基于神经网络和开关电容技术提出了一种并行算法型ＣＭＯＳ模数转换器电路结构．文中详细分析了电路的工作原理，并采用通用电路模拟程序进行了仿真．结果表明转换功能正确，输入电压递增和递减均无错码，转换过程中不需要复位．该转换器充分发挥开关电容精确处理和神经网络并行处理的优点，因而精度高、速度快，而且所需元件少，工艺完全与常规ＣＭＯＳ兼容，是一种有巨大发展潜力的新型ＣＭＯＳ转换器 Based on the neural network and switched capacitor technology, a parallel arithmetic CMOS analog-digital converter circuit structure is proposed. In this paper, the working principle of the circuit is analyzed in detail, and the simulation is carried out by the general circuit simulation program. The results show that the conversion function is correct, the input voltage increases and decreases without error code, the conversion process does not require reset. The converter gives full play to the advantages of switching capacitor precision processing and neural network parallel processing, so high precision, fast, and less components required, the process is fully compatible with conventional CMOS is a great potential for development of a new type of CMOS converter