Telephonic communications – Echo cancellation or suppression – Using digital signal processing
Reexamination Certificate
1999-06-03
2003-09-30
Nguyen, Duc (Department: 2643)
Telephonic communications
Echo cancellation or suppression
Using digital signal processing
C379S406080, C379S406120
Reexamination Certificate
active
06628781
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to communications systems, and more particularly, to echo suppression in bi-directional communications systems.
BACKGROUND OF THE INVENTION
Adaptive filtering arrangements are prevalent in communications systems of today. Such arrangements are typically used to reduce or remove unwanted signal components and/or to control or enhance signal components of interest.
A common example of such a filtering arrangement relates to hands-free telephony, wherein the built-in earphone and microphone of a conventional telephone handset are replaced with an external loudspeaker and an external microphone, respectively, so that the telephone user can converse without having to physically hold the telephone unit in hand. Since sound emanating from the external loudspeaker can be picked up by the external microphone, adaptive filtering is commonly performed in order to prevent the loudspeaker output from echoing back and annoying the far-end user at the other end of the conversation. This type of adaptive filtering, or echo canceling, has become a basic feature of the full-duplex, hands-free communications devices of today.
Typically, echo cancelation is achieved by passing the loudspeaker signal through an adaptive Finite Impulse Response (FIR) filter which approximates, or models, the acoustic echo path between the hands-free loudspeaker and the hands-free microphone (e.g., a passenger cabin in an automobile hands-free telephony application). The FIR filter thus provides an echo estimate which can be removed from the microphone output signal prior to transmission to the far-end user. The filtering characteristic (i.e., the set of FIR coefficients) of the adaptive FIR filter is dynamically and continuously adjusted, based on both the loudspeaker input and the echo-canceled microphone output, to provide a close approximation to the echo path and to track changes in the echo path (e.g., when a near-end user of an automobile hands-free telephone shifts position within the passenger cabin).
Adjustment of the filtering characteristic is commonly achieved using a form of the well known Least Mean Square (LMS) adaptation algorithm developed by Widrow and Hoff in 1960. The LMS algorithm is a least square stochastic gradient step method which, as it is both efficient and robust, is often used in many real-time applications. The LMS algorithm and its well known variations (e.g., the Normalized LMS, or NLMS algorithm) do have certain drawbacks, however. For example, the LMS and other known algorithms can sometimes be slow to converge (i.e., approach the target filtering characteristic, such as the acoustic echo path in a hands-free telephony application), particularly when the algorithm is adapted, or trained, based on a non-white, or colored, input signal such as a human speech signal. Moreover, the order of the FIR filter (i.e., the number of filter taps) can be quite high in the context of acoustic echo cancelation, and implementation of the adaptive filtering algorithm can therefore be computationally complex.
Consequently, recent work has focused on performing the adaptive filtering in sub-bands. In other words, filter banks are used to divide both the microphone signal and the loudspeaker signal into a number of frequency sub-bands. Each sub-band signal is then decimated, or down-sampled, and adaptive filtering is performed in each sub-band to provide a number of echo-canceled sub-band output signals. The resulting sub-band output signals are then expanded, or up-sampled, and combined to reconstruct the overall echo-canceled microphone signal for transmission to the far-end user. Advantageously, the sub-sampling results in greater computational efficiency as compared to the full-band processing approach and, since variations in the spectral content of the input signals are less severe within each sub-band, overall convergence speed is also improved.
However, known sub-band adaptive filtering systems suffer from certain disadvantages as well. For example, signal aliasing between sub-bands can result in slow overall convergence and/or errors in the reconstructed microphone signal. Consequently, there is a need for improved methods and apparatus for performing sub-band adaptive filtering in echo suppression systems.
SUMMARY OF THE INVENTION
The present invention fulfills the above-described and other needs by providing sub-band adaptive filtering techniques wherein the bandwidth of each of a number of sub-band analysis filters (i.e., bandpass filters used to divide an echo-containing signal and/or an echo causing signal into a number of adjacent frequency sub-bands), as well as the bandwidth of each of a number of synthesis filters (i.e., bandpass filters used in reconstructing a full-band echo-canceled signal from a number of sub-band echo canceled signals), is increased as compared to a corresponding bandwidth in a conventional system. More specifically, −3 dB bandwidths of adjacent filters in an analysis or synthesis filter bank according to the invention are designed to overlap one another. For example, whereas each of M sub-band filters in certain conventional analysis and synthesis filter banks includes a −3 dB bandwidth of 1/M times the total bandwidth of interest (M an integer), each of M sub-band filters in an analysis or synthesis filter bank according to the invention can include a −3 dB bandwidth which is somewhat greater than 1/M times the total bandwidth of interest. Advantageously, the increase in bandwidth tends to cancel aliasing effects resulting from down-sampling, and a sub-band adaptive filtering device according to the invention can therefore provide superior output signal reconstruction as compared to conventional sub-band adaptive filtering devices.
An exemplary communications device according to the invention includes an adaptive echo canceler receiving a near-end audio signal and a far-end audio signal and providing an echo-canceled near-end audio signal for transmission to a far-end user via the communications channel. The adaptive echo canceler includes a first bank of analysis filters for filtering the near-end audio signal, a second bank of analysis filters for filtering the far-end audio signal, and a bank of synthesis filters for filtering sub-band echo-canceled signals generated within the adaptive echo canceler.
According to the invention, a −3 dB bandwidth of a first of the synthesis filters and a −3 dB bandwidth of a second of the synthesis filters overlap in frequency. For example, the adaptive echo canceler can include an integer number M of synthesis filters, wherein −3 dB bandwidths of the M synthesis filters collectively span an overall bandwidth B, wherein a −3 dB bandwidth of one of the synthesis filters is centered at each one of M frequencies (the M frequencies being distributed across the bandwidth B at intervals of B/M), and wherein the −3 dB bandwidth of each of the synthesis filters is greater than B/M and less than 2B/M.
Additionally, a −3 dB bandwidth of a first analysis filter in the first bank of analysis filters and a −3 dB bandwidth of a second analysis filter in the first bank of analysis filters can overlap in frequency, and a −3 dB bandwidth of a first analysis filter in the second bank of analysis filters and a −3 dB bandwidth of a second analysis filter in the second bank of analysis filters can also overlap in frequency. For example, each of the first and second banks of analysis filters can include an integer number M of analysis filters, wherein −3 dB bandwidths of the M analysis filters in each bank of analysis filters collectively span an overall bandwidth B, wherein a −3 dB bandwidth of one analysis filter in each bank of analysis filters is centered at each one of M frequencies (the M frequencies being distributed across the overall bandwidth at intervals of B/M), and wherein the −3 dB bandwidth of each analysis filter is greater than B/M and less than 2B/M.
REFERENCES:
patent: 4956838 (1990
Ehrenstråle Fredrik
Grundström Patrik
Burns Doane Swecker & Mathis L.L.P.
Nguyen Duc
Telefonaktiebolaget LM Ericsson (publ)
LandOfFree
Methods and apparatus for improved sub-band adaptive... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and apparatus for improved sub-band adaptive..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and apparatus for improved sub-band adaptive... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3077599