Analysis of Embedding Dimension of Built-in sensor Nonlinear Output

This paper studies the nonlinear characteristics of the output signal of a built-in sensor in a smart device when the speech signal is excited. According to the lyapunov exponent calculation, the chaotic characteristics of the output signal are verified. Based on the theory of phase space reconstruction, the FNN and Cao algorithms are used to solve the embedding dimension. Comparing the two results, when the speech signal is used as an input, the embedding dimension of the nonlinear signal output by the sensor is more accurate than the FNN by the Cao solution. It laid the foundation for the subsequent establishment of relevant mathematical models.