Multiplex communications – Wide area network – Packet switching
Patent
1983-06-16
1986-02-18
Orsino, Jr., Joseph A.
Multiplex communications
Wide area network
Packet switching
1791702, 1791706, H04J 1500, H04L 514
Patent
active
045717192
DESCRIPTION:
BRIEF SUMMARY
TECHNICAL FIELD
The present invention relates to a balance filter included in a telecommunication system for duplex data transmission. More precisely, there is intended a filter which is incorporated in an echo canceller for inhibiting the effect of an echo signal from the transmit side via the hybrid coupler to the receive side.
BACKGROUND ART
A method and apparatus are described in the Swedish Patent Application No. 8106444-6 for achieving adaptive echo cancellation. The principle of the apparatus will be apparent from the appended FIG. 1. The data sequences b.sub.n and a.sub.n are transmitted on the conductor pair to and from the remote end, repectively. The local end receive units not only receive the signal w(t) from the remote end but also a signal h(t) coming from the local transmitter S. In practical cases, the amplitude of the h(t) signal may be very much greater than that of the w(t) signal. To enable correct detection in the detector D a signal y(t) is subtracted from the received signal. The former is the output signal from the balance filter B, the input signal of which comes from the data sequence b.sub.n. The balance filter B is adjusted adaptively with the aid of the correlator unit KB so that y(t) and h(t) become equal in the sampling instants kT. After subtraction there only remains the w(t) signal from which data from the remote end can be extracted.
The balance filter should be of the FIR type (Finite Impulse Response). The output signal of such a filter is defined thus: correction of the filter parameters: a correction vector.
The conventional implementation of an FIR filter utilizes the vector multiplication: ##EQU1## Here the number of parameters is thus N.
Another way in which to realize a filter of the FIR type is to utilize table look-up. All conceivable output signals are stored in a memory. If each input signal sample x.sub.n can assume L discrete levels the number of possible output signals will be:
For the memory size to be reasonable, L and N must be relatively small. The implementation can be interesting, particularly in binary data transmission (L=2). The filter input signal vector x.sub.k with N components determines an address j which points out a definite filter parameter c.sub.j in the memory.
In the table look-up the correction vector will be:
It will be seen from the above that the table filter results in large memory space if N is large (the memory size increases exponentially with n). Since in many practical cases it may be necessary to have a relatively large value of N, the table filter discussed above may be unsuitable.
The transmitter unit S, coupler G and lowpass filter LP as well as other units not shown in FIG. 1 in the transmitter-receiver unit give rise to nonlinearities. The signal w(t) coming to the hybrid coupler G from the remote end is comparatively weak. The signal h(t) from the transmitter unit S through the coupler G (the leal signal) is, on the other hand, at least at the beginning of transmit-receive sequence, relatively strong before echo cancellation has managed to come into action. In practical cases, the signal h(t) may be much greater than w(t), e.g. a hundred times as great. Since the signal h(t) passed a nonlinear signal path, components from a given transmitted data symbol b.sub.k arise in the signal h(t), for which a linear balance filter normally cannot compensate. Since the leak signal h(t) can be very much greater than the signal w(t) from the remote end, a minor nonlinearity in the system may also be devastating for echo cancellation if the nonlinearity cannot be compensated.
DISCLOSURE OF INVENTION
In accordance with the present invention, the total filter length N in the balance filter is divided into two, possibly more, parts where N=N1+N2, the part of the filter which is to compensate the greatest echo signal amplitude values, and thereby the greatest nonlinear components also, has a relatively short filter length and table look-up is utilized for this part of the filter. Remaining parts N2 may utilize table look-up or vect
REFERENCES:
patent: 4162378 (1979-07-01), Baudoux et al.
patent: 4261051 (1981-04-01), Ohnishi et al.
patent: 4425483 (1984-01-01), Lee et al.
Mueller--"A New Digital Echo Canceler"--IEEE Trans. on Comm. vol. COM-24, #9, Sep. 1976, pp. 956-962.
Holte et al.--"A New Digital Echo Canceler"--IEEE Trans. on Comm. vol.-COM 29, No. 11, Nov. 1981, pp. 1573-1579.
Carlqvist Bengt R.
Svensson Lars T. E.
Orsino Jr. Joseph A.
Telefonaktiebolaget L M Ericsson
LandOfFree
FIR-Type balance filter incorporated in the transmitter-receiver does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with FIR-Type balance filter incorporated in the transmitter-receiver, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and FIR-Type balance filter incorporated in the transmitter-receiver will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1554341