Fault detection

Abstract

The invention relates to fault detection. A method of detecting a fault along a cable or transmission line is provided. In the method, the following steps are performed: obtaining an echo response of the cable or transmission line using a time domain reflectometer (TDR), identifying a plurality of reflections in the echo response, generating a representation of the plurality of reflections, normalizing the representation or one of thresholds, and determining a fault condition based on the normalized representation. By doing so, faults in cables of different lengths can be accurately determined.

Description

Technical Field

[0001] The present invention relates to fault detection and, more particularly, to a method of detecting faults along a cable or transmission line. Background Art

[0002] Cables or transmission lines, such as Ethernet cables, can experience faults. An open circuit fault can occur when one or more of the wires of a cable or transmission line are interrupted. A short circuit fault can occur when one wire of a cable or transmission line comes into contact with another wire. Other softer faults can occur when aging, poor cable or transmission line construction, mechanical interaction with external objects, humidity, etc. create impedance discontinuities that affect signal integrity on the cable or transmission line. It is desirable to detect the occurrence of a fault and its location along the cable or transmission line, allowing the fault to be repaired soon after it is detected and reducing downtime for communications sent over the cable or transmission line. Summary of the invention

[0003] A method of detecting faults along a cable or transmission line is provided. In the method, the steps of obtaining an echo response of the cable or transmission line using a time domain reflectometer (TDR), identifying a plurality of reflections in the echo response, generating a representation of the plurality of reflections, normalizing the representation or one of a threshold, and determining a fault condition based on the normalized representation are performed. By doing so, faults in cables of varying lengths can be accurately determined.

[0004] According to a first aspect of the present disclosure, a method for detecting faults along a cable or transmission line is provided, the method comprising: obtaining an echo response of the cable or transmission line using a time domain reflectometer TDR; identifying multiple reflections in the echo response; generating a representation of the multiple reflections; obtaining a normalized representation of the representation of the multiple reflections or obtaining a normalized representation of a fault threshold; and determining a fault condition based on the normalized representation and the other of a first indication or a fault threshold.

[0005] According to a second aspect of the present disclosure, a method for detecting faults along a cable or transmission line is provided, the method comprising: obtaining an echo response of the cable or transmission line using a time domain reflectometer TDR; identifying multiple reflections in the echo response; generating representations of the multiple reflections; obtaining a normalized representation of the representations of the multiple reflections; and determining a fault condition based on the normalization and a fault threshold.

[0006] According to a third aspect of the present disclosure, a method for detecting faults along a cable or transmission line is provided, the method comprising: obtaining an echo response of the cable or transmission line using a time domain reflectometer TDR; identifying multiple reflections in the echo response; generating a representation of the multiple reflections; obtaining a normalized representation of a fault threshold; and determining a fault condition based on the representation of the multiple reflections and the normalized representation of the fault threshold. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Aspects of the present disclosure will now be described, by way of example only, with reference to the accompanying drawings, wherein like reference numerals represent like parts, and wherein:

[0008] Figure 1 A schematic diagram of the time domain reflectometry system is shown;

[0009] Figure 2 A plot of the echo response of a number of cables of varying lengths is shown;

[0010] Figure 3a A flow chart showing a method of detecting a cable or transmission line fault;

[0011] Figure 3b A flow chart showing a method for detecting faults along a cable or transmission line, including reflection verification;

[0012] Figure 4 Normalized echo response plots for several cables are shown;

[0013] Figure 5 The reflection of the echo response and the undershoot of the reflection are shown;

[0014] Figure 6 An exponential graph showing cable attenuation;

[0015] Figure 7a shows the impulse response provided by the time domain reflectometer with no cables connected;

[0016] Figure 7b shows the impulse response provided by a time domain reflectometer connected to an unterminated cable;

[0017] Figure 8 A flow chart showing a method of obtaining a cable echo response;

[0018] Fig. 9 shows another schematic diagram of a time domain reflectometer system;

[0019] Fig.10 Shows the use Fig. 9 A flowchart of a method for detecting a cable or transmission line fault in a system;

[0020] Fig.11 Shows Fig. 9 The echo response of the system's cables;

[0021] Fig. 12A The raw echo response of a fault occurring at 0 m of the reflectometer sampled at sampling frequency Fo is shown;

[0022] Fig. 12B yes Fig. 12A A magnified version of the reflection shown in ;

[0023] Fig. 12C The polynomial approximation for sampling reflections at sampling frequencies less than Fo is shown, and the reduced error in fault location is shown;

[0024] Fig.13 A heat map showing faults at different locations along a cable or transmission line. DETAILED DESCRIPTION

[0025] Time domain reflectometry can be used to measure reflections along a cable. A time domain reflectometer transmits a first signal down the cable and receives an echo response showing multiple reflections. In theory, in a perfect cable that is properly terminated, there would be no reflections, indicating that there are no faults. Reflections will appear if there is an open or short fault, or an impedance mismatch caused by a softer fault.

[0026] Some cables or transmission lines allow communication over long distances. For example, the 10BASE-T1L standard allows communication over cables up to 1 km long. After a fault is detected, the cable can be replaced. However, replacing a long cable may not be economical, so it is valuable to know the location of the fault within the cable, allowing only a small section of the cable to be removed and replaced.

[0027] There can be large variations between different cables, even those intended for the same standard. This is particularly problematic in long cables, where even small changes in cable performance, such as changes in attenuation per unit length of cable, can result in large variations in the operation of the cable.

[0028] In the case of time domain reflectometry, these changes in cable characteristics can result in large variations in the echo response. This results in fault detection accuracy that varies depending on the cable used.

[0029] The inventors have realised that by normalising the echo response or a threshold to which the echo response is compared, the accuracy of fault detection can be increased. This enables the system to detect faults in a variety of cables and accurately determine the size or type of the fault as well as the location of the fault.

[0030] Figure 1A schematic diagram of a time domain reflectometer system 300 is shown. System 300 includes a time domain reflectometer 302. Time domain reflectometer 302 can be implemented as part of a larger system. Time domain reflectometer 302 is coupled to a cable or transmission line 304. The time domain reflectometer is configured to output a signal to the cable or transmission line 304 and receive an echo response including reflections. Time domain reflectometer 302 is coupled to a control system 306. The time domain reflectometer outputs the received echo response to the control system 306. Time domain reflectometer 302 and control system 306 can be part of a system 308 that combines the capabilities of both systems.

[0031] Figure 2 The echo response is shown for multiple cables that have faults at different locations along cable 304. The X-axis shows the time at which the reflection was detected, with longer times representing greater distances along cable 304 from time domain reflectometer 302. The Y-axis shows the impulse response or amplitude of the reflection.

[0032] The first reflection 102 is detected at a first distance from the time domain reflectometer 302. The first reflection 102 represents the reflection provided by the time domain reflectometer 302 when no cables or transmission lines are connected to the time domain reflectometer 302.

[0033] Figure 2 304. The remaining reflections shown in 304 represent reflections that occur when a cable 304 having a fault is connected to the time domain reflectometer 302. The location of each fault is located at a different distance along the corresponding cable 304. A second reflection 104 is detected at a second distance along the cable 304. A third reflection 106 is detected at a third distance along the cable 304. A fourth reflection 108 is detected at a fourth distance along the cable 304. A fifth reflection 110 is detected at a fifth distance along the cable 304. A sixth reflection is located at a sixth distance along the cable 304. A seventh reflection 114 is located at a seventh distance along the cable 304.

[0034] The attenuation of the cable causes each reflection to have a different amplitude and shape. The greater the distance along the cable from the time domain reflectometer, the lower the peak impulse response or peak amplitude of the reflection. Other factors may also affect the amplitude of the reflection, such as the severity of the fault, the attenuation per unit length of cable, the impedance of the cable, and the length of the cable. Therefore, the same fault may look different depending on the length of the cable, the location of the fault, and the attenuation per unit length of cable. Figure 2 The reflections 104-114 shown in FIG. 1 may represent the same fault at different locations along the cable, with the reflections having different amplitudes due to cable attenuation.

[0035] The reflection may not represent a fault event. For example, the reflection may occur due to noise. Alternatively, the communication system may tolerate a fault that is not severe and therefore not defined as a fault. Therefore, the reflection may be compared to the fault threshold 116. In the case where the peak amplitude of the reflection is greater than the threshold 116, it may be determined that the reflection represents a fault. For example, the amplitudes of the second reflection 104 and the third reflection 106 at their peaks are greater than the fault threshold 116. Therefore, the second reflection 104 and the third reflection 106 may be determined to represent a fault condition. However, the amplitudes of the fourth reflection 108, the fifth reflection 110, the sixth reflection 112, and the seventh reflection 114 at their peaks are less than the threshold. Therefore, even if these reflections are caused by a fault, it may be determined that the reflections 106-114 do not represent a fault condition.

[0036] Threshold 116 may only accurately determine if a fault exists in short cables, where insertion loss is limited and cable attenuation has minimal effect on reflection amplitude. However, in some systems, such as 10BASE-T1L Ethernet, the cables may be longer, resulting in faults that provide reflections with minimal impulse response. Furthermore, not every reflection corresponds to a fault. Therefore, an alternative method is needed to determine if a reflection corresponds to a fault.

[0037] Figure 3a A flow chart showing a method of detecting a cable or transmission line fault.

[0038] In step S402, a time domain reflectometer is used to obtain an echo response of a cable or a transmission line.

[0039] In step S404, a plurality of reflections are identified in the echo response.

[0040] At step S406, a representation of the plurality of reflections is generated. The representation of the plurality of reflections may include a first waveform including the plurality of reflections, or it may include a list of the plurality of reflections and the corresponding sample numbers of the reflections, for example, the sample number of the peak value, or the peak amplitude of the reflections.

[0041] In step S408, a normalized representation is obtained. Obtaining the normalized representation may include normalizing the reflections of the echo response or a fault threshold to which the reflections are compared. Thus, a normalized representation of the representations of the plurality of reflections is obtained, or a normalized representation of the fault threshold is obtained.

[0042] At step S410, a fault condition is determined based on a normalized representation of the plurality of reflections and a fault threshold. Alternatively, a fault condition is determined based on a representation of the plurality of reflections and a normalized representation of the fault threshold. For example, a fault is determined by comparing the fault threshold and a peak value of the reflection. Although the reflections are described herein as positive reflections, the reflections may have a negative polarity. Therefore, comparing the peak value of the reflection to the threshold may include comparing the absolute maximum value of the reflection to the threshold, thereby detecting a fault at a sample number where the absolute highest value of the reflection is above the threshold.

[0043] Figure 4 The echo response of a cable with a fault at different locations along the cable is shown. The echo response can be a sampled discrete signal. The X-axis shows the samples at which the reflection was detected, with higher sample numbers indicating that the reflection occurred farther along the cable from the time domain reflectometer. The Y-axis shows the impulse response magnitude or reflection amplitude at each data point. While the echo response is sampled, it is obvious that the sample number can also be viewed as the time at which the reflection was detected.

[0044] Figure 4 The echo response of is normalized with respect to the number of samples, indicating Figure 3a This can also be considered the same as normalizing the echo response according to time or distance from a time domain reflectometer. Therefore, the reflections 202-214 representing the fault all have a peak amplitude greater than the threshold 216. Although Figure 4 Normalization of the echo response is shown, but it is also possible to normalize to a threshold 216 and compare the reflection to the normalized threshold.

[0045] By normalizing to one of the representations of multiple reflections or to a fault threshold, fault determination is independent of the attenuation of the cable or the length of the cable. This enables accurate fault detection even with long cables.

[0046] Reflections generated by impedance mismatches and indicating faults usually consist of a main reflection peak and an undershoot or overshoot. Figure 4 The reflection 204 is followed by an undershoot 218. The undershoot 218 does not represent a valid fault and can be considered an invalid reflection or a subsequent reflection. Therefore, it is beneficial to remove these invalid reflections from the fault analysis.

[0047] Figure 3b Shows Figure 3a A modified version of the method further comprises step S412. Step S412 comprises verifying the plurality of reflections identified in step S404. Then, the representation of the plurality of reflections generated in step S406 is based on the verified reflections, such that it is a representation of the plurality of verified reflections.

[0048] A reflection may be determined to be invalid for a number of reasons. Figure 4A second reflection 204 is shown, followed by a reflection 218 . Figure 5 Shows Figure 4 Another view of , only the second reflection 204 and the subsequent reflection 218 are shown.

[0049] A reflection may be invalid if it is adjacent to a previous reflection and has the opposite polarity to the previous reflection. Figure 5 As shown, the polarity of the following reflection 218 (negative) is opposite to the polarity of the second reflection 204 (positive). In addition, if the subsequent reflection 218 is adjacent to the second reflection 204, the subsequent reflection may be invalid. Adjacent may mean that the final sample of the first reflection 204 is adjacent to the first sample of the next reflection 218. Adjacent may mean that the final sample of the first reflection and the first sample of the subsequent reflection are within a small number of samples of each other. This may be a threshold, and if the first sample of the following reflection is within the following threshold number of samples, it is considered an overshoot or undershoot, rather than an independent fault. The following threshold number of samples can be determined by experimentally testing a cable containing a single fault to determine the number of samples after the final sample of the reflection for the following reflection to occur. For example, after the second reflection 204, the echo response of the cable or transmission line 304 may not return to a stable state, but may overshoot or undershoot.

[0050] A sample where the reflection crosses the zero axis or has a zero impulse response can define whether the sample is part of a reflection or part of an overshoot or undershoot. The final or ending sample of a reflection and the first or starting sample of an undershoot or overshoot can be defined as the samples on either side of the sample where the zero crossing occurs. Alternatively, the start and end of a can be defined as samples where the absolute value of the slope or gradient is less than a fraction of the average slope of the reflection. For example, a sample or samples may have a slope that is 1 / 4 the average slope of the reflection. If the following reflection is of opposite polarity and is adjacent to the previous reflection, it can be assumed to be caused by an undershoot or overshoot and therefore does not represent a fault.

[0051] If the width 402 of the reflection 218 is less than the sample width threshold, the reflection 218 may also be considered invalid. The sample width threshold may be defined by calibration based on the type or severity of faults that the communication system can tolerate. The sample width threshold may be defined as half the minimum intolerable fault sample width reflection width.

[0052] If a reflection 218 is adjacent to a previous reflection 204 and its average starting difference or gradient is less than the ending difference or gradient of the previous reflection, the reflection 218 may be considered invalid. The starting differential of a reflection may be considered to be the reflection gradient between the beginning of the reflection and the peak amplitude of the reflection. The ending differential of a reflection may be the reflection gradient between the peak amplitude of the reflection and the end of the reflection. For example, the ending differential of the second reflection 204 is the differential over the sample or time period 404. The starting differential of the reflection 218 is the differential over the sample or time period 406. The peak amplitude of the reflection is the maximum absolute value of the reflection. For example, the peak amplitude of the second reflection 204 occurs at sample number or time 408. Since the starting and ending differentials of adjacent reflections may have opposite polarities, a comparison may be made between the absolute values ​​of the starting and ending reflections.

[0053] By considering these factors, it is possible to determine if the subsequent reflection represents a second fault or if it is simply an overshoot or undershoot associated with the first fault.

[0054] like Figure 3a As shown in step S408, a normalized representation of multiple reflections is obtained, or a normalized representation of a fault threshold is obtained. Figure 6 It is shown that the reflection obtained by the time domain reflectometry echo response has an approximately exponential relationship with the time or number of samples to receive or detect the reflection 602. This is due to the fact that the insertion loss of the cable increases exponentially with the length of the cable. This can be seen from the following equation:

[0055] The length is L 0 The attenuation of the cable segment is set to A 0 , the attenuation A of a cable with a length of L at a given frequency f is:

[0056]

[0057] The length of a cable can also be expressed in terms of the nominal velocity of propagation (NVP) of the cable, where t is the round-trip time required for a signal to travel along the length of the cable, reflect, and return to its starting point. L can be thought of as:

[0058]

[0059] The received echo response may contain multiple reflections. For discrete signals, the sampling period T s The received echo response is sampled at a sampling rate of n, and the peak amplitude or impulse response of the reflection may appear at sample number n. The round-trip time t from the reflection peak can be rewritten as:

[0060]

[0061] Parameter n 0is the offset of the reflectometer, which may be caused by the front end of the reflectometer and can be inferred from the calibration of the reflectometer, and Fs is the sampling frequency of the discrete signal. Figure 7a An example of this parameter is shown, where Figure 7a The response shown is the echo response obtained by the time domain reflectometer 302 when there is no cable 304 connected and the reflectometer is not terminated or is in an open circuit state. Therefore, the parameter n 0 is the number of samples from the time the TDR transmits the signal to the first reflection 702 when no cable is connected to the TDR 302. This offset calibration takes into account any delays within the TDR 302 and any connections between the TDR 302 and the cable 304. Parameter b is the amplitude of the first reflection 702.

[0062] Therefore, for a given frequency, the total attenuation of a cable of length L is an exponential function of its length and can be expressed as:

[0063]

[0064] where v 0 is the propagation speed of the signal, and t is the round-trip time of the signal from one end of the cable to the other. Therefore, A can be expressed as:

[0065]

[0066] where β is the insertion loss per sample.

[0067] Therefore, each sample n of the echo response can be obtained by multiplying its amplitude or the value of that sample by a factor Equation 6 shows the sample-dependent normalization that can be applied to sample n of the echo response to normalize its amplitude or impulse response. This factor can be easily generated by cumulative products, so the processing power required is low.

[0068]

[0069] Therefore, the normalized representation can be obtained by normalizing the reflection representation to the insertion loss of the cable. Alternatively, the same effect can be achieved by normalizing the threshold value to the insertion loss of the cable using the inverse function of equation six.

[0070] The normalization factor β can be determined based on the reflections received by the time domain reflectometer when the cable is not terminated. β is the cable insertion loss for each sample. β can be obtained by performing a calibration, connecting a given type of cable, leaving the other end open or unconnected or shorted, and obtaining the TDR response of the cable. Beta is calculated as the ratio of the peak value corresponding to the open (or short) condition at the end of the cable to the peak value corresponding to the open condition when the TDR device is not connected to the cable, and then raising this ratio to the inverse of the negative difference between its sample locations.

[0071] Figure 7b The echo response of a cable with an open connection at the end of the cable is shown. Since the cable is not terminated, there is reflection 704. m is the peak value or amplitude of the reflected echo response. M is the sample number where the peak occurs. Amplitude P m It can be expressed as the normalized version of Equation 6:

[0072]

[0073] Therefore, without knowing the cable length, the normalization factor β can be determined according to Equation 11:

[0074]

[0075] Obtaining the normalized representation includes normalizing the reflection using a single point normalization centered about a frequency or an average frequency centered about the power of the transmitted signal.The normalization of Equation 6 may be applied to the received echo response, or its inverse may be applied to a fault threshold to which the echo response is compared.

[0076] In block S410, the reflection is compared to a fault threshold. If the peak value of the reflection (i.e., the maximum amplitude or impulse response of the reflection) is above the fault threshold, then the reflection is considered to represent a fault. Figure 2 The reflections shown in are all positive reflections, but a negative reflection may be indicative of a fault. Depending on the type of fault, the reflection may have positive or negative polarity, so the absolute value of the peak amplitude can be compared to a fault threshold, such as the absolute value of the minimum or maximum amplitude of the reflection.

[0077] The fault threshold may be selected based on the maximum reflection coefficient, maximum impedance mismatch, or minimum return loss that the cable or transmission line can tolerate. The value that the cable or transmission line can tolerate may be the value that the protocol or communication standard being used is designed to accept.

[0078] For example, the 10BASE-T1L standard states that the return loss of a given cable must be above 13dB. Therefore, a threshold corresponding to a 13dB return loss can be defined. The reflection coefficient r corresponding to a 13dB return loss can be expressed as:

[0079]

[0080] Therefore, the fault threshold should be defined as

[0081] R=th / b

[0082] th=r*b=0.22387*b

[0083] where b is Figure 7b The reflection amplitude shown, for a given reflectometer, can be compared with n 0 Calibrate once at the same time.

[0084] Likewise, if the fault to be detected is one corresponding to an open or short circuit condition, this corresponds to a reflection coefficient r=1, so the threshold should be close to b, leaving some margin for error.

[0085] The above example shows that the choice of threshold is well within the limit allowed by the standard. However, one may want to provide some margin for detecting faults before they reach the limit.

[0086] A similar analysis can be performed if the threshold is based on impedance mismatch, e.g., instead of return loss, the limit is given in terms of the tolerance of the nominal impedance of the transmission line. The reflection coefficient can be calculated as:

[0087]

[0088] Zo=Line characteristic impedance

[0089] Z: minimum or maximum allowed impedance.

[0090] exist Figure 3a and 3b In step S402, the echo response of the cable 304 is obtained using the time domain reflectometer 302 and provided to the control system 306. Figure 8 The time domain reflectometer 302 can obtain the echo response.

[0091] The time domain reflectometer 302 is coupled to a cable or transmission line 304. In block S802, the time domain reflectometer 302 transmits a frequency shaped pseudo-random transmission symbol sequence into the cable 304.

[0092] In block S804 , the time domain reflectometer 302 receives a signal from a cable or transmission line.

[0093] In block S806, the time domain reflectometer 302 determines or obtains the echo response of the cable by cross-correlating the received signal with the frequency-shaped pseudo-random sequence of the transmission symbol. The echo response is constructed point by point by performing the cross-correlation each time. The sampling start of the received signal is delayed by a time Ts, and a new echo response point is generated each time, thereby generating an echo response with a sampling period Ts.

[0094] During normal operation of the cable, a communication signal may be transmitted along the cable. The properties of the communication signal depend on the specific standard followed by the communication protocol. The transmission frequency shaping pseudo-random sequence of transmission symbols used to obtain the echo response can have the same power spectral intensity or a scaled version thereof as the power spectral density of the communication signal used for the cable or transmission line. For example, where the cable or transmission line is intended for use with the 10BASE-T1L Ethernet standard, the power spectral density of the frequency shaping pseudo-random sequence of transmission symbols can match the power spectral intensity of the 10BASE-T1L signal described as PMA test mode 3. This ensures that any faults found in the cable or transmission line are faults that will affect the intended communication signal. Other faults not represented in the echo response may be considered unrelated to the use of the cable or transmission line for its intended communications.

[0095] Figure 1 A system is shown in which a signal time domain reflectometer 302 is used to obtain the TDR or echo response of a cable 304. Fig. 9 As shown, the system 800 may include a second time domain reflectometer 902. The second time domain reflectometer 902 is coupled to a different portion of the cable or transmission line 304 than the first time domain reflectometer 302. For example, the first time domain reflectometer 302 is coupled to a first end of the cable 304 and the second time domain reflectometer 902 is coupled to a second end of the cable 304. The second time domain reflectometer 902 may be used in conjunction with the first time domain reflectometer 302 to provide improved fault detection accuracy.

[0096] Fig.10 A flow chart is shown of a method for detecting faults on a cable or transmission line using two received echo responses.

[0097] In step S402, a first echo response of the cable or transmission line 304 is obtained from a first end of the cable 304 using a time domain reflectometer. The first time domain reflectometer 302 may be used to obtain the first echo response.

[0098] In step S404, a plurality of reflections are identified in the first echo response.

[0099] In step S1002, a second echo response of the cable or transmission line 304 is obtained from a second end of the cable 304 using a time domain reflectometer. The second time domain reflectometer 902 may be used to obtain the second echo response.

[0100] In step S1004, a second plurality of reflections is identified in the second echo response.

[0101] At step S1012, the plurality of reflections identified in the first echo response are verified using the second plurality of reflections identified in the second echo response. The verification may include comparing the positions of the plurality of reflections in the first echo response with the positions of the plurality of reflections in the second echo response. For example, if the first echo response includes a first reflection and the second echo response includes a second reflection corresponding to the first reflection, the first reflection may be considered valid at least because two separate echo responses of the cable 304 indicate the reflection.

[0102] Fig.11 An example of a first echo response and a second echo response is shown, with the sample number on the X-axis and the impulse response on the Y-axis. The first echo response includes a first reflection 1102 and the second echo response includes a second reflection 1104. The first reflection 1102 and the second reflection 1104 occur at different sample numbers of the respective echo responses. This is because the echo responses are obtained from different ends of the same cable 304.

[0103] Fig. 9 A representation of a first fault 904 on the cable 304 is shown. Let D be the length of the cable or the number of samples required to represent the length of the cable. The first fault 904 is the distance or number of samples D1 from the location 302 (the first end of the cable) where the first echo response was collected. The first fault 904 is the distance or number of samples D-D1 from the location 302 (the second end of the cable) where the second echo response was collected.

[0104] Therefore, the same fault represented by the reflection appears at a different sample number in the first echo response compared to the second echo response. If the first echo response acquired from the first end of the cable shows a first reflection 1102 at a distance D1 from the first end of the cable 304, and the second echo response acquired from the other end of the cable shows a second reflection 1104 at a distance D-D1 from the other end of the cable 304, then the first reflection 1102 and the second reflection 1104 can be considered to represent the same reflection, and the first reflection 1102 is considered to be valid.

[0105] Furthermore, if the first reflection 1102 and the second reflection 1104 have the same polarity, they may be considered valid.

[0106] Therefore, in block S1012, the reflections in the first echo response are identified and verified by comparison with the reflections in the second echo response.

[0107] In step S406, a plurality of representations of verification reflections are generated.

[0108] In step S408, a normalized representation of the representations of the plurality of reflections is obtained, or a normalized representation of the fault threshold is obtained.

[0109] In step S410, a fault condition is determined based on a normalized representation of the plurality of reflections and a fault threshold. Alternatively, a fault condition is determined based on a representation of the plurality of reflections and a normalized representation of the fault threshold.

[0110] In step S1012, when the first reflection 1102 and the second reflection 1104 are determined to represent the same fault, these faults can be used to calibrate the cable and determine the attenuation of the cable.

[0111] The first reflection 1102 has a first amplitude or impulse response 1106, whose value is P 1 ,exist Fig. 9 It is observed in the TDR response obtained from the time domain reflectometer 302. When obtained from the time domain reflectometer 902, Fig.11 The second reflection 1104 in has a peak value P 2 The second amplitude or impulse response 1108 of the cable 304 is shown. The reflections are found at different distances from the respective time domain reflectometer or cable end from which the echo response is obtained. The first reflection 1102 is a distance D1 from the first end of the cable 304. The second reflection 1104 is a distance D-D1 from the second end of the cable 304. Essentially, the two different reflections represent the same fault measured from different points in the cable 304. Therefore, the amplitude difference between the two reflections is caused by the attenuation of the cable.

[0112] Between the first reflection 1102 and the second reflection 1104, the cable 304 attenuates the signal by the ratio of the first amplitude 1106 to the second amplitude. The attenuation corresponds to the attenuation over the length of the cable D-(2xD1). Therefore, if the distance is measured in samples, the attenuation or normalization factor β for each sample is given by:

[0113]

[0114] This in turn can be expressed as a function of the number of samples in which each peak occurs:

[0115]

[0116] This attenuation coefficient β can then be used for the normalization of the reflection or threshold described above. Thus, the attenuation or normalization coefficient β used to obtain the normalized representation can be determined based on the ratio of the maximum values ​​of the first reflection 1102 and the second reflection 1104.

[0117] Although the determination of the normalization coefficient is described here based on a fault in the cable 904, the determination can also be made in the absence of a fault. For example, the reflection caused by the equipment between the corresponding time domain reflectometer and the cable end to the cable joint can be used. In some cases, the characteristic impedance of the cable matches the output impedance of the time domain reflectometer, and the reflection caused by the joint may be too small to be determined in the noise of the cable, or the accuracy of the determination may be reduced due to a low signal-to-noise ratio. However, if the reflection is too small, it indicates that there is no fault in the cable.

[0118] The foregoing description considers the echo response of the system. Although the echo response is described as being obtained by a time domain reflectometer, it may be obtained by any device capable of transmitting a signal to a cable or transmission line and receiving an echo response from the cable or transmission line. For example, the system may not include a dedicated time domain reflectometer, but rather the echo response of the system may be obtained while the cable 304 is being used for communications. For example, device 302 may be a communications device capable of sending and receiving communications signals. Similarly, device 902 may be a communications device capable of sending and receiving communications signals. The echo response may be obtained using an active time domain reflectometer, where the echo response is obtained using the filter coefficients of an echo canceller of a device having such a filter during communications, as is the case with some Ethernet transceivers.

[0119] The representation of the plurality of reflections may include a first waveform including the plurality of reflections. Alternatively, it may include a list of the plurality of reflections including the sample number of the peak of each reflection. The list may also include other properties of the reflections, such as the width of each reflection; the maximum of each reflection; the location of the maximum of each reflection; the starting point of each reflection; the end point of each reflection; the average starting slope of each reflection; and the average ending slope of each reflection.

[0120] If it is determined that the reflection represents a fault, the location of the fault can be determined based on the number of samples where the reflection peak amplitude occurs and the sample frequency:

[0121]

[0122] The echo response may be a sampled echo response. Thus, the location of the fault may be the sample number where the reflection peak indicating the fault is located. When the echo response is sampled, the peak may not be accurately represented in the echo response.

[0123] Fig. 12AThe echo response of a fault occurring at 0 m of the reflectometer sampled at sampling frequency F0 is shown. As shown, the reflection 1202 is composed of multiple individual samples (represented by the circles in the figure). The line connecting these samples provides an approximation of the reflection. Based on the samples alone, the peak impulse response of the reflection is selected as the peak impulse response of one of the samples. However, as mentioned above, this may not be an accurate representation of the samples in the case of low sampling rates.

[0124] Polynomial fitting can be used to more accurately represent the location of the peak and the shape of the reflection. Fit a polynomial function to the reflection of the echo response and then determine the fault location based on the number of samples at the maximum value of the polynomial function or the number of interpolated samples.

[0125] Fig. 12B Shows Fig. 12A A peak sample may be sample 1204. Fig. 12C The same reflection sampled at a much lower sampling frequency is shown. It can be seen that there is a gap between the samples where no data was collected. Therefore, the true peak of the reflection may lie between the two samples.

[0126] Fig. 12C Also shown is a polynomial approximation 1206 applied to the reflections. Fig. 12A and 12B The approximation is then applied to the 12C data, which is Fig. 12A and 12B As shown in the figure, the maximum value 1208 of the polynomial function is different from the sample 1204 that was previously considered to represent the maximum value. This value is closer to Fig. 12B 1204. Thus, the polynomial approximation 1206 improves the accuracy of peak detection in the reflection. Although the polynomial fit described here is with respect to the maximum value of the polynomial function, it can also be with respect to the lowest value or minimum value of the polynomial function in the case where the reflection has a negative polarity. In other words, the number of samples of the reflection peak can be selected to be the number of samples of the interpolated number of samples when the absolute maximum of the polynomial function occurs.

[0127] Any detected fault can be represented using a heat map, showing the location of the fault. Fig.13 A heat map 1300 is shown. The heat map provides a graphical representation of the cable or transmission line 304. One or more faults detected by the control system 306 are represented on the cable or transmission line. In the event that no faults are detected, there may be no faults on the cable or transmission line 304. In the event that multiple faults are detected, each fault may be represented to show the severity of the fault. For example, a fault 1302 is represented by a dense dashed area, indicating a larger magnitude fault. A fault 1304 is represented by a different dashed area, indicating a relatively smaller magnitude fault.

[0128] The graphical representation can be presented on a digital or information display. The control system 306 provides instructions related to the location and severity of the fault to the display. Fig.13 The heat map shown is a black-and-white line drawing. However, the heat map can also be represented by colors. Thus, these faults can be represented by an increase in the color gradient near the fault and a decrease in the color gradient after the fault. The color gradient and the color itself can be used to represent the severity of the fault, such as how far the reflection peak is above the fault threshold. This allows the user to quickly determine the location and severity of the fault in the cable or transmission line.

[0129] Various modifications can be made to the above examples, whether by adding, deleting, or replacing features, to provide further examples, any and all of which should be included in the appended claims.

[0130] Numbering aspects:

[0131] As a non-limiting example, some aspects of the present disclosure are listed in the following numbered clauses.

[0132] 1. A method for detecting a fault along a cable or transmission line, the method comprising:

[0133] Obtaining an echo response of the cable or transmission line using a time domain reflectometer TDR;

[0134] Identifying a plurality of reflections in the echo response;

[0135] Generating a representation of the plurality of reflections;

[0136] Obtaining a normalized representation of the representation of the plurality of reflections or obtaining a normalized representation of a fault threshold;

[0137] Determining a fault condition based on the representation of the plurality of reflections or the normalized representation of the fault threshold and another.

[0138] 2. The method according to aspect 1, wherein obtaining the normalized representation of the representation of the plurality of reflections or the normalized representation of the fault threshold comprises obtaining the normalized representation of the representation of the plurality of reflections, and

[0139] wherein determining the fault condition comprises determining the fault condition based on the normalized representation and the fault threshold.

[0140] 3. The method according to aspect 1, wherein obtaining the normalized representation of the representation of the plurality of reflections or the normalized representation of the fault threshold comprises obtaining the normalized representation of the fault threshold, and

[0141] wherein determining the fault condition comprises determining the fault condition based on the representation of the plurality of reflections and the normalized representation of the fault threshold.

[0142] 4. A method for detecting faults along a cable according to any one of the preceding aspects, wherein the method further comprises:

[0143] The plurality of reflections are validated, wherein validating the plurality of reflections comprises determining whether a reflection of the plurality of reflections is invalid such that it does not represent a fault condition.

[0144] 5. The method according to aspect 4 includes determining a reflection among the multiple reflections as invalid if the reflection is adjacent to a previous reflection and has a polarity opposite to the previous reflection, wherein the reflection is adjacent to the previous reflection if a sample of the previous reflection is at a later time than a sample of the reflection.

[0145] 6. A method according to aspect 4 or aspect 5, comprising determining a reflection in the plurality of reflections as invalid if the width of the reflection is less than half the width of a reflection caused by a fault occurring at zero distance from the end of the cable or transmission line in the sample.

[0146] 7. A method according to any one of clauses 4 to 6, comprising determining a reflection in the plurality of reflections as invalid if a width of the reflection is greater than a sample width threshold.

[0147] 8. A method according to any one of clauses 4 to 7, comprising determining a reflection of the plurality of reflections as invalid if the reflection is adjacent to and opposite in polarity to a previous reflection and has an average starting differential that is smaller than an ending differential of the previous reflection.

[0148] 9. The method according to any one of aspects 4 to 8, wherein the method further comprises:

[0149] Invalid peaks are discarded from the obtained echo response and a representation of the multiple reflections is obtained based on any remaining reflections of the identified multiple reflections.

[0150] 10. The method according to any preceding aspect, wherein the representation of the plurality of reflections comprises:

[0151] a first waveform comprising the plurality of reflections; or

[0152] List of multiple reflections and the corresponding sample counts for multiple verified reflections.

[0153] 11. The method of clause 10, wherein the list of the plurality of reflections and the corresponding number of samples of the plurality of reflections further comprises characteristics of the verified reflections, wherein the characteristics comprise one or more of the following:

[0154] The width of each reflection;

[0155] The maximum value of each reflection;

[0156] The maximum position of each reflection;

[0157] The starting point of each reflection;

[0158] The end points of each reflection;

[0159] The average initial slope of each reflection; and

[0160] The average end slope of the individual reflections.

[0161] 12. A method according to any one of aspects 2 to 11, wherein determining a fault condition comprises determining whether a reflection in the normalized representation is above the threshold.

[0162] 13. A method according to any one of aspects 3 to 11, wherein determining a fault condition comprises determining whether a reflection in the representation of the plurality of reflections is above a normalized representation of the threshold.

[0163] 14. The method according to any of the preceding aspects, wherein the method further comprises:

[0164] Determine the location of the fault condition based on the number of samples of the reflection for which the fault condition is determined.

[0165] 15. The method according to any one of the preceding aspects, wherein the method further comprises:

[0166] A polynomial function is fitted to the reflection of the echo response and the position of the reflection is determined based on the maximum or minimum value of the polynomial function and the number of samples or interpolated samples at which the maximum or minimum value occurs.

[0167] 16. The method according to aspect 14 or aspect 15, wherein the method further comprises:

[0168] Get a heat map representation of the fault condition location.

[0169] 17. The method of clause 16, wherein the heat map representation further comprises one or more of the following:

[0170] Reflection coefficient for fault conditions;

[0171] Fault condition return loss; and

[0172] Impedance mismatch for fault conditions.

[0173] 18. The method of any preceding aspect, wherein obtaining a normalized representation comprises normalizing with respect to an insertion loss of the cable.

[0174] 19. The method of any preceding aspect, wherein obtaining the normalized representation comprises normalizing the reflectance using a single point normalization centered around a frequency or an average frequency centered around the power of the transmitted signal.

[0175] 20. A method according to any of the preceding aspects, wherein obtaining a normalized representation includes normalizing using sample-dependent normalization, such that the applied normalization is an exponential function of the number of samples (n) and depends on the amplitude and number of samples of the reflection corresponding to an open condition when no cable is connected, and the amplitude and number of samples of the reflectivity corresponding to an open or short-circuit condition at the end of the cable when the cable is connected.

[0176] 21. The method according to clause 20, wherein the normalization is a function where β is the cable insertion loss per sample, n is the number of samples corresponding to the normalized sample, and n 0 is the number of samples of the first reflection without the cable connected.

[0177] 22. A method according to any preceding aspect, wherein the echo response of the cable or transmission line is obtained by cross-correlation of a frequency shaped pseudo-random sequence of transmission symbols.

[0178] 23. The method according to any one of the preceding aspects, wherein the method further comprises:

[0179] A frequency shaped pseudo-random transmission symbol sequence is sent to obtain the echo response of the cable by cross-correlation.

[0180] 24. A method according to clause 23, wherein the transmitted frequency shaped pseudo-random transmission symbol sequence has a power spectral intensity that is the same as or a scaled version of the power spectral density of the communication signal.

[0181] 25. A method according to any preceding aspect, wherein the fault threshold is selected based on a maximum reflection coefficient, a maximum impedance mismatch or a minimum return loss that can be tolerated by the cable or transmission line.

[0182] 26. The method according to any one of the preceding aspects, wherein the method further comprises:

[0183] obtaining a second echo response of the cable or transmission line, wherein the second echo response is from a different location than the echo response;

[0184] identifying a second plurality of reflections in the second echo response;

[0185] A plurality of reflections in the echo response is verified, wherein verifying the plurality of reflections includes comparing positions of the plurality of peaks to positions of a second plurality of peaks.

[0186] 27. The method according to aspect 26, wherein the method further comprises:

[0187] identifying a first reflection of the plurality of reflections;

[0188] identifying a second reflection in the second plurality of reflections;

[0189] The first reflection and the second reflection are compared to determine whether the first reflection is a valid reflection.

[0190] 28. A method according to clause 27, wherein the first reflection is a valid reflection if the first reflection and the second reflection overlap in position and have the same polarity.

[0191] 29. A method according to clause 27 or 28, wherein a normalization coefficient used to obtain the normalized representation is determined according to a ratio of maximum values ​​of the first reflection and the second reflection.

[0192] 30. A method of detecting faults along a cable or transmission line, the method comprising:

[0193] Obtaining an echo response of the cable or transmission line using a time domain reflectometer TDR;

[0194] identifying a plurality of reflections in the echo response;

[0195] generating a representation of the plurality of reflections;

[0196] obtaining a normalized representation of the representations of the plurality of reflections; and

[0197] A fault condition is determined based on the normalization and a fault threshold.

[0198] 31. A method of detecting faults along a cable or transmission line, the method comprising:

[0199] Obtaining an echo response of the cable or transmission line using a time domain reflectometer TDR;

[0200] identifying a plurality of reflections in the echo response;

[0201] generating a representation of the plurality of reflections;

[0202] obtaining a normalized representation of the fault threshold; and

[0203] A fault condition is determined based on the representation of the plurality of reflections and a normalized representation of the fault threshold.

Claims

1. A method for detecting faults along a cable or transmission line, the method comprising: Obtaining an echo response of the cable or transmission line using a time domain reflectometer TDR; identifying a plurality of reflections in the echo response; generating a representation of the plurality of reflections; obtaining a normalized representation of the representations of the plurality of reflections or obtaining a normalized representation of a fault threshold; and A fault condition is determined based on either the representation of the plurality of reflections or the normalized representation of the fault threshold and the other.

2. The method of claim 1 , wherein obtaining the representation of the plurality of reflections or the normalized representation of the fault threshold comprises obtaining a normalized representation of the representation of the plurality of reflections, and Wherein determining a fault condition comprises determining a fault condition based on the normalized representation and the fault threshold.

3. The method of claim 1 , wherein obtaining the representation of the plurality of reflections or a normalized representation of the fault threshold comprises obtaining a normalized representation of the fault threshold, and Wherein determining a fault condition comprises determining a fault condition based on a representation of the plurality of reflections and a normalized representation of the fault threshold.

4. The method according to claim 1, wherein the method further comprises: The plurality of reflections are validated, wherein validating the plurality of reflections comprises determining whether a reflection of the plurality of reflections is invalid such that it does not represent a fault condition.

5. The method of claim 4 , comprising determining that a reflection of the plurality of reflections is invalid if the reflection is adjacent to a preceding reflection and has an opposite polarity to the preceding reflection, wherein the reflection is adjacent to the preceding reflection if a sample of the preceding reflection is at a later time than a sample of the reflection.

6. The method of claim 4, comprising determining that a reflection of the plurality of reflections is invalid if the width of the reflection is less than half the width of a reflection caused by a fault occurring at zero distance from the end of the cable or transmission line in the sample.

7. The method of claim 4, comprising determining that a reflection of the plurality of reflections is invalid if the reflection is adjacent to and opposite in polarity to the preceding reflection and has an average starting differential that is smaller than an ending differential of the preceding reflection.

8. The method according to claim 4, wherein the method further comprises: Invalid peaks are discarded from the obtained echo response and a representation of the multiple reflections is obtained based on any remaining reflections of the identified multiple reflections.

9. The method of claim 1 , wherein the representation of the plurality of reflections comprises: a first waveform comprising the plurality of reflections; or A list comprising the plurality of reflections and corresponding sample numbers of the plurality of reflections.

10. The method of claim 2, wherein determining a fault condition comprises determining whether a reflectance in the normalized representation is above the threshold.

11. The method of claim 3, wherein determining a fault condition comprises determining whether a reflection in the representation of the plurality of reflections is above a normalized representation of the threshold.

12. The method according to claim 1, wherein the method further comprises: The location of the fault condition is determined based on the number of samples of the reflection for which the fault condition is determined.

13. The method according to claim 1, wherein the method further comprises: A polynomial function is fitted to the reflection of the echo response, and the position of the reflection is determined based on a maximum value of the polynomial function and a sample number or an interpolated sample number at which the maximum value occurs.

14. The method of claim 1, wherein obtaining a normalized representation comprises normalizing with respect to an insertion loss of the cable.

15. The method of claim 1, wherein obtaining the normalized representation comprises normalizing the reflection using a single point normalization centered around a frequency or an average frequency centered around a power of a transmitted signal.

16. The method of claim 1, wherein the normalization is a function ,in β is the cable insertion loss per sample, n is the sample number corresponding to the sample being normalized, and n0 is the sample number of the first reflection without the cable connected.

17. The method according to claim 1, wherein the method further comprises: obtaining a second echo response of the cable or transmission line, wherein the second echo response is from a different location than the echo response; identifying a second plurality of reflections in the second echo response; A plurality of reflections in the echo response is verified, wherein verifying the plurality of reflections comprises comparing positions of a plurality of peaks to positions of the second plurality of peaks.

18. The method according to claim 17, wherein the method further comprises: identifying a first reflection of the plurality of reflections; identifying a second reflection in the second plurality of reflections; The first reflection and the second reflection are compared to determine whether the first reflection is a valid reflection.

19. A method of detecting faults along a cable or transmission line, the method comprising: Obtaining an echo response of the cable or transmission line using a time domain reflectometer TDR; identifying a plurality of reflections in the echo response; generating a representation of the plurality of reflections; obtaining a normalized representation of the representations of the plurality of reflections; and A fault condition is determined based on the normalization and a fault threshold.

20. A method of detecting faults along a cable or transmission line, the method comprising: Obtaining an echo response of the cable or transmission line using a time domain reflectometer TDR; identifying a plurality of reflections in the echo response; generating a representation of the plurality of reflections; Obtain a normalized representation of the fault threshold; and A fault condition is determined based on the representation of the plurality of reflections and a normalized representation of the fault threshold.

Images