Active Noise Control For Broadband And Narrowband Indoor Noise Based On Online Secondary Path Identification

Noise pollution has seriously affected the production and living conditions of humankind.The low frequency noise with high energy is especially harmful.So it is particularly important to control noise effectively.With the progress of digital signal research and electronic circuit technology,active noise control method has developed rapidly in recent years.Owing to its excellent physical characteristics and good control effect,ANC becomes more and more popular.The traditional active noise control system divides the ambient noise into broadband and narrowband,and designs the broadband ANC system and the narrowband ANC system respectively.In recent years,the new hybrid control algorithm is devoted to the control of wide and narrow band noise,which has developed rapidly with its excellent control effect.However,the existing active noise control algorithms still have some difficulties,for example,the secondary channel modeling error is large,the primary noise is non-stationary,slow convergence,poor stability,low anti-interference ability and less research on indoor noise problem.In this paper,the research of indoor noise active control system is carried out.Firstly,after analyzing the influence mechanism of secondary path modeling error on ANC control effect,this paper designs a feedback wide and narrowband online modeling active noise control algorithm based on FXLMS adaptive algorithm,and adopts variable step strategy and improvement.The cost function approach optimizes the ANC algorithm.Afterwards,simulation experiments on MATLAB software verify the effectiveness of the algorithm.Secondly,the active noise controller is built on the TMS320c6748 DSP,including hardware structure design and software development.The hardware structure mainly includes three major modules:primary noise input,algorithm processing,and secondary signal output.To ensure the real-time performance of the controller,the ping-pong transmission mode of the EDMA3 is used to carry data.In addition,NAND FLASH is used for data access,and GPIO is used to complete the key switch function.The software development uses CCS software,mainly including module initialization,hardware parameter configuration,algorithm logic and other programming work.Finally,multiple sets of indoor noise control experiments were conducted to test the performance of the built controller.The results show that the controller has a noise reduction of about 20 dB for single-frequency sine wave noise and double sine wave superimposed noise,and a noise reduction of about 10 dB for broadband indoor noise.In addition,the ANC space control experiment shows that the position of the secondary path has a great influence on the noise reduction effect of the active noise control system.When the secondary sound source is placed on the same side as the noise source,the space control effect is optimal.