in is in the order of O(logn). Finally, a partition technique, that substantially reduces the overall cost of the implementation for all the schemes in terms of delay, latches, weights, and fan-in with some few additional threshold gates, is also presented.
منابع مشابه
Low Weight and Fan-In Neural Networks for Basic Arithmetic Operations
In this paper we investigate low weight and fan-in neural networks for the precise computation of some basic arithmetic operations. First we assume one bit per serial cycle LSB first operand reception and introduce a pipeline network performing serial binary addition in O(n) time constructed with 11 threshold gates, a maximum weight of 2 and a maximum fan-in of 4. Further we prove that serial m...
متن کاملSerial Binary Addition with Polynominally Bounded Weights
This paper presents a new approach to the problem of serial binary addition with feed-forward neural networks. It is shown that the serial binary addition, an important component to a number of applications such signal processing, of two n-bit operands with carry can be done by a feed-forward linear threshold gate based network with polynomially bounded weights, i.e. in the order of O(n k), ass...
متن کامل2-1 Additions and Related Arithmetic Operations with Threshold Logic
In this paper we investigate the reduction of the size for small depth feed-forward linear threshold networks performing binary addition, comparison, and related functions. For n bit operands we propose a depth-3 O(n 2 log n) asymptotic size network for the binary addition with polynomially bounded weights. We propose also a depth-3 addition of optimal O(n) asymptotic size network and a depth-2...
متن کاملPeriodic symmetric functions, serial addition, and multiplication with neural networks
This paper investigates threshold based neural networks for periodic symmetric Boolean functions and some related operations. It is shown that any n-input variable periodic symmetric Boolean function can be implemented with a feedforward linear threshold-based neural network with size of O(log n) and depth also of O(log n), both measured in terms of neurons. The maximum weight and fan-in values...
متن کامل1 Redundant Addition with Threshold Logic
In this paper we investigate threshold based networks performing 2 1 addition using redundant binary operand representations. In particular we concern with establishing the networks with the smallest network depth assuming small weight and fan-in values. First we propose a depth-4 implementation and we show that the 2 1 addition can be computed by a threshold network with the size in the order ...
متن کامل