International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395 -0056
Volume: 04 Issue: 04 | Apr -2017
p-ISSN: 2395-0072
www.irjet.net
Highly Reconfigurable Trimmed Mean Filter for Multiband Noise Cancellation Jefrin Hannah.D1, Nirmal Jothi.J2 1M.E.
Student, Dept. of ECE, SCAD College of Engineering and Technology, Cheranmahadevi, Tamil Nadu, India. Professor, Dept. of ECE, SCAD College of Engineering and Technology, Cheranmahadevi, Tamil Nadu, India.
2Assistant
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The active noise cancellation headphone use sound to cancel sound. The existing ANC system use Multi Band Filter (MBF) which provide variable and multi band filtering by tuning cutoff frequency. The existing system has an adaptive structure with single order and multi order coefficient generation. The Existing system use 41 logic elements and has the total power dissipation of .154W. This paper proposed the designing of trimmed mean filter for the cancellation of noise in the in-ear headphones. The trimmed mean filter is designed to cancel the multiband noise without affecting the actual input audio. The proposed design is expected to reduce the size and power consumption than the existing system.
Key Words: Active noise cancellation, multi band filter, trimmed mean filter, in-ear headphone, trimming.
1.INTRODUCTION Active noise cancellation is done by introducing anti noise wave through secondary sources which cancel out the noise wave. Destructive interference notion is applied. In [8] the basic adoptive algorithm for ANC is developed and analyzed based on single- channel broad band feed forward control. This algorithm is then modified for the narrow band feed forward and adaptive feedback control. These single channel ANC algorithms were then expanded to multiple channel cases for controlling the noise field in an enclosure or a large dimension duct. Various adaptive algorithms such as the lattice, frequency domain, sub band and RLS algorithms were also modified for ANC applications. [6] and [7] focuses on an active noise cancellation system for a home window using a transparent acoustic transducer. In a traditional active noise cancellation system, direct microphone measurements are used for reference and error signals. In the case of window application, both external and internal sound would be picked up by such microphones. This leads to adverse effects on the performance of the active noise cancellation system and also to distortion of the internal sound. To address this problem, [6] and [7] proposed a wave separation technique to separate the internal and external component of sound. The wave separation algorithm is based on the use of two microphones and an algorithm that © 2017, IRJET
|
Impact Factor value: 5.181
|
separates components based on their direction of travel. The theory for active cancelling of noise is simple, but the realization of an efficient ANC system is challenging due to several physical constraints. The existing ANC systems often use high-speed digital signal processors to cancel out disturbing noise, which results in high power consumption for a commercial ANC headphone. The contribution of the paper [3] can be classified into: 1) Proper filter length selection; 2) Low-power storage mechanism for convolution operation; and 3) High-throughput pipelining architecture. H-S Vu and K-H Chen [3] have developed an area-/powerefficient ANC circuit by using the TSMC 90-nm CMOS technology for in-ear headphone applications. The performance of ANC systems is determined by the magnitude response of the secondary path, which can be affected by the placement of the error microphone. In [9], the optimum location of the error microphone is subjected to the flat magnitude response of the secondary path, is found by experiments, and explained by the spectral filtering of pinna. The emphasis of S.M Kuo, S Mitra and W-S Gan [9] is on the design and experiment of the AFANC headphone in real time, with the goal to achieve higher noise cancellation as compared to high-end commercial ANC headphones. Multi-band filters are used to separate the frequency components of a signal and pass certain selected frequency ranges while filtering out other frequency ranges. It uses Z transform and inverse functions to achieve multiband filtering. In this paper a trimmed mean filter is designed to overcome all the shortcomings of the previous designs and to have better sound quality. Trimmed Median Filter (TMF) [5] is a decision based asymmetric filter. TMF is a two stage filter. First it detects the noisy pixels and then restores them. TMF considers all saturated pixels (0 or 255) as noisy pixels. If a pixel value lies within the dynamic range then it is considered a noise free pixel. Noise free pixels are left unchanged in the restoration stage. For each noisy pixel, the neighboring pixels within the 3X3 window are analyzed in the restoration stage. If all the pixels of the selected 3X3
ISO 9001:2008 Certified Journal
|
Page 854