Active noise reduction system and method

Abstract

The invention discloses an active noise reduction system and a method. The active noise reduction system comprises a sound acquisition device, a motion signal acquisition device, a controller and a sound playing device, wherein the sound acquisition device is used for acquiring error noise signals in the environment; the motion signal acquisition device is used for acquiring motion signals at a specific position on a device generating primary noise signals and related to the primary noise signals; the controller is used for using a support vector machine to generate reference noise signals according to the motion signals, adjust the reference noise signals according to the error noise signals and then generate secondary noise signals; and the sound playing device is used for playing the secondary noise signals in the environment. The error noise signals are generated by overlapping the primary noise signals and the secondary noise signals. In the above active noise reduction system and the method, the support vector machine is used for predicting noise in the environment in real time. The generated reference noise signals have real-time performance and accuracy, and the noise reduction effects are improved significantly.

Description

technical field [0001] The invention relates to the field of signal processing, in particular to an active noise reduction system and method. Background technique [0002] With the development of the economy and the improvement of people's requirements for the quality of life, noise pollution is becoming more and more intolerable. However, many devices generate noise, especially in the environment around these devices, which can cause great distress to people. [0003] For example, more and more cars come into people's life. But the noise in the car has always been a major factor affecting people's ride comfort. In recent years, the noise in the car has become one of the important indicators to evaluate the comfort of the car and product quality, which will have a great impact on the consumer's car purchase orientation and the market competitiveness of the car products. [0004] In the prior art, passive and active noise reduction methods can control the noise to a certai...

AI Technical Summary

Problems solved by technology

The sound feedback problem will greatly affect the noise reduction effect of the active noise reduction system, and even cause the system to vibrate and fail to work in severe cases

[0008] 2. The reference noise signal cannot reflect the change of the original noise signal in real time

The resulting problem is that the reference noise signal cannot accurately reflect the original noise signal in real time, reducing the effect of noise cancellation

If the noise sample signal is too different from the original noise signal, it may even aggravate the noise

[0009] 3. The problem that the reference noise signal cannot be generated accurately

However, the systems in the actual application environment are nonlinear systems, so the reference noise signal generated by linear model prediction cannot accurately reflect the original noise signal

At present, there are also some papers that use various nonlinear models (artificial neural network models) to predict noise, but it is easy to fall into local minimum and other problems, which will also lead to inaccurate prediction results of nonlinear models, and cannot accurately Generate Reference Noise Signal

Therefore, the noise reduction effect is poor

Images