【摘要】 本文提出了一种新的利用跨导运放OTA精确实现电压-电流到硬限幅函数神经元方法，它的主要特点是容易实现、且神经元的输出幅值可调。本文所提的基于OTA的硬限幅函数神经元（简称OHL）非常适合于求解L1问题的神经网络跨导模式综合实现。实例和计算机仿真结果与理论分析相吻合。更多还原

【Abstract】 In this paper, a new approach for exact implementation of neuron with signum transfer function using Operational Transconductance Amplifier (OTA) is proposed. The main advantage of the proposed approach in comparison with the voltage or transconductance comparator is its simple implementation. The proposed OTA-Based Hard Limiter (OHL) is very well suited for VLSI implementation of neural networks for solving least absolute value problem. A good agreement with theoretical analysis has been obtained by PSPICE simulation results.更多还原