• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    - A complete injection signal having a digital spectrum formed of components weighted by a set of coefficients, in a preliminary step is attenuated the coefficients of the components in said frequency band leading to obtaining an incomplete digital spectrum; - In a next step (21) is injected into the power line the analog signal formed from said incomplete digital spectrum; - In a next step (22) the echo corresponding to said injected signal is measured; In a next step (23), the digital spectrum (Ycancel) of said echo is calculated; In a next step (24), the digital spectrum (Yfull) of the echo of the complete analog signal is estimated if it had been injected, the kth component (Yfullk) of the estimated complete digital spectrum being equal to the kth component ( Ycancelk) of the digital spectrum of the received echo multiplied by the ratio (ζk) between the kth component (Xfullk) of the digital spectrum of the complete injection signal on the kth component (Xcancelk) of the incomplete digital spectrum; In a following step (25), the reflectectogram of said line is obtained from the estimated complete numerical spectrum (Yfull).
    公开号:FR3037147A1
    申请号:FR1555080
    申请日:2015-06-04
    公开日:2016-12-09
    发明作者:Laurent Sommervogel
    申请人:Win Ms；
    IPC主号:
    专利说明:

    [0001] The present invention relates to a method and a reflectometry device, in particular of the multicarrier type, for diagnosing cables in operation. BACKGROUND OF THE INVENTION The diagnosis of an electric line or cable by reflectometry is a well-known method of injecting a broadband signal into the line and detecting echoes to trace characteristic impedance variations. the line corresponding to singularities, including defects such as for example open circuits or short circuits. In its simplest version, the probe signal is a pulse whose duration depends on the bandwidth and the length of the cable to be diagnosed. Among the known reflectometry methods, the Multi Carrier Time Domain Reflectometry (MCTDR) method has the advantage of being able to precisely control the spectrum of the injected signal and thus be able to respond to electromagnetic compatibility constraints (EMC). ) imposed. Subsequently, the following conventions will be adopted: - * denotes the convolution product 25 - * denotes the cross correlation (inter correlation) - h denotes the impulse response of the cable - h denotes the estimated response of the cable: the reflectogram - x designates the signal injected - ydesigns the reflected signal 30 - The uppercase notation denotes the Fourier transforms from the lowercase ones - F {} denotes the Fourier transform operation - F-1 {} denotes the inverse Fourier transform operation The injected x and reflected signals are linked thereto by the following equation: y = x * h. A particular purpose of a reflectometry measurement is to measure the signal h, if possible with a maximum of precision. h can be considered as a parsimonious signal, that is to say that it contains only peaks at the location of singularities on the cable.
    [0002] In an ideal simplified case for a cable whose end is in an open circuit, the difference in position between the first two peaks of h makes it possible to go back to the knowledge of the length of the cable. When it is desired to monitor or diagnose a cable in operation, a non-invasive, non-intrusive measurement is required. Two conditions must then be respected: - It is necessary to find an adequate means of coupling, especially in terms of impedance; - Do not inject signals with frequency components overlapping frequency bands already used in the target application. In other words, it is necessary to avoid any interference between the signals related to the operation of the target application and the signals related to the reflectometry, this resulting in various constraints at the CEM level, spectral occupancy and noise robustness. The second condition makes it possible in particular to obtain the following double guarantee: the diagnostic signals do not interfere with those of the application in operation, in transmission; - The signals of the application do not degrade the quality of the diagnosis, in terms of susceptibility. By way of example: For the diagnosis of an antenna cable for the application of an FM radio reception, the 88-108 MHz frequency band should be avoided for the diagnostic signal; - For CAN bus diagnostics of a vehicle, the 0.1-2 MHz band should likewise be avoided. The MCTDR technique, in particular described in the patent application FR 2 931 323 A1 makes it possible to construct a probe signal corresponding to a given spectral mask. The described solution can handle any number of frequency bands with variable attenuation coefficient in each band, up to total extinction. However, in the solution described, it is necessary to face an important technical limitation. In fact, the more attenuated or more frequency bands are removed, the less the richness of the frequency content on the measurement of the reflected signal y is reduced, and the more the quality of the diagnosis is degraded.
    [0003] In particular, if we considered an ideal pulse, represented by a Dirac x = δ, to diagnose the cable, we would obtain h = y. However, the MCTDR is a pulse compression technique. That is, the energy is not concentrated on a very short time but spread over the duration of the signal. We can show mathematically in this case that: h = R. * h, with K. = x * x, this autocorrelation being called the pattern. The ideal case R. = Ô is only possible if x contains all the frequencies. Now, the more frequency components are removed, the less the pattern looks like an impulse.
    [0004] There are several techniques for diagnosing without interfering with a target application. STDDR (Spread Spectrum Time Domain Reflectometry) is a variant of the STDR (Sequence Time Domain Reflectometry). STDR is also a pulse compression technique. With this technique, instead of injecting an impulse, we inject into the cable a binary sequence of slots, composed of +1 and -1, making sure that the autocorrelation of the sequence is close to a pulse. .
    [0005] The STDR does not make it possible to overcome the constraints CEM but the SSTDR brings an answer to it. For this purpose, the STDR sequence is modulated by a sinusoidal carrier of frequency fo. So it's actually a modulated STDR sequence. To meet the operational constraints, we choose fo so as to move the modulated signal away from the forbidden bands.
    [0006] A reflectometry method described in document FR 2 931 323 A1 is an iterative method aiming to go back to h from the estimated impulse response h. It is a post-processing algorithm which can take as input a measurement of reflectometry made with a probe signal 5 of the MCTDR type. He inherits his safety from the CEM point of view. Multi Carrier Reflectometry (MCR) uses the same waveforms as the MCTDR, with a weighted sum of sinusoids. For this reason, it provides the same quality of response to EMC problems. The processing of the reflected signal is however very different, since it is an optimization algorithm (least squares type) directly applied in the frequency domain in order to adapt the coefficients (a; r) of a model of simplified parsimonious impulse response for the cable: 15 h (t) = IcrAt-2ir) This treatment has the advantage of being able to be performed for a limited number of excitation frequencies. The MCR technique is described in particular in the article "Multicarrier 20 Reflectometry", IEEE SENSORS JOURNAL, VOL. 6, No. 3, June 2006. One problem to be solved is in particular to allow the flexibility of the injection methods with a controlled spectral mask, such as the MCTDR in particular, while avoiding the denaturation problems of the pattern which introduce a bias. in the result of the diagnosis. The reflectometry methods previously described do not solve this problem or solve it insufficiently. The SSTDR reflectometry seems to answer this problem, but it nevertheless has several drawbacks: the modulation system around the frequency introduces a complexity of realization and additional costs. 3037147 5 - It is difficult to change fo "on the fly", which does not allow a dynamic reconfiguration of the diagnostic system. In some applications, it may even be difficult to find a frequency fo available, it should be noted that it is limited by the bandwidth of the cable and fo can not be chosen arbitrarily high. Likewise, the algorithm described in document FR 2 931 323 A1 as post-processing could be suitable. However, its robustness is caused when peaks are too close to each other in the impulse response h of the cable. On the other hand, it begins to converge erroneously when about more than one-sixth (1/6) of the useful band of the test signal has been suppressed.
    [0007] For its part, although the MCR method seems insensitive to the cancellation of its coefficients, it suffers in particular the following crippling defects: Its use is restricted to point-to-point cables.
    [0008] It results in a location inaccuracy which can reach 3% to 5% of the length of a cable. An object of the invention is in particular to overcome the aforementioned drawbacks. For this purpose, the subject of the invention is a method for diagnosing at least one electrical line by reflectometry measurements, said electrical line being in operation in a given frequency band, the diagnosis being made by analysis of the reflectogram of said line, said method comprising in particular the following steps: - A complete injection signal having a digital spectrum formed of components weighted by a set of coefficients, in a preliminary step the coefficients of the components included in said frequency band leading to a reduction are attenuated obtaining an incomplete digital spectrum. In a following step, the analog signal formed from said incomplete digital spectrum is injected into the electrical line (10). In a following step, the echo corresponding to said signal injected is recovered by sampling measurements of said echo. In a next step, the digital spectrum (Y, -cancel) of said echo is calculated. In a next step, the digital spectrum (Mali) of the echo of the complete analog signal is estimated if it had been injected, the kth component ullk / (r7, 1 of the estimated complete numerical spectrum being equal to the kth component (Y, -cancelk) of the digital spectrum of the received echo multiplied by the ratio (k) between the kth component (Xfuiik) of the digital spectrum of the complete injection signal on the kth component (Xcanceik) of the digital spectrum In a next step, the measurement of the reflectogram of said line is obtained from the estimated complete digital spectrum. The injected signal being calculated on N points, the echo is for example sampled on NP measurement points, the digital spectrum (Y, -cancel) of the echo being calculated on said NP sampled points, P being strictly greater than 2, P being able to be greater than or equal to 10. The ratios (fic) are for example calculated on N points, the Ncalculated ratios being taken again P times In the case where the signal injected not of the MCTDR type, said ratio is for example defined according to the following relation: 25 = iim Xfull Xc: anchel e () X cancel + £ Xfu11 and X cancel being respectively the digital spectrum of the complete injection signal and the digital spectrum of the injected signal. The subject of the invention is also a reflectometry device capable of being used for the diagnosis of a line operating in a given frequency band, said device comprising at least: ( Fuu). A parameter block for parameterizing the signal to be injected in said line, a complete injection signal having a digital spectrum formed of components weighted by a set of coefficients, the coefficients of the components included in said frequency band being attenuated; to obtain an incomplete digital spectrum; a block synthesizing in the digital domain the signal to be injected from said incomplete digital spectrum; a digital-to-analog converter converting the synthesized digital signal into an analog signal; coupling means to said line for injecting said analog signal into said line; an analog-digital converter converting the echo corresponding to the injected signal; Processing means: performing measurements sampled from said echo and storing the samples; calculates the digital spectrum (Y, -cancel) of said echo; estimating the digital spectrum (Mali) of the echo of the complete analog signal if it had been injected, the kth component (12j, 1 ullk / 25 estimated being equal to the kth component (Y of the -cancelk) spectrum digital number of the received echo multiplied by the ratio (k) between the kth component (Xfuiik) of the digital spectrum of the complete injection signal on the kth component (Xcanceik) of the incomplete digital spectrum; - obtaining the reflectogram of said line to from the estimated full digital spectrum (full).
    [0009] Other characteristics and advantages of the invention will become apparent with the aid of the description which follows, given with reference to the appended drawings which represent: FIG. 1, an example of a reflectometry device coupled to a cable, suitable for placing implement the method according to the invention; FIG. 2, the possible steps of a method according to the invention; of the complete digital spectrum 3037147 8 - Figure 3, an example of a digital spectrum of a solid injection signal and a spectrum of an attenuated signal; - Figures 4a and 4b, reflectograms respectively obtained without and with the implementation of the method according to the invention.
    [0010] FIG. 1 shows an exemplary architecture of a multi-carrier MCTDR reflectometry device composed of different blocks, able to implement the invention. In this embodiment, the device comprises a first block 1 for setting the test signal x to be injected. This analog signal is obtained from a sampled signal parameterized in this first block. This signal being parameterized on N samples, a sample xn is expressed according to the following relation: N / 2-1 (nk (1) 15 xn = 2 1 ck cos Dr- + 01 k = 0 N () The samples xn are The parameter block is followed by a block 2 which synthesizes the test signal 20 by inverse Fourier transform, at the output of this block 2 the digitized test signal C) is thus obtained. A digital-to-analog converter 3 converts this signal into the analog domain to obtain the signal to be injected x. This MCTDR signal, x, is a sum of sinusoids weighted by a set of coefficients ck, where 61k is the phase of the kth sinusoid. The signal x is injected into the cable 10 to be diagnosed via coupling means 4. These same means make it possible to receive the signal reflected therein. The latter is digitally converted by an analog-digital converter 5, then it is analyzed by processing means 6 to deduce the estimated impulse response h of the cable 10.
    [0011] It can be shown that the module of the analog spectrum X (f) of the injected MCTDR signal is given by the following relation (2), f being the frequency variable: 1 / -1 +00 IX Cd = fmaxICkI k = 0 n = ## EQU1 ## where f max is the maximum frequency constituting the signal MCTDR, being the Dirac function.
    [0012] 10 This relation shows that when canceling or attenuating a coefficient ck, it impacts in particular f = kfmax By way of example, if N = 256 and fmax = 200 MHz, and if it is desired to cancel the signal MCTDR in the band 10 to 20 MHz: 15 - k1 = 256 x 10 2 T 13 and k2 = 256 x 2 2T ° 0 26 - and one parameter c13 = c14 = = c26 = 0 is calculated.
    [0013] The MCTDR signal is therefore strongly denatured by the very high cancellation or attenuation of these coefficients, which introduces bias in the analysis of the impulse response and therefore in the diagnosis of the cable. The object of the invention is in particular to overcome this problem. For this, the invention advantageously uses the fact that the kth component, Xk, of the digital spectrum of the injected MCTDR signal can be written according to the following relation (3): P-1 'sin J 8k-k7cNP) (3 ) P represents the ratio between the sampling frequency 1 / Te and the maximum frequency fmax of the signal MCTDR. To respect Shannon's condition, we must choose P> 2, but an oversampling can be done by taking for example P equal to 10 or 20.
    [0014] In the rest of the description, Xfun is the numerical spectrum for which none of the coefficients ck and Xcancei are attenuated, the spectrum that must be used to comply with the EMC constraints imposed in the target application. Xcancei corresponds to the case where coefficients ck are attenuated. In the invention, the coefficients ck are not canceled but only very attenuated. They may have a very low value, for example of the order of 0.01 but remain non-zero. A coefficient ck not attenuated is equal to 1. Using the above notations, one can also write: Xfun = FIxfuid and Xcancei = F {xcancei}, F {} being as indicated above the Fourier Transform. xfuu and xcancel are respectively the signals with all the coefficients ck not attenuated and with coefficients ck attenuated.
    [0015] We also note Y full and v cancel the respective echoes of the signals xfull and xcancei. Similarly, Yfull = F {Y full} and Ycancel = F {Y cancei}. An example of a digital spectrum Xfun and an example of a digital spectrum X cancel are illustrated in FIG. 3, respectively by a first curve 11 and a second curve 12. The invention makes it possible to obtain an estimate as close as possible to the echo Y full corresponding to the complete MCDTR signal xfuu, without attenuations, from the echo signal v cancel obtained from the incomplete MCTDR signal Xcancel.
    [0016] Yfull and Ycancel verify the set of two relations (4): Yfull = Xfull - H Ycancel = Xcancel - H where H = F {h}, h being the impulse response of the cable.
    [0017] In order to obtain H and thus h = F-1 {H}: - having injected the signal Xcancel; - measured the corresponding echo v cancel; 10 = - - full,, Xcancel - then calculated Xfun Ffx = F {Xcancel}, and Y - cancel = F {Y cancel} a solution could be to estimate Yfull according to the following calculation: Yfull = x full x Ycancel 5 rfu Where 1 1 is the estimated value of Y, -11. Xcancel 15 Unfortunately, this solution poses implementation problems because Xcancel can cancel out due to the coefficients ck but also vanishes systematically at all N points because of the term sin 20 If we note: - (ck; Ok) the coefficients and phases pairs associated with Xcancel; - (1; ÇOk) the coefficient and phase pairs associated with Xfun, the coefficients ck being all equal to 1 for xfull. Advantageously using the relation (3) described above, the kth component of Xfun and Xcancel can be linked according to the following relation, k varying from 0 to NP-1: (4) (1 (71-N) X fek P- 1) J n-kzNP eJ (Çek-Bk) (5) e X cancelk ek-kz-NP Ck cke 3037147 12 -k, which represents the ratio between the kth component of the Xfull digital spectrum and the kth component of the digital spectrum Xcancei, is easily computable from the values of the pairs (ck; Ok) and (1; çok) that are known.
    [0018] Since the injected signal is calculated on N points, to obtain the NP fic values, the signal on N points is calculated and the N calculated values are taken P times, thus obtaining P periods of N points, giving a periodic character to the set. NP reports -k. In the time domain, this amounts to maintaining the injection of each of the N samples P times, which is precisely at the origin of the oversampling. Thus, the kth component of Fui / is: ffullk = fic-Ycancelk (6) 15 N being the number of points or samples and P being the oversampling factor, at least P = 2, in practice P can have a value equal for example to 10 or 20. We will describe later how can be oversampling.
    [0019] 20 The only condition of existence for rifulik is that ck is non-zero, regardless of rank k. Having described the principle of the invention, an exemplary embodiment can now be described. FIG. 2 thus illustrates possible steps for carrying out the method according to the invention. The method can be applied by a device as described in FIG.
    [0020] In a first step 21, an Xcancel probe signal parameterized on N points is injected. In a preliminary step, depending on the conditions or the constraints to be respected, one calculates the coefficients ck which must be attenuated, by attributing for example the value 0,1 to these attenuated coefficients, and 1 to the other coefficients. This first step 21 may be carried out by means of the components 1, 2, 3 of the device of FIG. 1 making it possible to obtain the incomplete MCTDR signal X cancel and coupling means 4. The parameterization of the coefficients ck, and 61k phases in the block 1 can be achieved by means of a suitable interface, well known elsewhere.
    [0021] For practical reasons, in particular at the level of the coupling means 4, the injected signal xcanced can not have a DC component. So, we need co = 0. To avoid problems of division by 0, we therefore impose for example the following condition: 10 E [0, P -1] = 0 (7) pN The relation (7) means that the components Frequencies corresponding to k = 0, k = N, k = 2N, k = 3N, k = (P-1) N of k are zero. This is particularly the case when one parameter co = 0, due in particular to the periodicity of samples'k (P periods of N samples). The signal having been injected into the cable and then reflected, in a second step 22, the reflected signal Ycancel corresponding to the signal X cancel injected is measured. This measurement is performed on NP samples. In practice, the NPs are measured, for example, by applying N first sampling at the sampling frequency 1 / Te described above and by repeating P times these N sampling measurements, the N measurements of each series being out of phase with the N measurements. from the previous series. In this way, the echo v canceller is always sampled at a clock frequency equal to 1 / Te but the P series of out-of-phase measurements actually make it possible to obtain an equivalent sampling frequency equal to 1 / (P.Te. ). An example of oversampling that can be applied is described in particular in document WO 2009/087045 A1.
    [0022] In the exemplary device of FIG. 1, this second step 22 may be implemented by the coupling means 4, the analog-digital converter 5 and by the processing means which receive the digitized echo v cancel. To perform the oversampled measurements as described above, the device comprises, for example, a clock generating signals at the frequency of 1 / Te and a counter. It is also possible to use phase-locked components of the PLL or DLL type. The clock signals transmitted at 1 / Te control the sampling of the echo v cancel digitized at the output of the converter 5, the signal v cancel being sampled at each transmitted clock signal. The counter counts clock ticks. When N clock ticks have been counted, the clock signal is out of phase before engaging a next set of N sampled measurements and so on. The application of such oversampling, realizing NP measurements of samples, advantageously makes it possible to improve the estimation. In a next step 23, Y -cancel = F {is calculated. Y cancel}, from the sample NP samples sampled Ycancel, this calculation being performed by the processing means 6. The Fourier transform used is in fact a discrete Fourier transform. In a next step 24 step, estimate the complete numerical spectrum Yfull on NP points using the ratios' k according to the relations (5) and (6). This estimate can also be made by the processing means 6. In a next step 25, the cable reflectectogram is calculated which is the estimated impulse response h as a function of the estimated value. Full: 25 h = F (8) The simple product in the frequency domain actually represents a correlation between xfuu and Pfull in the time domain.
    [0023] Figures 4a and 4b show the estimated impulse responses h respectively obtained without and with the implementation of the method according to the invention. This is an example of an application in which it is desired to diagnose a 10-meter long twisted-pair cable by avoiding the 48 MHz to 92 MHz band which has been attenuated by 90% (i.e. That the corresponding coefficients ck in this band were set to 0.1) and that the maximum frequency of the MCTDR signal was 200 MHz. For these two figures: The abscissa axis represents the samples from 0 to NP-1, a zoom having been made to focus on the interesting part of the reflectogram. It is possible to convert this axis into distance by knowing the speed of propagation on the cable; - The y-axis represents the reflection coefficient and is graduated in arbitrary units between -100% and + 100%; - A first curve 31 represents the impulse response corresponding to an injected signal without attenuated coefficients, that is to say F-X 7211-Y, 1 {; In FIG. 4a, a second curve 32 represents the estimated impulse response corresponding to a signal injected with attenuated coefficients without compensation. This second curve 32 therefore represents the signal F-1 f * ullYcancell In FIG. 4b, a third curve 33 represents the estimated impulse response corresponding to a signal injected with the attenuated coefficients with compensation, that is to say with implementation of the method according to the invention. This second curve 32 thus represents the signal F. In an alternative embodiment of the invention, it is possible to use an injection signal which is not of the MCTDR type. Indeed, it can be shown that the invention can advantageously be generalized by the use of a ratio defined by the following relationship, corresponding to a regularization of Tikhonov 30 = - X full X c: ancel e. () cancel + e The invention can therefore also be applied to signals such as pulses of any width or, for example, slot-like binary sequences. In this case, we choose in relation (8) an s for 1 {Xf * ullr7full}. (9) 3037147 16 avoid packaging problems. We choose it arbitrarily low, and we increase it as long as the spectral division calculus does not converge. This is the same as that of the previous fourth stage.
    权利要求:
    Claims (12)
    [0001]
    REVENDICATIONS1. A method of diagnosing at least one electrical line by reflectometry measurements, said electrical line (10) being in operation in a given frequency band, the diagnosis being made by analyzing the reflectogram of said line, characterized in that: A complete injection signal having a digital spectrum formed of components weighted by a set of coefficients, in a preliminary step is attenuated the coefficients of the components included in said frequency band leading to obtaining an incomplete digital spectrum; - In a next step (21) is injected into the power line (10) the analog signal formed from said incomplete digital spectrum; In a next step (22), the echo corresponding to said injected signal is recovered by sampling measurements of said echo; - In a next step (23), the numerical spectrum (Y, -cancel) of said echo is calculated; - In a next step (24) we estimate the digital spectrum (17, uu, 1, -, 20 of the echo of the complete analog signal if it had been injected, the kth component ( J ullk / 1 of the spectrum total numerical estimate being equal to the kth component (Y, -cancelk) of the digital spectrum of the received echo multiplied by the ratio (k) between the kth component (Xfuiik) of the digital spectrum of the complete injection signal on the kth 25 component (Xcanceik) of the incomplete digital spectrum - In a next step (25), the reflectectogram of said line is obtained from the estimated complete digital spectrum (ffu11).
    [0002]
    2. Method according to claim 1, characterized in that the injected signal 30 being calculated on N points, the echo is sampled on NP measurement points, the numerical spectrum (Y, -cancel) of the echo being calculated on said NP sampled points, P being strictly greater than 2.
    [0003]
    3. Method according to claim 2, characterized in that P is greater than or equal to 10. 3037147 18
    [0004]
    4. Method according to any one of claims 2 or 3, characterized in that the ratios (-k) are calculated on N points, the N calculated ratios being repeated P times.
    [0005]
    5. Method according to any one of the preceding claims, characterized in that the injected signal is a signal of the MCTDR type.
    [0006]
    6. Method according to any one of the preceding claims, characterized in that the injected signal is not of the MCTDR type, said ratio is defined according to the following relationship: f. X 11 X 1 = 11111 u ciency e. () Xcancel 1 + e 1 0 Xfu11 and Xcancel being respectively the digital spectrum of the complete injection signal and the digital spectrum of the injected signal.
    [0007]
    7. Reflectometry device, suitable for use in diagnosing a line in operation in a given frequency band, characterized in that it comprises at least: a parameterization block (1) for parameterizing the signal at injecting into said line (10) a complete injection signal having a digital spectrum formed of weighted components by a set of coefficients, the coefficients of the components included in said frequency band being attenuated to obtain an incomplete digital spectrum; a block (2) synthesizing in the digital domain the signal to be injected from said incomplete digital spectrum; a digital-to-analog converter (3) converting the synthesized digital signal into an analog signal; Coupling means (4) to said line (10) for injecting said analog signal into said line; an analog-digital converter (5) converting the echo corresponding to the injected signal; processing means (6): performing sampled measurements of said echo and storing the samples; Calculates the digital spectrum (Y, -cancel) of said echo; estimating the full spectrum of the echo of the complete analog signal if it had been injected, the kth component ullk / (r7, 1 of the estimated complete digital spectrum being equal to the kth component (Y of, - cancelk) digital spectrum of the received echo multiplied by the ratio (k) between the kth component (Xfuiik) of the digital spectrum of the complete injection signal on the kth component (Xcanceik) of the incomplete digital spectrum; - obtaining the reflectogram of said line from the estimated full digital spectrum
    [0008]
    8. Device according to claim 7, characterized in that the injected signal being calculated on N points, the echo is sampled on NP measurement points, the numerical spectrum (Y, -cancel) of the echo being calculated on said NP sampled points, P being strictly greater than 2.
    [0009]
    9. Device according to claim 8, characterized in that P is greater than or equal to
    [0010]
    10. Device according to any one of claims 8 or 9, characterized in that the ratios (k) are calculated on N points, the N calculated ratios being taken P times.
    [0011]
    11. Device according to any one of claims 8 to 10, characterized in that the injected signal is a signal of the MCTDR type.
    [0012]
    12. Device according to any one of claims 8 to 10, characterized in that the injected signal is not of the MCTDR type, said ratio is defined according to the following relation: f. X 11 X 1 = 11111 u ciency e. () Xcancel 1 + £ Xfu11 and X cancel being respectively the digital spectrum of the complete injection signal and the digital spectrum of the injected signal. 5 (ffuu).
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3304107B1|2019-04-10|Reflectometry method and device for diagnosing cables in use
    EP2277271B1|2017-07-05|Multicarrier reflectometry device and method for the on-line diagnostic testing of at least one transmission line
    EP0295160B1|1993-02-03|Method for numerically evaluating the frequency and phase of signals and device using such a method
    EP3238210B1|2019-08-07|Method for processing and analysing a signal, and device implementing said method
    WO2013057131A1|2013-04-25|Reflectometry method for detecting soft faults in an electrical cable, and system for implementing the method
    EP2361389B1|2012-08-15|Device and method of distributed reflectometry for diagnosing a transmission network
    FR2897441A1|2007-08-17|DIGITAL RECEIVER LARGE FREQUENCY MEASUREMENT BAND
    Bhandari et al.2018|Unlimited sampling of sparse sinusoidal mixtures
    EP0309336A1|1989-03-29|Method for measuring the frequency of a periodic signal and frequency meter using such a method
    FR2559586A1|1985-08-16|METHOD AND DEVICE FOR MEASURING FREQUENCY WITH COUNTING OF THE NUMBER OF ZERO CROSSINGS
    EP1940023A2|2008-07-02|Bank of cascadable digital filters, and reception circuit including such a bank of cascaded filters
    Kusuma et al.2006|Multichannel sampling of parametric signals with a successive approximation property
    EP2091149A1|2009-08-19|Method and device for downconverting the sampling frequency of a digital signal.
    FR2938083A1|2010-05-07|METHOD FOR IMPROVING RESOLUTION AND CORRECTING DISTORTIONS FOR SIGMA-DELTA MODULATOR AND SIGMA-DELTA MODULATOR USING THE METHOD
    FR2946149A1|2010-12-03|Electrical cable analyzing method for electrical system, involves injecting probe signal with total duration in cable, and locating reference measurement point selected on measurement of injected signal at center of middle portion of signal
    EP1246373A1|2002-10-02|Method and apparatus for processing interferences in signals received by an antenna array
    Santos et al.2020|Imaging Transmission Line Impedance Profiles using Passband Signals and Adaptive Sequence Design
    FR3088436A1|2020-05-15|OPTIMIZED DYNAMIC REFLECTOMETRY DEFECT ANALYSIS SYSTEM
    EP2666026A1|2013-11-27|Method and device for estimating the impulse response of a cable network by means of myopic deconvolution
    FR3050849A1|2017-11-03|METHOD AND DEVICE FOR REDUCING NOISE IN A MODULE SIGNAL
    Lenkov2008|2D input signal reconstruction from results of measuring by digital optical systems
    EP2622360A1|2013-08-07|Method of detecting defects of an array by reflectometry and system implementing the method
    FR2965625A1|2012-04-06|Method for estimating pulse response of e.g. optical fiber network, by short-sighted deconvolution in reflectometry system in electronics field, involves carrying out fine estimation of pulse response by deconvolution using core
    FR2890749A1|2007-03-16|Electric signal period measuring device for interfering system, has correlation circuit calculating autocorrelation function of signal, and circuit modifying delay value based on another signal to find delay value canceling function
    FR2937750A1|2010-04-30|Transponder for receiving continuous wave signal in e.g. electronic field, has local oscillator arranged such that initial phase error is compensated by applying compensation command on input signal before emission
    同族专利:
    公开号 | 公开日
    FR3037147B1|2017-06-02|
    US20180149693A1|2018-05-31|
    EP3304107A1|2018-04-11|
    EP3304107B1|2019-04-10|
    US10345365B2|2019-07-09|
    CA2988169A1|2016-12-08|
    WO2016192980A1|2016-12-08|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    FR2931323A1|2008-05-14|2009-11-20|Commissariat Energie Atomique|MULTI-PORTABLE REFLECTOMETRY DEVICE AND METHOD FOR ONLINE DIAGNOSIS OF AT LEAST ONE TRANSMISSION LINE|
    FR2926141B1|2008-01-03|2010-03-19|Commissariat Energie Atomique|METHOD FOR IMPROVING THE PRECISION FOR DETECTING AND LOCATING DEFECTS BY REFLECTOMETRY IN A CABLE ELECTRICAL NETWORK|
    FR2937146B1|2008-10-15|2011-02-11|Commissariat Energie Atomique|DEVICE AND METHOD FOR DISTRIBUTED REFLECTOMETRY FOR DIAGNOSING A TRANSMISSION NETWORK|
    US20110181295A1|2010-01-22|2011-07-28|Livewire Test Labs, Inc.|Fault detection using combined reflectometry and electronic parameter measurement|
    FR2979994B1|2011-09-09|2013-10-11|Commissariat Energie Atomique|SYSTEM AND METHOD FOR TEMPORAL REFLECTOMETRY FOR NON-AMBIGOUS LOCATION OF AN ELECTRICAL FAULT IN A CABLE|FR3081561B1|2018-05-23|2020-06-12|Commissariat A L'energie Atomique Et Aux Energies Alternatives|BINARY REFLECTOMETRY SYSTEM FOR ANALYSIS OF DEFECTS IN A TRANSMISSION LINE|
    WO2019224137A1|2018-05-23|2019-11-28|Commissariat A L'energie Atomique Et Aux Energies Alternatives|Binary reflectometry system for analysing faults in a transmission line|
    FR3093812B1|2019-03-15|2021-05-07|Safran Electrical & Power|FAULT DETECTION BY REFLECTOMETRY|
    FR3093811B1|2019-03-15|2021-05-14|Safran Electrical & Power|Arc detection by reflectometry|
    CN111610410A|2020-05-27|2020-09-01|上海岩芯电子科技有限公司|SSTDR technology-based photovoltaic cable sub-health detection and positioning method|
    法律状态:
    2016-05-26| PLFP| Fee payment|Year of fee payment: 2 |
    2016-12-09| PLSC| Search report ready|Effective date: 20161209 |
    2017-05-30| PLFP| Fee payment|Year of fee payment: 3 |
    2018-06-12| PLFP| Fee payment|Year of fee payment: 4 |
    2019-06-03| PLFP| Fee payment|Year of fee payment: 5 |
    2021-03-12| ST| Notification of lapse|Effective date: 20210205 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1555080A|FR3037147B1|2015-06-04|2015-06-04|METHOD AND DEVICE FOR REFLECTOMETRY FOR THE DIAGNOSIS OF CABLES IN OPERATION.|FR1555080A| FR3037147B1|2015-06-04|2015-06-04|METHOD AND DEVICE FOR REFLECTOMETRY FOR THE DIAGNOSIS OF CABLES IN OPERATION.|
    CA2988169A| CA2988169A1|2015-06-04|2016-05-17|Reflectometry method and device for diagnosing cables in use|
    US15/578,211| US10345365B2|2015-06-04|2016-05-17|Reflectometry method and device for diagnosing cables in use|
    PCT/EP2016/061004| WO2016192980A1|2015-06-04|2016-05-17|Reflectometry method and device for diagnosing cables in use|
    EP16726807.7A| EP3304107B1|2015-06-04|2016-05-17|Reflectometry method and device for diagnosing cables in use|
    [返回顶部]