A neural network systems component

Mueller, Dietmar; Hammerstrom, Dan

doi:10.1109/icnn.1993.298738

Cited by 8 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each chip has 64 fixed point, RISC processors that currently operate at 20 MHz. These processors are designed to operate in an SIMD configuration where several CNAPS chips may be under the control of a single sequencer chip [4]. Each of the 64 processing nodes ('PNs) on each CNAPS chip has an adder, a multiplier, a logic unit, 4K bytes of local memory, several general purpose registers, and inter-PN bussing.…”

Section: Mavis Hardware Overviewmentioning

confidence: 99%

<title>Missileborne artificial vision system (MAVIS)</title>

1994

View full text Add to dashboard Cite

Several years ago when INTEL and China Lake designed the ETANN chip, analog VI.,SI appeared to be the only way to do high density neural computing.In the last five years, however, digital parallel processing chips capable of performing neural computation functions have evolved to the point of rough equality with analog chips in system level computational density. The Naval Air Warfare Center, China Lake has developed a real time, hardware and software system designed to implement and evaluate biologically inspired retinal and cortical models.The hardware is based on the Adaptive Solutions Inc. massively parallel CNAPS system COHO boards. Each COHO board is a standard size 6U VME car d featuring 256 fixed point, RISC processors running at 20 MHz in a SIMD configuration.Each COHO board has a Companion board built to support a real time VSB interface to an imaging seeker, a NTSC cam'era and to other COHO boards. The system is designed to have multiple SIMD machines each performing different Corticomorphic functions.The system level software has been developed which allows a high level description of Corticomorphic structures to be translated into the native microcode of the CNAPS chips. Corticomorphic structures are those neural structures with a form similar to that of the retina, the lateral geniculate nucleus or the visual cortex.This real time hardware system is designed to be shrunk into a volume compatible with air launched tactical missiles. Initial versions of the software and hardware have been completed and are in the early stages of integration with a missile seeker.

show abstract

Section: Mavis Hardware Overviewmentioning

confidence: 99%

<title>Missileborne artificial vision system (MAVIS)</title>

1994

View full text Add to dashboard Cite

show abstract

“…However, the use of neural network remains contingent on the availability of powerful hardware to provide adequate speed. Fortunately, the high density of modern technologies lets us implement a large number of identical, concurrently operating processors on one chip, thus exploiting the inherent parallelism of neural networks [6]. The regularity of neural networks and the small number of well-defined arithmetic operations used by neural algorithms greatly simplify the design and layout of VLSI circuits [7].…”

Section: Introductionmentioning

confidence: 99%

Implementation of a neural netwok module for fourth generation mobile equipements

Damergi

Rabaa²,

Bouallègue³

2006

International Conference on Design and Test of Integrated Systems in Nanoscale Technology, 2006. DTIS 2006.

View full text Add to dashboard Cite

Pipelined Recurrent Neural Network (PRNN) has been used with lot of success in many applications. In recent works, we have also proven that the PRNN exhibits good performances when used for interference cancellation and channel parameters estimation for the different multiple access schemes proposed as physical layer of the Fourth Generation (4G) networks: Wideband Code Division Multiple Access (WCDMA), Orthogonal Frequency Division Multiplexing (OFDM) and Multi Carrier CDMA (MC-CDMA). The use of a unique PRNN based module for the three multiple access techniques addresses a major challenge for the 4G mobile terminals: embedding many access techniques for reduced area and resources costs. In this paper, we investigate the feasibility of practical hardware implementation of the proposed structure, this paper aims at implementing the Pipelined Recurrent Neural Network structure by using the VHDL. VHDL is the name of the IEEE 1076 Hardware Description Language standard for very high-speed digital circuit design. The RTL/logic synthesis tool; Galileo has been used in order to generate the gate level of the proposed structure. The Xilinx Virtex II family is chosen as target technology.

show abstract

“…However, the use of neural network remains contingent on the availability of powerful hardware to provide adequate speed. Fortunately, the high density of modern technologies lets us implement a large number of identical, concurrently operating processors on one chip, thus exploiting the inherent parallelism of neural networks [4]. The regularity of neural networks and the small number of well-defined arithmetic operations used by neural algorithms greatly simplify the design and layout of 'V'LSI circuits [ 5 ] .…”

Section: Introductionmentioning

confidence: 99%

Implementation of an acoustic echo canceller based on NLMS-neural networks structures by using the VHDL

Rabaa

Zitouni²,

Abid³

et al.

ICC '98. 1998 IEEE International Conference on Communications. Conference Record. Affiliated With SUPERCOMM'98 (Cat. No.98CH362

View full text Add to dashboard Cite

In this paper, a cascaded Pipelined Recurrent Neural NetworMNLMS structure is employed in an attempt to model the roomfspeakerphone transfer function and acoustic path in a hands-gee environment. To investigate the feasibility of practical hardware implementation of the proposed structure, this paper aims at implementing the Pipelined Recurrent Neural Network and NLMS algorithms by using the VHDL. VHBL is the name of the IEEE 1076 Hardware Description Language standard for very high-speed digital circuits design. Galileo, an RTLAogic synthesis tool, has been used in order to generate the gate level of the proposed structure. The Xilinx family is chosen as target technology.

show abstract

A neural network systems component

Cited by 8 publications

References 4 publications

<title>Missileborne artificial vision system (MAVIS)</title>

<title>Missileborne artificial vision system (MAVIS)</title>

Implementation of a neural netwok module for fourth generation mobile equipements

Implementation of an acoustic echo canceller based on NLMS-neural networks structures by using the VHDL

Contact Info

Product

Resources

About