The pRAM: an adaptive VLSI chip

Clarkson, T.G.; Ng, C.K.; Guan, Y.

doi:10.1109/72.217182

Cited by 49 publications

(7 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…5 Setup column bus, if k > n -1; 2.6 PE(n-l j) broadcast(F(T(n-l j))), for 0 < j < N, if k > n -1; 2.7 0[layer+l](ij)=content(bus), for i = (k mod n), if k > n -1; end end One is for k < n and the other is for k > n, as described in the two clauses for the logical or operator in step 2.2. Steps 2.3 and 2.4 compute the partial sums for the weighted inputs and move the results to the next row.…”

Section: Second Mapping Methodsmentioning

confidence: 97%

See 1 more Smart Citation

Implementing and Mapping Anns on Reconfigurable Mesh Massively Parallel Architectures

Jenq

2002

Series in Machine Perception and Artificial Intelligence

View full text Add to dashboard Cite

Artificial neural networks (ANNs) have been used successfully for solving hard problems in many disciplines. Multi-layer feed-forward ANNs with back-propagation learning is one of the most widely used models. Methods for efficiently implementing and mapping this model on general-purpose, massively parallel reconfigurable mesh architectures are presented. IntroductionParallel implementation of artificial neural networks (ANNs) involves mapping of neural network models onto parallel architectures. Due to the large, diverse, and growing number of learning algorithms and architectures in the ANN literature, it is impossible to come up with a single classification scheme that can categorize all the essential features of all these paradigms. Nonetheless, ANN models are frequently characterized by their learning rules (supervised, un-supervised, hybrid), their network topology (feed-forward, recurrent, single layer, multi-layer), and neuron characteristic. Some of the most commonly used and discussed ANN models include multi-layer feed-forward networks with supervised 373 Neural Networks and Systolic Array Design Downloaded from www.worldscientific.com by MONASH UNIVERSITY on 04/23/17. For personal use only.

show abstract

Section: Second Mapping Methodsmentioning

confidence: 97%

“…Research and development efforts fall into three general categories: mapping and implementing ANNs directly in hardware [2,5], mapping and implementing ANNs on special-purpose processor architectures and computing devices that are designed for ANN simulation [21,22], mapping and implementing ANNs on generalpurpose parallel architectures [4,8,10,14,16,17].…”

mentioning

confidence: 99%

Implementing and Mapping Anns on Reconfigurable Mesh Massively Parallel Architectures

Jenq

2002

Series in Machine Perception and Artificial Intelligence

View full text Add to dashboard Cite

show abstract

“…However, the state of technology today is not mature enough for implementing large ANNs directly in hardware. A few, simple, small size ANNs have been implemented in VLSI as exemplified by those reported in [2], [5], [8], [9]. Most current implementations are software simulations on single processor systems.…”

Section: Introductionmentioning

confidence: 99%

Algorithmic mapping of feedforward neural networks onto multiple bus systems

El-Amawy

Kulasinghe

1997

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

This paper addresses the problem of mapping a feedforward ANN onto a multiple bus system, MBS, with p processors and b buses so as to minimize the total execution time. We present an algorithm which assigns the nodes of a given computational layer (c-layer) to processors such that the computation lower bound  l N /p p l t and the communication lower bound  l N /bt c are achieved simultaneously, where l N is the number of nodes in the mapped c-layer , and p l t and t c are the computation and communication times, respectively, associated with a node in the layer. When computation and communication are not overlapped, we show that the optimal number of processors needed is either 1 or p, depending on the ratio p l t /t c . When computation and communication are overlapped, we show that the optimal number of processors needed is either 1 or ( p l t /t c )b. We show that there is a unique arrangement of interfaces such that the total number of interfaces is minimum and the optimal time is reached. Finally, we compare the relative merits of the MBS simulating ANNs over the recently introduced checkerboarding scheme.

show abstract

“…have made them of extreme use in solving various mathematical problems. Neural networks have been successfully applied to broad spectrum of data-intensive applications, such as Signal Processing [1], Chip Designing [2], optimization problems [3] and in many engineering problems [4]. ANNs are also used for solving problems that are too complex for conventional technologies e.g., problems that do not have an algorithmic solution or for which an algorithmic solution is too complex to be found.…”

Section: Introductionmentioning

confidence: 99%

Performance Analysis of Cellular Radio System Using Artificial Neural Networks

Gupta¹,

Jain²

2017

AJNNA

View full text Add to dashboard Cite

Abstract:In this paper, we exploit one of the fastest growing techniques of Soft Computing, i.e. Artificial Neural Networks (ANNs) for obtaining various performance measures of a cellular radio system. A prioritized channel scheme with subrating is considered in which a fixed number of channels are reserved for handoff calls and in case of heavy traffic, these reserved channels are subrated into two channels of equal frequency to deal with more handoff calls. Two models dealing with infinite and finite number of subscribers are considered and the blocking probabilities of new and handoff calls are computed analytically as well as by using ANNs. A feedforward two-layer ANN is considered for obtaining the blocking probabilities. The backpropagation algorithm is used for training the ANN. The analytical and ANN results are compared by taking the numerical illustrations.

show abstract

The pRAM: an adaptive VLSI chip

Cited by 49 publications

References 5 publications

Implementing and Mapping Anns on Reconfigurable Mesh Massively Parallel Architectures

Implementing and Mapping Anns on Reconfigurable Mesh Massively Parallel Architectures

Algorithmic mapping of feedforward neural networks onto multiple bus systems

Performance Analysis of Cellular Radio System Using Artificial Neural Networks

Contact Info

Product

Resources

About