An analog local cluster neural net for a 3 V supply

Author

Körner, Tim ; Sitte, Joaquin ; Rückert, Ulrich

Author_Institution

Heinz Nixforf Inst., Paderborn Univ., Germany

fYear

1999

fDate

1999

Firstpage

292

Lastpage

298

Abstract

The local cluster neural net (LC net) is a feedfoward net suitable for continuous function approximation and discrete classification tasks. All operations of the LC net can be realized in analog circuits. Therefore we implemented the LC net in analog VLSI hardware for a 3 V power supply. Main applications of the LC net are control tasks in autonomous systems that can be battery powered. In this paper we describe the CMOS VLSI realization and present the results of the comprehensive test measurement for characterising the performance of the implementation

Keywords

CMOS analogue integrated circuits; VLSI; analogue processing circuits; feedforward neural nets; function approximation; integrated circuit noise; low-power electronics; neural chips; parallel processing; 3 V; CMOS VLSI realization; analog VLSI hardware; analog circuits; analog local cluster neural net; autonomous systems; battery powered application; continuous function approximation; control tasks; discrete classification tasks; feedfoward net; performance evaluation; test measurement; Analog circuits; Analog computers; Circuit testing; Concurrent computing; Control systems; Function approximation; Information technology; Neural networks; Parallel processing; Very large scale integration;

fLanguage

English

Publisher

ieee

Conference_Titel

Microelectronics for Neural, Fuzzy and Bio-Inspired Systems, 1999. MicroNeuro '99. Proceedings of the Seventh International Conference on

Conference_Location

Granada

Print_ISBN

0-7695-0043-9

Type

conf

DOI

10.1109/MN.1999.758877

Filename

758877