Gunjan PandeyHarini Sundararaman
Sudhanshu Vyas
OutlineObjectivesArchitectureDesign StagesIssues (Expected and Faced)ProgressFinal week targets
ObjectivesThe aim is to receive speaker's voice + noise at
both the microphone which are kept at a known distance and then try to make an estimate of noise using the microphone which is farther from the speaker.
The known information includes the x(t)+ n(t) value at M1 and x(t + ∆t1) + n(t + ∆t2) at the other microphone.
Current assumption is that both the microphones are far enough so that speaker's voice doesn't reach the microphone no 2
Architecture
Design Stages
Theory
Issues We Expected• Synchronization of the microphones• ADC• Position of noise source with respect
to the system• microphones characteristics• Level of Noise the system will be
able to handle• Proximity of the microphones wrt each
other
Additional Issues FacedNo tech support for AD1981BChip is in secondary mode on the ML507Doubt regarding two microphone supportMade a generic controller instead of specific
oneCustom debug methods
ProgressAssumed Actual
Study time Mid-NovemberArchitecture design Mid-NovemberImplementationBoard Testing
October 30th
November 14th
November 21st November 25th
November 29th December 3rd
Made AC'97 controller
Integrated with MP0 for testing
Read register values from AD1981B
Long way to go
Plans for last weekTest Implementation of register writesRemove UARTChange estimator algorithm
Questions