JPS6343014B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention relates to an echo canceller that cancels echoes generated in a 2-wire to 4-wire conversion section of a telephone line.

(Prior art) Conventionally, in the case of long-distance lines via communication satellites, telephone lines have generally been constructed using a four-wire line (receiving line 5 and transmitting line 6) as shown in Figure 1 to amplify the signal. (forming a 4-wire line) is inserted. In FIG. 1, 1 is a telephone, 2 is a hybrid circuit, 3 is a pseudo-echo generator, 4 is an adder, 14 is a subscriber line (two-wire line), 15 is an echo canceller, and 15 is an echo canceller. The two-wire line) 14 and the four-wire lines 5 and 6 are tangentially connected through a hybrid circuit 2. If the impedance matching between the hybrid circuit 2 and the subscriber line 14 is perfect, the received signal x(t) will not leak to the transmission line 6 side, but since the impedance of the subscriber line 14 varies depending on the line type, line length, etc. , it is difficult to always achieve perfect consistency. Therefore, the received signal x
A part of (t) leaks to the transmitting side, causing an echo.

Echo can significantly degrade speech quality as the propagation time on the line increases. Furthermore, in recent years, dramatic advances in integrated circuit technology have led to the digitization of various devices, and even in telephone lines, high-speed digital transmission is being adopted, which aims to provide a wider range of services, instead of the conventional analog transmission method that mainly uses the voice band (4kHz band). A shift to this method is being considered in various fields, but although this would shorten the line propagation time, the echoes described above may cause data bit errors and degrade transmission quality. An echo canceller is used to eliminate the deterioration in transmission quality caused by this echo. As shown in FIG.
The echo is forcibly canceled by generating y^(t) and subtracting it from the echo y(t).

The configuration of the pseudo echo generating section 3 of the conventional echo canceller 15 is shown in FIG. In Figure 2, 5
is the receiving line, 7-1, 7-2,..., 7-(N-
1) are delay elements, 8-1, 8-2,..., 8-
N is a multiplier, 9 is an adder, and the received signal x(t)
is input to the delay element 7-1, and is sequentially input to the delay element 7-1.
2,..., 7-(N-1), and at the same time, the delay elements 7-1,..., 7-(N-1) pass through the taps coming out from between each delay element 7-1,..., 7-(N-1) to the respective corresponding multipliers. 8-1,..., 8-N, each coefficient h^ _k (k=0,...,
N-1) and the results are added in an adder 9 to generate a pseudo echo y^(t).
Each coefficient h^ _k (k=0,
..., N-1) is obviously set equal to the impulse response h _k (k=0,..., N-1) of the echo path. In the case of the conventional configuration of the pseudo-echo generation section, an analog delay line is required to configure it using an analog system, so it is not suitable for LSI implementation, and as described above, to configure it using a digital system. Considering the high-speed digitization of telephone lines, problems arose in the processing speed and power consumption.

(Objective of the Invention) An object of the present invention is to achieve high processing speed and high quality transmission by connecting R and C primary circuit networks in parallel, multipliers, and adders in order to eliminate the drawbacks of the prior art described above. The present invention provides an echo canceller that can be obtained, and will be described in detail below.

(Structure of the Invention) The present invention comprises N RC primary circuit networks connected in parallel between an input end of a received signal and a common potential point, each RC primary circuit network consisting of a resistance and a capacitance. The k-th (where 1≦k≦N) RC primary circuit has a time constant of CR/k with RC as a constant.

Further, a multiplier is provided corresponding to each of the RC primary circuit networks, and outputs a signal obtained by multiplying the signal at the connection point between the resistor and the capacitor in the RC primary circuit network by a coefficient d _k , and a transmitter. summing means for subtracting the output signals of all the multipliers from the echoes to the side, the N coefficients d _k being determined to minimize the power of the echo residual at the output of the summing means; It is something to do.

(Example) FIG. 3 is an explanatory diagram showing an example of the present invention,
10-1, 10-2, ..., 10-N are resistors each having a resistance value R/k (k = 1, 2, ..., N), 11
-1,11-2,...,11-N is capacity, 12-
1, 12-2,..., 12-N are each coefficient d _k (k=
1, 2, . . . , N), 13 is an adder, and other symbols are the same as in FIG. Received signal x
(t) is a R, C circuit (resistor 1
0-1 and capacitance 11-1, resistance 10-2 and capacitance 11
-2,..., each consisting of a resistor 10-N and a capacitor 11-N), and after passing through each R and C circuit, they are added together in an adder 13 to produce a pseudo echo y^
(t) is generated. The principle of the present invention will be described below.

Impulse response h(t) appearing in the echo path
In order to approximate
h^(t) is generated by h^(t)= _N 〓 ^k=1 C _k φ _k (t)...(1). Here {φ _k (t)} (k=1,
2,...,N) form an orthonormal function system,
Each φ _k (t) is given by the following equation.

φK(t)=√2 _K 〓 ⁱ⁼¹
(-1) ^Ki (i+K-1)! /(i-1)! (K-i
)! i! e ^-ipt …(2-1) = _K 〓 ⁱ⁼¹ a _ki e ^-ipt …(2-2) where It is. The _coefficient C _k ( _k =1, ^{... ,} N) in equation (1) ^is the square error power J shown by the following equation h^(t)} ² dt...(3) is determined to be the minimum. That is, C _i is ∂J/∂C _i = ∂/∂C _i ∫ ^∞ ₀ {h(t)−h^(t)} ² dt =−2∫ ^∞ ₀ h(t)φ _i (t)dt+2C _i =0 C _i =∫ ^∞ ₀ h(t)φ _i (t)dt (i=1,...,N) ...(4). C _i (i=1,
..., N), equations (2-2), and equations (1), the pseudo echo generation circuit shown in FIG. 3 can be obtained. However, each coefficient _dk (k=1,...,N) of the multipliers 12-1, . . . , 12-N is calculated from the following equation.

d _k = _N 〓 ^i=k a _iK゜C _i /KP …(5) Equation (5) is expressed using the coefficient C _i as a medium, but the third
Since the echo canceller shown in the figure has only the coefficient d _k , the coefficient d _k is determined so as to minimize the power of the echo residual.

Next, specific numerical examples will be shown. The subscriber line is a paper insulated cable with a wire diameter of 0.5 mm and a total length of 2 km, with a terminating resistance of 110 Ω.As the received signal x(t), a unit impulse band-limited to 150 kHz is used to generate a pseudo echo. The part is N=7, P=
FIG. 4 shows the frequency characteristics of the pseudo echo and the echo frequency characteristics when 1/CR=2π×200×10 ³ /7. In Fig. 4, the frequency characteristics of the echo shown by the solid line and the frequency characteristics of the pseudo echo from the pseudo echo generation section according to the present invention completely match up to 20kHz, and even in the low frequency range beyond that, it is almost practical. This shows a degree of approximation that does not cause any problems. Therefore, the echo can be canceled by subtracting the pseudo-echo generated by the pseudo-echo generator according to the present invention from the echo. This embodiment is composed of R, C circuits, multipliers, and adders, and has the advantage of being able to provide an echo canceller that eliminates the drawbacks of the prior art.

(Effects of the Invention) Since the echo canceller according to the present invention can be configured with R, C circuits, multipliers, and adders, it can be implemented as an LSI, which has the advantage of improving processing speed and power consumption.

[Brief explanation of the drawing]

Figure 1 is an explanatory diagram showing the outline of an echo canceler that cancels echoes generated in the 2-wire to 4-wire conversion section of a telephone line, Figure 2 is a configuration diagram of the pseudo echo generation section of a conventional echo canceller, and Figure 3 is FIG. 4 is an explanatory diagram showing an embodiment of the present invention, and is an explanatory diagram comparing the frequency characteristics of a pseudo echo and the actual echo frequency characteristics. DESCRIPTION OF SYMBOLS 1... Telephone, 2... Hybrid circuit, 3... Pseudo-echo generation part, 4, 13... Adder, 5, 6... 4-wire line, 7-1,..., 7-(N-1)... Delay element, 8-1,...,8-N...multiplier, 10-1,
..., 10-N...resistance, 11-1,..., 11-N...
Capacity, 12-1,..., 12-N... Multiplier, 14...
Subscriber line, 15...Echo canceller.

Claims (1)

[Claims] 1. N RC primary circuit networks connected in parallel between the input end of the received signal and a common potential point, each of which consists of a resistance and a capacitor. , and the k-th (1≦k≦N) RC primary circuit is RC
10, which has a time constant of CR/k with CR/k as a constant.
11, and a multiplier 12, which is provided corresponding to each of the RC primary circuit networks and outputs a signal obtained by multiplying the signal at the connection point between the resistor and the capacitor in the RC primary circuit network by a coefficient d _k . and adding means 4, 13 for subtracting the output signals of all the multipliers from the echoes to the transmitting side, and the N coefficients _dk calculate the power of the echo residual at the output end of the adding means. An echo canceller characterized by being determined to minimize.