IEEE Trans Biomed Circuits Syst. 2011 Aug;5(4):380-90. doi: 10.1109/TBCAS.2011.2116786.
The bulkiness of current electronic nose (E-Nose) systems severely limits their portability. This study designed and fabricated an E-Nose signal-processing chip by using TSMC 0.18-μ m 1P6M complementary metal-oxide semiconductor technology to overcome the need to connect the device to a personal computer, which has traditionally been a major stumbling block in reducing the size of E-Nose systems. The proposed chip is based on a conductive polymer sensor array chip composed of multiwalled carbon nanotubes. The signal-processing chip comprises an interface circuit, an analog-to-digital converter, a memory module, and a microprocessor embedded with a pattern-recognition algorithm. Experimental results have verified the functionality of the proposed system, in which the E-Nose signal-processing chip successfully classified three odors, carbon tetrachloride (CCl4), chloroform (CHCl3), and 2-Butanone (MEK), demonstrating its potential for portable applications. The power consumption of this signal-processing chip was maintained at a very low 2.81 mW using a 1.8-V power supply, making it highly suitable for integration as an electronic nose system-on-chip.
目前的电子鼻（E-Nose）系统体积庞大，严重限制了其便携性。本研究采用 TSMC 0.18-μm 1P6M 互补金属氧化物半导体技术设计和制造了一种 E-Nose 信号处理芯片，克服了将设备连接到个人计算机的需要，这一直是减小 E-Nose 系统尺寸的主要障碍。所提出的芯片基于由多壁碳纳米管组成的导电聚合物传感器阵列芯片。信号处理芯片包括接口电路、模数转换器、存储模块和嵌入模式识别算法的微处理器。实验结果验证了所提出系统的功能，其中 E-Nose 信号处理芯片成功地对三种气味（四氯化碳（CCl4）、三氯甲烷（CHCl3）和 2-丁酮（MEK））进行了分类，展示了其在便携式应用中的潜力。该信号处理芯片在 1.8V 电源下的功耗保持在非常低的 2.81mW，非常适合集成作为电子鼻系统级芯片。
