MOS-BJT-NDR 元件之設計與應用

Design and Applications of MOS-BJT-NDR Devices

溫峻明(Chun-Ming Wen)

指導教授 ： 陳耀煌

崑山科技大學/工程學院/電子工程研究所/碩士(2006年)

摘要

共振穿透二極體(RTD)元件是以化合物半導體的技術製作而成,其材料有高電子移動率等優點,它已經廣泛和成熟應用在高速數位邏輯閘及類比式電路上,那是由於其獨特的負微分電阻的電壓-電流特性曲線的關係,而本篇論文中我們所製作的負微分電阻電路,是以金屬氧化半導體場效電晶體(MOS)元件所架構而成的,其結構是由兩個N型MOS元件、一個BJT元件、一個P 型MOS 元件所組成, 因此我們稱此負微分電阻電路為MOS-BJT-NDR元件,以其來實現RTD負微分電阻的電壓-電流特性曲線,並且利用這些基本的負微分電阻元件來模擬和製作不同元件電路為負載的多值記憶器、解碼器與多工器、除頻器與細胞式類神經網路(CNN)電路。

關鍵字

負微分電阻元件 ； 多值記憶器 ； 細胞式類神經網路除頻器 ； 解碼器與多工器

並列摘要

The resonant-tunneling-diode (RTD) device is fabricated by the technique of compound semiconductor. This kind of NDR device is possess of the high electron mobility. Resonant tunneling diodes (RTD’s) have found various applications in high-speed digital and analog circuits due to their
specific advantages associated with the unique folded-back negative differential resistance (NDR)I-V characteristics. In this thesis, we proposed a new NDR device which is composed of metal-oxide-semiconductor field-effect-transistor (MOS) devices. Because this NDR device is consisted of MOS and BJT devices, we call his novel NDR device as MOS-BJT-NDR device. The MOS-BJT-NDR device can realize the negative resistance I-V curve of RTD. Because MOS-BJT-NDR devices are fabricated by standard CMOS processes, we can easily use them to design many electronic circuits of applications. Among them, we are interesting in the multiple-valued memory circuit, decode circuit, multiplexers circuit, frequency divider circuit, and the cellular neural network. The simulations and fabrications of multiple-valued memory circuit and discrete-time cellular neural network are finished by using
MOS-BJT-NDR devices.

並列關鍵字

Decode ； Negative Differential Resistance Devices ； Cellular Neural Network ； Multiplexers ； Multiple-ValuedMemory ； Frequency Divider

參考文獻

[1]R.Tsu and L.Esaki,“Tunneling in a finite superlattice”,

1979.pp.1094-1101.

device with multiple negative differential resistance: Digital and signal

roomtemperature peak-to-valley ratio,” IEEE Electron Device Letr., vol. 9,

[5]S. J. We; and H. C. Lin, “Multivalued SRAM cell using resonant