Amoeba-inspired nanoarchitectonic computing implemented using electrical Brownian ratchets

M. Aono, S. Kasai, S. J. Kim, M. Wakabayashi, H. Miwa, M. Naruse

研究成果: Article査読

28 被引用数 (Scopus)

抄録

In this study, we extracted the essential spatiotemporal dynamics that allow an amoeboid organism to solve a computationally demanding problem and adapt to its environment, thereby proposing a nature-inspired nanoarchitectonic computing system, which we implemented using a network of nanowire devices called 'electrical Brownian ratchets (EBRs)'. By utilizing the fluctuations generated from thermal energy in nanowire devices, we used our system to solve the satisfiability problem, which is a highly complex combinatorial problem related to a wide variety of practical applications. We evaluated the dependency of the solution search speed on its exploration parameter, which characterizes the fluctuation intensity of EBRs, using a simulation model of our system called 'AmoebaSAT-Brownian'. We found that AmoebaSAT-Brownian enhanced the solution searching speed dramatically when we imposed some constraints on the fluctuations in its time series and it outperformed a well-known stochastic local search method. These results suggest a new computing paradigm, which may allow high-speed problem solving to be implemented by interacting nanoscale devices with low power consumption.

本文言語English
論文番号234001
ジャーナルNanotechnology
26
23
DOI
出版ステータスPublished - 2015 6月 12
外部発表はい

ASJC Scopus subject areas

  • バイオエンジニアリング
  • 化学 (全般)
  • 材料科学(全般)
  • 材料力学
  • 機械工学
  • 電子工学および電気工学

フィンガープリント

「Amoeba-inspired nanoarchitectonic computing implemented using electrical Brownian ratchets」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル