实时最小二乘算法的神经网络实现

Neural Networks Implementing Via Real-time Least Squares Algorithms

本文提出了一种求解最小二乘问题的神经网络模型。理论分析和模拟结果表明：该网络是稳定的并且能在电路时常数数量内获得近似解。

In this paper, we presents a neural networks modeling to solve real-time least squares(RLS)problems. We show that by theoretical analysis and by simulatlons this networks isstable . and that it is guaranteed to provide arbitrarily approximate solutlons to the accurateRLS ones within a few of the circuit time constants of this network,therefore this network isvery applicable to process the RLS problems.