RU2406224C2 - Method, device and software for evaluating properties of transmission line of communication system - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: properties of a transmission line are evaluated (204) from the relationship between a signal transmitted over the line (200) and the resultant received signal (201) or signals at frequencies for which the absolute value of the propagation coefficient of the line multiplied by the length of the line is less than π. The input impedance of the line is expressed through linear constants (capacitance, resistance, induction and conductivity of the line) using Taylor's theorem. After dropping higher order terms and including measured values, the resultant system of equations is solved and linear constants are obtained provided that at least two frequencies are used.

EFFECT: high accuracy ensuring operation on long lines.

18 cl, 6 dwg

Description

FIELD OF TECHNOLOGY

The present invention relates to the field of transmission line analysis.

BACKGROUND OF THE INVENTION

For the operation of communication networks, it is of great interest to measure the properties of transmission lines of networks. The results of such measurements can be used, for example, to detect failures, determine the location of failures, predict some failures that are close to occurrence, and evaluate the suitability and throughput of a line for some services such as digital subscriber line (DSL).

Many methods and devices can be used to evaluate line properties.

Single-ended line testing, SELT, is an important test method. In this method, the properties of the line are evaluated from measurements taken at only one end of the line.

One of the SELT equipment classes is specialized measuring equipment with physical access for testing. The measuring device for special purposes during the measurement is galvanically connected to the transmission line.

Such measurements can be carried out in various ways. The capacitance can be estimated by applying a voltage to the line and turning it off, while measuring the voltage drop time.

The line length can be measured, for example, by sending a pulse and measuring the arrival time of the reflected pulse, the so-called reflection coefficient measurement method, TDR.

However, galvanic access requires special measures. It can be provided either by connecting manually to the line to be measured, or using the linear panels of the communication center serving the lines that need special equipment (for example, relays) to provide galvanic access when necessary. This is a major disadvantage.

In addition, the test device can usually perform one measurement at a time. The use of many such test devices is expensive, and it may be problematic to simultaneously provide galvanic access to several lines, several test devices, depending on the organization of providing galvanic access.

A more attractive solution is to get access to the lines through conventional signal paths of the linear panels of communication stations for testing. Access can be obtained to many lines simultaneously, while specialized equipment is not required.

Existing line panels often have built-in functionality for simple line testing, for example, measuring resistance and voltages between the cores of a pair and between each cores and the ground. However, the accuracy of such embedded functionality is often low.

PCT / SE2004 / 000718 discloses a method in which a transmission line length can be determined by analyzing the relationship between a sent signal and a received resulting signal.

The ratio is adjusted to account for the effect of the transceiver on the signals, and the input line impedance is calculated as a function of frequency. The length is then determined by the periodic behavior of the absolute value of the impedance. The method does not work well for line lengths above about 1.5 km and uses frequencies in the range from 30 kHz to 1 MHz.

A reflected pulse, similar to a pulse in a TDR measurement, can be calculated by the inverse Fourier transform of the line input impedance, in which the line input impedance is first determined, as described above. Broadband input line impedance measurement required. In this way, line lengths of up to about 6.4 km can be measured. The implementation of this method is disclosed in EDA 2.1 Line Testing, EN / LZT 108 7773 R1A, Ericsson AB, March 2005.

Known methods for assessing the properties of transmission lines of communication systems suffer from the disadvantages that require specialized equipment or physical access during testing, have low accuracy or do not work for long lines. Typically, only one or more line properties can be estimated from a single measurement result.

SUMMARY OF THE INVENTION

An object of the present invention is to remedy these drawbacks when evaluating transmission properties for a transmission line of communication systems.

An object of the present invention is to provide a method independent of specialized equipment or physical access for testing, having the required accuracy, working for longer lines, and capable of: evaluating more properties from a single measurement. An additional task is to create a device and software product for performing the method.

The problem is solved by creating a method in which the signal is sent to the transmission line, the resulting signal is received, while the properties of the transmission are evaluated by the ratio between the sent and received signals, and the sent signal contains at least two frequencies for which the absolute value of the constant line propagation times the line length is less than π.

If at the time of measurement it is not known whether the frequencies used meet this criterion, a first estimate can be made using a predetermined set of frequencies. Frequencies for use in the second or subsequent further evaluation are then selected based on the results of the first or previous evaluation.

In slightly more detail, the properties of the line can (as described above), in general, be determined from the relationship between the transmitted signal and the received resulting signal. Today it is determined that if the transmitted signal meets a certain criterion, then longer lines can be measured than was previously possible. At the same time, reasonable accuracy is achieved and several properties can be estimated from the results of a single measurement. If desired, the line panel used for normal line operation can be used to transmit and receive signals without the need for physical access for testing. The line panel can even be a standard POTS panel.

The criterion is that the transmitted signal (or signals) contains at least two frequencies for which the absolute value of the line propagation constant multiplied by the line length is less than π. When this criterion is fulfilled, it becomes possible to calculate the properties of the line in a different way than the known one, and giving better results in important aspects.

The method is as follows. The input impedance of the line, as a function of frequency, is determined from the ratio between the transmitted and received signals. For a frequency range that meets the criterion, Taylor series expansion can be used to express the line input impedance using constant lines and frequency. After discarding higher-order terms and substituting the measured values, the resulting system of equations can be solved, and the line constants are obtained provided that there are at least two frequencies. If there are more than two frequencies, the system of equations is overridden and the solution is found using the least squares method. More frequencies usually give better accuracy.

Thus, an advantage of the present invention is that the method can be used using conventional signal paths (e.g., voice paths in the case of a POTS panel) through a line panel for normal transmission line operation, and thus no physical access during testing not required.

An additional advantage is that the present invention uses the range of POTS, and it can be used through the POTS panel.

The advantage of using POTS panels is that such panels are widely distributed in communication centers.

Another advantage is that if the POTS range is used for measurement (whether through the POTS panels or not), this can be done without breaking the current DSL schedule.

Additionally, an important advantage is that the properties can be evaluated for longer lines than was previously possible using a comparison between the transmitted and received resultant signals measured through the line panel.

Also, the advantage of using the POTS band (i.e., relatively low frequencies) is that the attenuation of the signal in the transmission line is usually lower for lower frequencies. Low attenuation means that an acceptable received signal power can be obtained for longer lines, that is, longer lines can be measured.

Another advantage is that the present invention has better accuracy than the measuring functionality built into the POTS panels of the prior art.

Also, the advantage is that more parameters can be determined from the results of a single measurement than before.

An additional advantage is that many measurements can be carried out simultaneously, without requiring large expenses. Large volumes of measurements can be planned and carried out automatically on a regular basis.

Another advantage is that line panels can be made cheaper if relays that provide galvanic line access to specialized test equipment are no longer needed.

Another advantage is that the calculations used in the method in accordance with the invention do not impose high requirements in numerical terms and can be implemented with low power computing tools.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is further explained in the description of the preferred embodiments with reference to the accompanying drawings, in which:

Figure 1 depicts a block diagram of a simplified installation of a Central communication node according to the invention;

Figure 2, 3 is a block diagram of a simplified structure of a linear POTS panel according to the invention;

Figure 4 is a graph of frequency versus cable length for a mode where the absolute value of the propagation constant times the line length is π for various types of ESTI cables according to the invention;

5 is a simplified block diagram of a linear POTS panel, transmission line and load impedance according to the invention;

6 is a flowchart of a method for determining transmission properties of a transmission line of a communication system.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The transmission lines 60 (Fig. 1) of a communication system are usually loaded at one end at a site 61 in the client’s premises, and at the other end, at the installation 130 of the central communication center. Here, the lines are usually connected to the line panels 100. The line panels 100 are usually controlled by the communication unit processor 110. Of great interest is the ability to measure the properties of lines 60 from the installation 130 of the central station.

The initial steps of the present method are sending a signal on a transmission line 60 and receiving a resulting signal. The ratio of the signals then calculates the complex input line impedance for each of the many frequencies present in the signals. By this impedance, other transmission properties of the line can be determined.

The following describes how to practice the method when signals are sent and received through the line panel 100 and when the line panel is a POTS line panel.

With some simplification, the POTS line panel 10 has the structure shown in FIG. 2. The transmission line 60 is connected to an analog input device 50, which is connected to a digital-to-analog (D / A) converter 30 and an analog-to-digital (A / D) converter 40.

The analog input device contains amplifiers and other analog circuits, such as a linear transformer and hybrid circuits for separating the sent and received signals.

The analog-to-digital and digital-to-analog converters together with the analog input device form a transceiver 20.

The signal S _in (t), which must be transmitted on line 60, is digitally transmitted to the transceiver 20, where in the digital-to-analog converter 30 it is converted to an analog voltage, which is transmitted through the line through an analog input device 50. An input signal is received from the line through an analog the input device by an analog-to-digital converter 40, which converts it into a digital signal and outputs from the transceiver as a signal S _out (t).

For measurement purposes, the measurement signal MS _in (t) can be digitally transmitted to the transceiver and then transmitted over the transmission line. The resulting signal will be received from the line and digitally transmitted by the transceiver as a signal MS _out (t).

The digital signals MS _in (t) and MS _out (t) are usually represented in the time domain. They can be converted to frequency domain representation using the usual fast Fourier transform (FFT), in which the signal MS _in (t) is converted to the signal V _in (f) and the signal MS _out (t) is converted to the signal V _out (f) .

The ratio between the transmitted and received signals as a function of frequency is called the transfer function of the reflected signal (echo), H _echo (f). It is defined as follows:

H _echo (f) = V _out (f) / V _in (f)

The transfer function of the reflected signal H _echo (f) is complex and depends on both the characteristics of the line and the characteristics of the transceiver. The signals V _in (f), V _out (f) are complex Fourier transforms of the transmitted and received signals.

Therefore, to determine the input impedance of the line Z _in (f) from the transfer function of the reflected signal H _echo (f), the characteristics of the transceiver 20 should be considered.

The effect of the transceiver, for the purpose of the method, can be characterized by three calibration parameters, Z _ho (f), Z _hyb (f) and H _∞ (f). All of them are complex and frequency dependent.

The parameter H _∞ (f) is the frequency-dependent transfer function of the reflected signal for a transceiver with an open line connection, i.e. no line connection. The parameter Z _hyb (f) is the transceiver impedance, measured at the line panel line connections, that is, the panel impedance when viewed from the line side. The parameter Z _ho (f) can be expressed as Z _ho (f) = Н ₀ (f) · Z _hyb (f), where the parameter H ₀ (f) is a complex and frequency-dependent transfer function of the reflected signal for a transceiver with short-circuited connections lines, and the parameter Z _hyb (f) are determined as described above.

In addition to the circuit of the analog input device 50, the transceiver 20 may also provide digital filtering of signals in the digital filters 70 and 80 (FIG. 3). The signal to be transmitted is then digitally filtered by the line panel before being transmitted, and the received resulting signal is digitally filtered after it has been converted by the analog-to-digital converter 40. For the purpose of the method, such filters may be turned off or remain on. In the latter case, the effect of digital filtering of signals can, of course, be included when the values of the calibration parameters are determined in order to compensate for the effect of the filters on the signals.

The signals MS _in (t) and MS _out (t) (as well as V _in (f) and V _out (f)) then refer to the transmitted signal before filtering and to the received signal after filtering, respectively (Fig. 3).

The procedure for determining and using the calibration parameters Z _ho (f), Z _hyb (f) and H _∞ (f) is described in sufficient detail in patent applications PCT / SE2004 / 000296, PCT / SE2004 / 000566 and PCT / SE2004 / 000718 (published as WO 2004/100512, WO 2004/100513 and WO 2004/099711).

The complex frequency-dependent input impedance of the transmission line from the side of the line panel interface can be calculated as

Z _in (f) = (Z _ho (f) -Z _hyb (f); H _echo (f) / (H _echo (f) -H _∞ (f))

where H _echo (f) = V _out (f) / V _in (f)

Thus, when the signal V _in (f) has been transmitted and the signal V _out (f) has been received, the line input impedance can be determined as shown above.

It was indicated above that a voltage is applied and the received resulting voltage is measured. However, in principle, one or both of the signals can be currents, so that the line input impedance can be calculated from the relationship between the signals.

At least two frequencies are required to satisfy the convergence criterion explained below. More frequencies usually give better accuracy. Of course, it is possible to transmit and receive several signals when the signals together contain the required number of frequencies, provided that it is possible to determine the input impedance of the line.

Signals can be transmitted and received by specialized measuring equipment or through conventional transmission paths of the linear panel to which the line is connected. The line panel may be a POTS type line panel, as described above, or another type.

Depending on the type of signals and equipment used, the formula for determining the input impedance of a line may differ from that described above.

Determination of line constants by input impedance

The transmission line can be described by specific parameters per unit length: series resistance R, series inductance L, shunt conductance G and shunt capacitance C (for example, in units of Ohm / km, GN / km, S / km and F / km, respectively) . They are called the primary parameters of the cable type.

For the frequencies used for the present invention, the primary parameters can reasonably be taken as constant and independent of the frequency.

The propagation constant γ and the characteristic impedance Z _{0 are} determined through the primary parameters R, L, G and C, that is,

и

,

and

,

respectively.

For open at the end of the transmission line of length d, the input impedance can be expressed as follows:

Z _in = Z ₀ coth (γd)

Using the expansion of γdcoth (γd) in a Taylor series and after some transformations, the input impedance can be expressed as follows:

=R*d,

=L*d,

=G*d,

=C*d, ω=2πf и f обозначает частоту.Where

= R * d,

= L * d,

= G * d,

= C * d, ω = 2πf and f denotes the frequency.

,

,

и

называются константами линии. Разложение в ряд действительно (или сходится) для |γd|<π.Specific values

,

,

and

are called line constants. Series expansion is valid (or converges) for | γd | <π.

In practice, the shunt conductivity of the transmission line can be neglected (G = 0). Similarly, members of a higher order, except for the first three or at most four members, are negligible. This allows us to express Z _in as a pure polynomial:

where the coefficients a _k can be explicitly related to the line constants, that is:

Representation of the line input impedance as a frequency polynomial simplifies the estimation of line constants for a subscriber line before solving ordinary equations. The quadratic error function can be represented as follows:

where Z _in (jω _m ) is the spectral component of the measured Z _in , which is used for estimation, and N is the total number of spectral components (frequencies) used for estimation.

,

, и

.The solution for the coefficients a _-1 ... a ₂ (for example, the usual least-squares method to find a set of coefficients that give the minimum value of ε ² ) leads to an estimate with minimal dispersion of the line constants, such as

,

, and

.

In the future, we consider the real and imaginary parts of the above equation separately, that is:

.

.

Conventional equations can be solved separately in the frequency domain for the real and imaginary parts, since even coefficients are associated exclusively with the real part, while odd coefficients are associated with the imaginary part. Only two spectral components (frequencies) of the input impedance of the Z _in line are required to find solutions for these four coefficients a _k . More frequencies will usually give better accuracy.

,

, и

могут быть определены следующим образом:When the coefficients a _{k are} determined, the constants

,

, and

can be defined as follows:

or alternatively

из a₁, а не из а₂ обычно является лучшим выбором с точки зрения точности.From experience, definition

of a ₁ rather than a _{2 is} usually the best choice in terms of accuracy.

Line length determination

линии известна, длина линии d вычисляется с помощью известного или вероятного значения С (емкость на единицу длины линии), например 50 нФ/км.When the specific capacity

the line is known, the line length d is calculated using the known or probable value C (capacitance per unit line length), for example 50 nF / km.

Frequency selection

The Taylor series expansion used in expressing the line input impedance is valid for | γd | <π (radius of convergence). Therefore, the frequencies should be chosen so that | γd | <π. This is shown in FIG. 4, which shows graphs | γd | = π for various ETSI cables. The area below and to the left of the curve for a certain type of cable is the area in which the convergence criterion is performed for a given cable at different cable lengths and frequencies.

The choice of frequencies of the measuring signal, therefore, depends on the length and primary parameters of the transmission line, which may be unknown. To solve this problem, a wide range of frequencies is first used for a rough estimate of the length and line constants. A more accurate estimate is then made using frequencies that satisfy the convergence criterion.

,

,

и

(где

можно пренебречь, то есть принять равной нулю) можно определить, какие из используемых частот удовлетворяют критерию сходимости |γd|<π, где для каждой частоты

.This can be done in the following sequential way. The first assessment is done using many frequencies. From the estimates obtained in this way

,

,

and

(Where

can be neglected, that is, taken equal to zero), it is possible to determine which of the frequencies used satisfy the convergence criterion | γd | <π, where for each frequency

.

(Thus, the fulfillment of the convergence criterion can be estimated directly from the line constants without first estimating the line length.)

Those frequencies that do not satisfy the convergence criterion are excluded, and the calculation is repeated. And again, those frequencies that according to the new estimate do not satisfy the convergence criterion are excluded from the calculations. The process is repeated until a set of frequencies remains for which estimates of the line constants indicate that the convergence criterion is satisfied for all frequencies used for the estimation.

It is also possible to start the evaluation with only a few low frequencies and sequentially add higher frequencies to the evaluation until the evaluation shows that the next frequency to be added does not meet the convergence criterion. For example, the first score may be based on tones 3 and 4 (as defined below) and higher tones will be added sequentially.

Accuracy will usually improve with each added tone, mainly due to the reduced effect of noise. However, the third tone will usually make the greatest contribution to improving accuracy, and the contribution of subsequent tones will be less. Therefore, an estimate based on three frequencies is optimal in some implementations.

In one embodiment, the following simple rules have been used. A range of frequencies, starting from 125 Hz and with an interval of 125 Hz up to 3625 Hz, is defined as tones 1-29.

The first rough estimate is made using all tones. Then

- for cables with an estimated length of more than 4 km, tones 1-14 (125-1750 Hz) are used,

- for cables with an estimated length of less than 3 km, tones of 14-29 (1750-3625 Hz) are used,

- in other cases, tones 1-25 (125-3125 Hz) are used.

In another embodiment, a first rough estimate is made using all tones, and then

- for cables with an estimated length of more than 6 km, tones 3-13 (375-1625 Hz) are used,

- for cables with an estimated length of more than 3 km, tones of 6-26 (750-3250 Hz) are used,

- in other cases, tones 3-15 (375-1875 Hz) are used.

Other factors that can be considered when choosing frequencies are frequency-dependent noise (noisy frequencies can be avoided) and the fact that at very low frequencies the conductivity G cannot be neglected for all types of cables (very low frequencies can be avoided).

For example, to avoid low frequencies, frequencies higher than or equal to 125 Hz or, for example, higher or equal to 375 Hz can be used.

Additionally, if you do not use specialized measuring equipment, it may not be possible to use direct current signals (for example, when sending signals through most linear panels), which in this case may impose a limitation that the frequency must be greater than zero.

Achieved and achievable accuracy

,

, и

была проверена на имитаторах кабелей и на реальных кабелях на кабельных барабанах, а также на реальных линиях передачи в полевых условиях.Rating

,

, and

It was tested on cable simulators and on real cables on cable reels, as well as on real transmission lines in the field.

и

можно оценить с точностью приблизительно 2-6%. Кабели длиной до 11 км были измерены с ошибками приблизительно 15-20%. Точность оценки длины является такой же, как ошибка

плюс любая ошибка, вводимая ошибочным значением емкости на один километр.Overall, for cables up to 9 km long

and

can be estimated with an accuracy of approximately 2-6%. Cables up to 11 km long were measured with errors of approximately 15-20%. The accuracy of the length estimate is the same as the error.

plus any error introduced by one kilometer erroneous capacity value.

обычно имеет значительно более низкую точность, часто порядка 30%.Rating

usually has significantly lower accuracy, often of the order of 30%.

Estimation of the type of load impedance

It was assumed above that the line at the end is open or that the load impedance is such that the conditions of the line are essentially equivalent to an open line. When the end of the line is not open, but instead, as shown in FIG. 5, a load impedance 62 is connected, the type of load impedance can be estimated.

The input impedance of the transmission line, one end of which is loaded on the impedance Z _T , can be expressed as follows

Neglecting G (G = 0), this expression can be expanded in a series in the variable jω:

The series is real (converges) for | γd | <π.

тогда как нагрузка в виде параллельного соединения R и С может аппроксимироваться как резистивная нагрузка

The load impedance Z _T can have both a resistive and a reactive part, that is, Z _T = R _T + jX _T. For the most common R and C values and the frequencies used for the present invention, the load in the form of a series connection of R and C can be approximated as a capacitive load

while the load in the form of a parallel connection of R and C can be approximated as a resistive load

Comparison of input impedance series expansion for a short-circuited transmission line (Z _T = 0)

series expansion for open transmission line (Z _T = 0)

или ω в зависимости от того, является ли линия разомкнутой/емкостно нагруженной или короткозамкнутой/ резистивно нагруженной.shows that the imaginary part of the input impedance changes approximately as

or ω, depending on whether the line is open / capacitively loaded or short-circuited / resistively loaded.

Therefore, the imaginary part as a function of frequency can be used to distinguish between an open / capacitively loaded line or a short-circuited / resistively loaded line. The absolute value of the imaginary part at a very low frequency (for example, 125 Hz) is compared with the absolute value of the imaginary part at a frequency of about 1 to 2 kHz. If the absolute value of the imaginary part at a low frequency is greater than the absolute value at a higher frequency, then the line is most likely open at the end or has a capacitive load. If the absolute value of the imaginary part at a low frequency is lower, then the line is most likely short-circuited or has a resistive load.

Description of a flowchart of a method

,

, и

линии вычисляют на основе коэффициентов а_-1-а₂. На этапе 207 длину линии вычисляют на основе емкости

линии.The method described above is shown in the flowchart of the method (FIG. 6). The method begins at step 200, in which the signal MS _in (t) is transmitted to the transmission line. At step 201, the resulting signal MS _out (t) is received. At step 202, the transmitted and received signals are converted into the frequency domain by means of a fast Fourier transform. Based on the signals thus transformed, in step 203, the transfer function of the reflected signal H _echo (f) is calculated. At step 204, the input impedance of the line Z _in (f) and the calibration parameters Z _ho (f), Z _hyb (f), and H _∞ (f) are calculated based on H _echo (f). At step 205, the coefficients a ₋₁ – a ₂ are found that minimize the quadratic error function ε ² by solving the corresponding ordinary equations. At step 206, the constants

,

, and

the lines are calculated based on the coefficients a _-1 -a ₂ . At step 207, the line length is calculated based on the capacitance

lines.

At step 220, the imaginary part Z _{in is} calculated for a very low frequency (f1, for example, 125 Hz) and high frequency (f2, for example, 1.5 kHz). At step 221, it is checked whether the absolute value of the imaginary part Z _in (f) for the lowest frequency f1 is greater than the absolute value of the imaginary part Z _in (f) for the high frequency f2. If so, then the line at the end is most likely open or has a capacitive load (step 222). If not, the line at the end is most likely short-circuited or has a resistive load (step 223).

Devices and circuits for performing the method

The method according to the invention can be carried out on the installation shown in figure 1. Transmission and reception of signals along transmission lines is performed using linear panels 100, while subsequent processing of signals is performed on a separate automated workstation 120. Line panels can have a standard function of transmitting a signal along a transmission line and receiving a resulting signal, which can be used in the present invention. A digital representation of the signal to be transmitted will be uploaded from the workstation 120 to a line panel. On command from the automated workstation to the line panel, the signal is transmitted to transmission line 60 and a resulting signal is received, which is then transmitted back to the automated workstation. Then, the automated workstation is evaluated according to the method.

Automated workstation 120 can manage line tests according to the method for several thousand lines. Tests can be planned and performed on a planned basis to control any changes over time in the properties of the lines. Comparison with the previous results for a particular line may indicate that an error has occurred, that an error may occur in the near future, or that some property of the line is gradually deteriorating. Such an indication can be detected automatically and trigger an alarm.

Naturally, the functions of an automated workstation can be distributed across multiple computers or reside on the same computer along with other functionalities. An automated workstation or computers may also be located in a different place than the installation 130 of the central node, and communicate through a network.

In a specific implementation, such other functionality is the general line panel management for DSL. In another implementation, such other functionality relates to another transmission line management, such that all or most of the transmission line management functionality is performed in one place. Such an automated workstation may be called the "Copper Plant Manager".

Alternatively, the method according to the invention can be implemented entirely in a specialized test device or in a linear panel (which may or may not be a POTS linear panel). It is also possible that part of the method is performed by a linear panel (in addition to sending and receiving signals), and the rest is performed by the processor 110 of the communication station to which the linear panel belongs. In an additional alternative, the method is carried out partially in a linear panel, partially in a processor of a communication station, and partially in a separate automated workstation.

Claims (18)

1. A method for determining an estimate of at least one transmission property of a transmission line of a communication system, the method comprising:
at least one signal is transmitted to the line, at least one resulting signal is received from the line, an estimate is determined from the relationship between the transmitted signal or signals and the received resultant signal or signals, characterized in that the transmitted signal or signals together comprise, at least two frequencies for which the absolute value of the line propagation constant multiplied by the line length is less than π. 2. The method according to claim 1, in which at least one transmission property to be determined is one or more of a line length, line resistance, line capacitance, line inductance or type of line load impedance. 3. The method according to claim 1 or 2, in which the transmitted signal or signals together contain at least three frequencies for which the absolute value of the line propagation constant multiplied by the line length is less than π. 4. The method according to claim 1 or 2, in which the second estimate is performed using only those frequencies of the first estimate for which, according to the first estimate, the absolute value of the line propagation constant times the line length is less than π. 5. The method according to claim 4, in which subsequent additional estimates are performed in the same way, until the absolute value of the line propagation constant multiplied by the line length remains less than π for the frequencies used in accordance with the estimate based on these frequencies. 6. The method according to claim 1 or 2, in which the frequencies used for the second estimate are selected depending on the line length or capacitance determined from the first estimate. 7. The method according to claim 1 or 2, wherein said at least two frequencies are within the frequency range of a conventional telephone network (POTS). 8. The method according to claim 3, wherein said at least three frequencies are within the frequency range of a conventional telephone network (POTS). 9. The method according to claim 1 or 2, wherein said signal or signals are transmitted and received by a transceiver for normal operation of the transmission line. 10. The method of claim 9, wherein said transceiver is a POTS transceiver. 11. The method according to claim 9, in which the ratio between the transmitted and received resulting signals is adjusted to compensate for the effect of the transceiver on the signals. 12. The method according to claim 1 or 2, wherein the method is used to evaluate at least one transmission property of a transmission line longer than about 6.4 km. 13. A device for performing the method according to any one of claims 1 to 12, comprising means for transmitting a signal and receiving a resultant signal from a transmission line of a communication system and means for determining an estimate from a relationship between a transmitted signal and a received resultant signal. 14. The device according to item 13, and the device is a linear panel of a communication node. 15. A device for performing the method according to any one of claims 1 to 12 in cooperation with at least one linear panel of a communication node, the linear panel comprising means for transmitting a signal and receiving a resulting signal from a transmission line of a communication system, and the device comprises means to determine the estimate from the ratio between the transmitted and received resulting signal. 16. The device according to clause 15, the device is intended to perform the method according to any one of claims 1 to 12 in conjunction with at least one linear panel and at least one other part of the communication node. 17. The communication node for performing the method according to any one of claims 1 to 12, comprising means for transmitting a signal and receiving a resultant signal from a transmission line of a communication system and means for determining an estimate from a relation between a transmitted signal and a received resultant signal. 18. A computer-readable medium containing a software product stored thereon for implementing, when executed, a method according to any one of claims 1-12 on a device according to any one of claims 13-17.

Images