SAM
SAM Benefits
SAM vs. Broadside Array Microphone
SAM in Automotive Hands-free Applications
SAM in VoIP Applications
SAM in Handheld Applications
History of Adaptive Filter
Strength & Pending Patents
 

1940s - Wiener filter

1960s - Kalman filter

1960s - LMS (Least Mean Square)

1970s - NLMS (Normalized LMS)

1980s - Affine Projection

2000s - Fortemedia SAM's Adaptive Step Size Sub-band, Non-Linear Filter (AS3NLF)

Chronicle
Echo cancellation has been through significant changes since the Wiener filter technology became available in 1940. The Kalman filter and LMS (Least Mean Square) algorithm replaced the Wiener filter during 1960s, because the Wiener filter was not suitable for real time applications. The Kalman algorithm increased the computational requirements. This in turn raised the cost of implementation in embedded systems. The LMS algorithm had less stability when applied to real human voice communication applications. A modified and more stable version of the algorithm, NLMS (Normalized LMS), was a big milestone that became prevalent in the ‘70s, while the Affine Projection algorithm was introduced in the 80’s to accelerate the convergence times. Today, with cancellation in both time and frequency domain and innovative non-linear process, Fortemedia SAM AS3NLF adaptive filter provides superior echo cancellation.

[ Back to TOP ]
© 1997-2004 Fortemedia, Inc. All Rights Reserved. Site MapTerms of Use & Disclaimers