Multiplex communications – Duplex – Transmit/receive interaction control
Reexamination Certificate
1999-09-27
2003-06-17
Cangialosi, Salvatore (Department: 2661)
Multiplex communications
Duplex
Transmit/receive interaction control
C370S290000
Reexamination Certificate
active
06580696
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of echo cancellers, and more particularly, to a multi-adaptation echo canceller. The present invention is also directed to a novel method and apparatus for canceling echoes associated with voice packet networks.
BACKGROUND OF THE INVENTION
In Time Division Multiplex (TDM) based telephone systems, echoes are generated when a sender's speech is reflected back to the sender from a hybrid interface.
FIG. 1
illustrates a conventional hybrid interface
4
between a two wire line that connects a near end subscriber
2
to the hybrid interface
4
and a four wire line (not shown) used for transmission of analog or digital signals within a local exchange. Any speech at point
6
from a far end subscriber that re-emerges at point
8
via the hybrid interface
4
is an echo. Speech at point
8
is transmitted to the far end subscriber and may include speech from the near end subscriber
2
and echoes generated by the hybrid interface
4
. Such echoes may be annoying, and under certain conditions can completely disrupt a conversation between the far end subscriber and the near end subscriber. To attenuate such echoes, digital filters are generally used in TDM networks.
Conventional echo cancellation techniques that are currently available are improvements in this field, but they do not eliminate the echoes completely. One conventional method uses a transformer with a number of passive elements. Echoes occur in telecommunication networks due to impedance mismatches at hybrid transformers that couple the two wire line to the four wire line. Ideally, the hybrid transformer transmits the far end subscriber's speech signals at the four wire receive port through to the two wire transmit port without leakage into the four wire transmit port. However, this would require exact knowledge of the impedance seen at the two wire port, which varies widely and can only be estimated. As a result, leakage signals in the form of echoes are transmitted to the far end subscriber.
Another conventional method, an adaptive digital echo canceller, will now be described with reference to FIG.
2
.
FIG. 2
is a block diagram illustrating an arrangement having a conventional adaptive echo canceller. In general, adaptive digital filters are used to replicate different impulse responses associated with telephones and to compensate for variations in the impulse responses caused by changes in the subscriber loops.
Digital speech signals at point
32
from the far end subscriber are converted to analog signals by a digital to analog converter
22
before the signals reach the hybrid interface
4
. Speech signals from the near end subscriber
2
and/or echo signals generated by the hybrid interface
4
are converted to digital signals by an analog to digital converter
24
. The digital signals are then transmitted from the A/D converter
24
to a transmission delay
26
, which is caused by other transmission devices. The digital speech signals from the near end subscriber
2
and/or the echo signals generated by hybrid interface
4
are then transmitted to a subtractor
30
.
Simultaneously, an adaptive filter
28
, i.e. Least Means Squares (LMS) digital filter
28
, receives the far subscriber speech signals from point
32
. Filters using other algorithms such as Recursive Least Squares (RLS) may be substituted for the LMS digital filter. The adaptive filter
28
generates synthetic echo signals based on the real speech signals. The synthetic echo signals are then subtracted from the real echo signals using the subtractor
30
, and thus reducing the echo signals transmitted to the far end subscriber. Subtracting the synthetic echo signals from the real echo signals will reduce the amplitude of the echo, and the remaining signal after this subtraction is called the “error signal.” The error signal is forwarded to the filter
28
where the coefficients of the filter
28
are adjusted accordingly in an effort to minimize future echoes from being transmitted to the far end subscriber. The magnitude of the echo and the time delay constitute the echo path transfer function.
Adaptive methods such as that described above generally rely on information contained in the speech signals from the far end subscriber. Echo cancellers have been used for TDM networks, which generally uses a sample by sample adaptation. Adaptive filters generally include coefficients that are adjusted based on the LMS or RLS algorithm. The coefficients are updated in a manner opposite to the gradient of the error signal multiplied by a constant number. This constant number is usually referred as an adaptation step size, which effectively controls adaptation speed. However, a large adaptation step size results in stability problems, which sometimes goes into signing or oscillation situations. Such problems get worse when the step size is increased excessively in order to increase adaptation speed.
There are generally two key performance parameters of an echo canceller: ERLE (echo return loss enhancement) and converge time. The ERLE is the degree to which the echo canceller suppresses the echo signal, i.e. the ratio between the echo signal and the error signal measure in dB. The converge time is the time required to reach the ERLE of 26 dB or greater. Currently, the ITU requirement for the converge time is less than 500 ms for a continuous excitation and less than 1 second for speech like bursts.
With the advent of voice packet technology, i.e. Voice over Internet Protocol (VoIP), echo cancellers are becoming increasingly important. This is because the latency and transmission delay of an IP network is much longer than a traditional TDM network. Because the human's perception of an echo signal is proportional to the delay it experiences, controlling its ERLE and converge time performance is highly desirable for a good voice quality.
There are differences in the design of echo cancellers for TDM and IP networks. For example, the echo canceller in the TDM network basically uses a sample by sample process, whereas an echo canceller for an IP network would use a block by block or packet by packet process.
Thus, there is a need for a system and method that can provide echo cancellation by way of a block by block or packet by packet process and can increase converge speed while maintaining robust stability.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide an echo canceller that can provide stability and fast converge time.
It is another object of the present invention to provide an echo canceller that can provide a multi-adaptation, compare, and select process for canceling echoes.
It is yet another object of the present invention to provide an echo canceller having multiple adaptive filters, and thus allowing the system to choose the optimal filter.
It is yet another object of the present invention to provide an echo canceller that can cancel echoes in an IP network.
It is yet another object of the present invention to provide an echo canceller that can cancel echoes using a block by block or packet by packet process.
These and other objects of the present invention are obtained by providing a multi-adaptation echo canceller that transfers coefficients from a first echo canceller to a second echo canceller as the initial coefficients for the second echo canceller. The system compares the single and double adaptation echo error signals and selects the echo canceller with the smaller error signal. This process is repeated so that the system can select the optimal echo canceller for canceling echoes associated with each voice block/packet. The present invention can be implemented with N number of adaptation echo cancellers so long as the each block or packet can be processed within a specified time frame.
REFERENCES:
patent: 4757527 (1988-07-01), Beniston et al.
patent: 4862450 (1989-08-01), Guidoux
patent: 4885737 (1989-12-01), Guidoux
patent: 5418848 (1995-05-01), Armbruster
patent: 5555310 (1996-09-01), Minami et al.
patent: 5812944 (1998-09-01), Mats
Chen Herbert M. K.
Liu Fuling
Cangialosi Salvatore
Cisco Systems Inc.
LandOfFree
Multi-adaptation for a voice packet based does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multi-adaptation for a voice packet based, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multi-adaptation for a voice packet based will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3134507