In various areas of modern society, the use of loudspeakers in a confined space, remains urgent problem of suppressing acoustic noise. It is known that if the system of sensor-receiver signal audible range, at a certain level, the gain in the system experiences a feedback loop that would create a spurious signal of high frequency. For example, the signal may vosprinimaetsya as \"sharp whistle\" was heard from the speaker. Known linearized approaches to the aforesaid problem, expressed in a number of devices, noise reduction, to some extent suppress acoustic feedback, but do not solve the problem completely.
The authors propose a new approach for raising the level of sound pressure to sound, closed prostranctve without distortion and without the appearance of excitation due to feedback. The method consists in the formation of useful output, as the differences and common parasitic signals.
In order to develop an algorithm to prevent excitation speaker, a series of experiments, the results of which were in the form of digital data files (the original signal and record the same signal is transmitted through the system gain and space), of which, by means of a comparative analysis, the authors can received pulse transient response is voiced space.
To obtain the results of the sound waves from the source to the receiver, was found mutual – correlation function (CCF) at intervals matching the two signals. CCF maxima were used as factors delay filter. Subtracting the signal values occurred with the correction factors that depend on the value of the function at its maximum for a particular order.
The results of the impulse response characteristics of finding and implementing filtering algorithms are displayed in the appropriate figures and tables.
Based on the analysis of the results should be noted that:
– It is possible to obtain a reasonable estimate of the impulse response characteristics of the signal dispersion in an enclosed space that by finding the cross-correlation function on adjacent intervals entering the receiving sensor signal.
– Revealed a clear enough parabolic relationship between power and an additional signal attenuation, which in turn leads to the adaptive filter.
– The occurrence of feedback loops affect the signals having the greatest Power steering, which, first of all, should include signals the forward direction, and as reflected by the most priblizhennnyh surfaces.
– There is a real opportunity to implement the proposed filtering algorithm based on the latest achievements mikroprotsesornoy techniques in digital signal processing in the form of specialized \"system on a chip\".
Galiev A. L., SHishkina A. F. Ustrojjstvo oslablenija akusticheskojj obratnojj svjazi s kompandirovaniem ogibajushhejj rechevogo signala // Promyshlennye ASU i kontrollery. 2011. № 6. S. 48 – 50.
Dzhonson D., Dzhonson Dzh. Spravochnik po aktivnym filtram. M.: EHnergoatomizdat. 2003.
SHtrapenin G. L. Operacionnye usiliteli: stremlenie k sovershenstvu // Kompanenty i tekhnologii. 2007. № 10. S. 46 – 50.
Nepomnjashhijj O. V., Kondratev K. V., Sergeevich V. N., Matjukha N. V., Dryzhak V. B. Metod podavlenija akusticheskojj obratnojj svjazi na osnove cifrovogo filtra predvaritelnogo vychislenija signala korrekcii // Informacionno-izmeritelnye i upravljajushhie sistemy. 2012. № 3. S. 19 – 22.
Glinchenko A. S. Cifrovaja obrabotka signalov: Ucheb. posobie. V 2-kh chastjakh. CH. 2. Krasnojarsk: IPC KGTU. 2001.
Kotkin G. L., CHerkasskijj V. S. Kompjuternoe modelirovanie fizicheskikh processov s ispolzovaniem MATLAB: Ucheb. posobie. Novokuzneck: Kuzbassvuzizdat. 2004.
Nepomnjashhijj O. V., Krasnobaev JU. V., Titovskijj S. N., KHabarov V. A. Mikroehlektronnye ustrojjstva upravlenija // ZHurnal Sibirskogo federalnogo universiteta. Ser.: Tekhnika i tekhnologii. 2012. T. 5. № 2. S. 162 – 168.