ULCEAR Digital Microphone Array eliminates the need for a close-talk microphone with a superior approach to noise suppression. It separates wanted audio information from unwanted noise with the use of multi stage processing:
implemented using multiple DSP based algorithms. Its purpose is to create an extremely versatile algorithm able to handle a huge variety of noise types and other acoustical issues disrupting mobile communications. This technology is well suited for implementation in any area where hands-free communications are required.
Our patented” Adaptive-Beam Forming” algorithm is used to optimize the SNR of the wanted signal and continue to track the user’s voice within the Sweet Spot Area. This ensures that the signal taken in by stage 1 is high in desired signal as noise and interference free as possible. Stage 1 processing does simple signal classification outputting classification results and filter parameters estimation for stage 2 processing. When comparing the UCLEAR solution to conventional beam former widely implemented our adaptive solution:
“Adaptive Interference-Cancellation” is performed in this stage, to suppress man generated noise. As man generated noise contains directional information, the array is able to identify those noise sources as outside the ‘sweet zone’ thus cancelling them. The cancellation is done by adaptively creating a ‘null’ or ‘dead spot’ in the direction of each identified noise source (Figure 1).The number of noise sources that can be suppressed by nulls is limited by the bandwidth of the noise source and the processing power of the DSP.
At the end of this stage, the array has separated out the desired signal from the noise and interference, using directional cues and other information embedded within the received audio information from the logarithmically spaced microphone array. Both the desired signal and the noise and interference are then sent to a third stage for processing.
▲ Drawings of polar response curve of sweet spot and others interference nulls
This stage of the signal processing primarily resolves the natural noise sources and other non-directional noise source induced into the desired audio information. These noise types are diffused or otherwise lack directional information. They would be identified and suppressed dynamically along with the signal level of the induced natural noise in the frequency domain. Also performed in this stage, is some form of signal recovery to the wanted signal information, speech enhancement, before outputting the final processed signal.
The operations in this stage are performed by a mathematically optimized algorithm that suppresses the natural noises from human speech has been developed to perform such non-direction noise suppression. This differs greatly from the commonly implemented simple spectral subtraction resulting in voice distortion – an issue that the BITwave algorithm does not face. This is because the algorithm manipulates the signal in both frequency and time domain and is able to suppress a lot more accurately and retaining maximum intelligibility compared to other common forms of frequency domain only suppression algorithms.
Together, these technologies produce astonishingly levels of noise suppression and interference cancellation never achieved before. The result of processing different types of noise by different intelligent adaptive processes layer by layer, as opposed to a single juncture cancellation, results in superior quality of the desired voice and greater suppression of unwanted noise being delivered. The resultant suppression of unwanted noise sources by the Digital Microphone Array with DSP processing is on average 24 dB and may go as high as 30+dB.
The two wave forms below illustrates the before and after effects of the Microphone Array Technology.