Active Noise Control Projects (ANC)

Outline of the problem.

Based on the figure below, the aim is to use the cancellation speaker to achieve y=0 using signals from the two microphones. To analyse this control problem, let the transfer function from the cancellation speaker to the cancellation area be F, from the noise source to the cancellation area G, and from the cancellation speaker to Microphone I H. Microphone I is not necessarily in close proximity to the noise source but is assumed to be somewhere between the noise source and the cancellation area. Real-time estimates of G, F , H can be used to compute a suitable control signal for the cancellation speaker. This simplified outline model neglects the geometry of wave interference, which is important to take into account in order to achieve good cancellation.

Complete Solutions

Current Projects (MPhil, PhD, Postdoctoral)

• Noise cancellation in air ducts
• Noise cancellation in vehicle interiors
• Noise filtering/cancellation for music listeners
• Airport noise protection
• Quiet living rooms
• Noise protection in open-plan offices
• Active vibration control of plates (various applications)
• Fuzzy-logic applied to intelligent supervision of control systems.
• Worst-case theory of the frquency domain adaptive methodology.
• Piezo-electric actuator enhancement
• High performance electromagnetic actuators.
• Information available upon request at s.m.veres@bham.ac.uk

Advanced features of the control schemes implemented in the Active Sound and Vibration Control (ASVC) Laboratory

• Frequency selective adaptive control
• Guaranteed stability of the control scheme based on real-time assessment of model uncertainty
• Good rejection of disturbances with novel techniques based on ellipsoidal and polytope gain bounding
• Vibration insulation and the delayed resonator.

Hardware interface

Experimental hardware is interfaced to an LSI TMS320C31 Real-Time Applications Board with A/D and D/A daughter modules. The control algorithms are usually implemented with sampling rates of 2kHz-20kHz to control sound below 1kHz. Real-time Processing Harware

MATLAB -based development system

• Prototyping of algorithms is carried out in MATLAB/SIMULINK. MATLAB compiler is used to generate C-code for microprocessors. Development system.
• Many routines are used from the Geometric Bounding Toolbox (GBT) and the Signal Processing Toolbox (SPT).