Locating passive intermodulation fault sources

Abstract

A method for locating the source of a passive intermodulation (PIM) fault in a radio frequency (RF) system, the method comprising: a) generating a PIM fault fingerprint database by: modelling an RF network; defining a plurality of PIM fault patterns for the modelled RF network by defining PIM faults at one or more sources; simulating a received RF signal for each respective PIM fault pattern; generating the PIM fault fingerprint database in dependence on the simulated received RF signals for each defined PIM fault pattern; b) measuring an RF signal for a real network having a PIM fault and abstracting the measured RF signal to define a given signal; c) searching and matching the given signal using the fingerprint database to determine the PIM fault pattern; and d) locating the source of the PIM fault, or giving a further measuring guideline in dependence on the determined PIM fault pattern.

Claims

1. A method for locating the source of a passive intermodulation (PIM) fault in a radio frequency (RF) transmission and receiving system, the method comprising: a) generating a PIM fault fingerprint database by: modelling an RF network; defining a plurality of PIM fault patterns for the modelled RF network by defining PIM faults at one or more sources; simulating a received RF signal for each respective PIM fault pattern; generating the PIM fault fingerprint database in dependence on the simulated received RF signals for each defined PIM fault pattern; b) measuring an RF signal for a real network having a PIM fault and abstracting the measured RF signal to define a given signal; c) searching and matching the given signal using the fingerprint database to determine the PIM fault pattern, wherein the searching and matching the given signal using the fingerprint database comprises using a closest distance algorithm denoted as: $\hat{p}=\arg \underset{p}{\min }\mathrm{Dist}\left(p_{},s\right)$ where {circumflex over (p)} is an estimated fault pattern, a fingerprint representing a p.sup.th PIM fault pattern, and s is the given signal; and d) locating the source of the PIM fault, or giving a further measuring guideline in dependence on the determined PIM fault pattern.

2. The method as claimed in claim 1, the method further comprising: if a further measuring guideline is given, repeating steps (b), (c) and (d) until the source of the PIM fault is located.

3. The method as claimed in claim 2, wherein generating the PIM fault fingerprint database further comprises: a) modelling the operation environment of the RF network; b) defining one output port for each of a plurality of passive sources as a respective signal point in the modelled network; c) defining PIM fault patterns in the modelled network by defining PIM faults at one or more of the plurality of passive sources; d) simulating the transmission and receipt of RF signals for each defined fault pattern; and e) abstracting PIM signal feature vectors in dependence on the simulated received signals for each defined signal point respectively, and storing the PIM fault patterns and the corresponding feature vectors in the fingerprint database as PIM fault signatures for the plurality of passive sources.

4. The method as claimed in claim 2, wherein abstracting the measured RF signal to define the given signal further comprises: measuring RF signals at one or more signal points, wherein each signal point is located in an output port of a passive source; abstracting PIM signal feature vectors at a source signal point, wherein the PIM signal feature vectors represent one or more of received noise floor, power level, spectrum and interference; recording the feature vectors for the given signal.

5. The method as claimed in claim 2, wherein locating the source of the PIM fault further comprises: a) locating the source of a PIM fault; and b) giving a further measuring guideline to further test for other PIM fault sources, measuring an RF signal for a real network and abstracting the measured RF signal to define a given signal, searching and matching the given signal using the fingerprint database to determine another PIM fault pattern, and locating another source of the PIM fault in dependence on the determined PIM fault pattern.

6. The method as claimed in claim 1, wherein generating the PIM fault fingerprint database further comprises: a) modelling the operation environment of the RF network; b) defining one output port for each of a plurality of passive sources as a respective signal point in the modelled network; c) defining PIM fault patterns in the modelled network by defining PIM faults at one or more of the plurality of passive sources; d) simulating the transmission and receipt of RF signals for each defined fault pattern; and e) abstracting PIM signal feature vectors in dependence on the simulated received signals for each defined signal point respectively, and storing the PIM fault patterns and the corresponding feature vectors in the fingerprint database as PIM fault signatures for the plurality of passive sources.

7. The method as claimed in claim 6, wherein modelling the operation environment of the RF network comprises modelling one or more of a building structure and materials, outdoor terrain and clutter.

8. The method as claimed in claim 6, wherein each PIM fault pattern can be defined as [0.sub.1, . . . , 1.sub.i, . . . , 0.sub.K], where an i.sup.th source has PIM faults, it is marked as 1, and others are marked as 0, and K is the total number of PIM sources in the modelled network.

9. The method as claimed in claim 6, wherein PIM signal feature vectors further include one or more of: a received noise floor, a power level, a spectrum, and an interference.

10. The method as claimed in claim 1, wherein the PIM fault fingerprint database further includes the PIM fault patterns and feature vectors, and the fingerprint of a p.sup.th fault pattern is represented by $p_{}=\left\{{\left[0_1,\mathrm{.Math.},1_i,\mathrm{.Math.},0_K\right]}_p,\left[\begin{array}{cccc}{f}_{p,1}\left(1\right),& f_{p,1}\left(2\right),& \mathrm{.Math.},& f_{p,1}\left(N\right)\\ f_{p,2}\left(1\right),& f_{p,2}\left(2\right),& \mathrm{.Math.},& f_{p,2}\left(N\right)\\ \mathrm{.Math.},& \mathrm{.Math.},& ,& \mathrm{.Math.}\\ f_{p,K}\left(1\right),& f_{p,K}\left(2\right),& \mathrm{.Math.},& f_{p,K}\left(N\right)\end{array}\right]\right\}$ where, in the p.sup.th fault pattern, the i.sup.th source comprises the defined PIM faults, K is the total number of PIM sources in the modelled network, f.sub.p,k(n) is an uplink receive noise floor at the n.sup.th sample time, and N is the total sample times.

11. The method as claimed in claim 1, wherein abstracting the measured RF signal to define the given signal further comprises: measuring RF signals at one or more signal points, wherein each signal point is located in an output port of a passive source; abstracting PIM signal feature vectors at a source signal point, wherein the PIM signal feature vectors represent one or more of received noise floor, power level, spectrum and interference; recording the feature vectors for the given signal.

12. The method as claimed in claim 1, wherein abstracting the measured RF signal to define the given signal further comprises: analysing mobile record data in an operation and maintenance mode to abstract PIM signal feature vectors.

13. The method as claimed in claim 1, wherein the given signal is represented by $s=\left[\begin{array}{cccc}{s}_{b_1}\left(1\right),& s_{b_1}\left(2\right),& \mathrm{.Math.},& s_{b_1}\left(N\right)\\ s_{b_2}\left(1\right),& s_{b_2}\left(2\right),& \mathrm{.Math.},& s_{b_2}\left(N\right)\\ \mathrm{.Math.},& \mathrm{.Math.},& ,& \mathrm{.Math.}\\ s_{b_{K^{}}}\left(1\right),& s_{b_{K^{}}}\left(2\right),& \mathrm{.Math.},& s_{b_{K^{}}}\left(N\right)\end{array}\right]$ where K (KK) is the number of measured target signal points, b.sub.k denotes the k.sup.th signal point, which can be mapped to one of a total of K signal points, i.e. b.sub.k=k, k1, . . . , K.

14. The method as claimed in claim 1, wherein the closest distance algorithm calculation comprises using a Euclidean distance method.

15. The method as claimed in claim 14, wherein the Euclidean distance method is expressed as $\mathrm{Dist}\left(p_{},s\right)=\frac{1}{K^{}}\underset{k^{}=1}{\overset{K^{}}{.Math.}}\left(\frac{1}{N}\sqrt{\underset{i=1}{\overset{N}{.Math.}}{\left({p,b_{k^{}}}_{}-s_{k^{}}\right)}^2}\right)$ where is a fingerprint representing a p.sup.th fault pattern at a k.sup.th signal point, s.sub.k is a given feature vector at the k.sup.th signal point, K (KK) is the number of measured signal points, where K is the total PIM sources in the network, and N is the total number of sample points.

16. The method as claimed in claim 1, wherein locating the source of the PIM fault further comprises: a) locating the source of a PIM fault; and b) giving a further measuring guideline to further test for other PIM fault sources, measuring an RF signal for a real network and abstracting the measured RF signal to define a given signal, searching and matching the given signal using the fingerprint database to determine another PIM fault pattern, and locating another source of the PIM fault in dependence on the determined PIM fault pattern.

17. The method as claimed in claim 1, wherein the further measuring guideline comprises one or more of: a) test procedures, such as test sequence, and test cases; and b) suggestions of parameters configuration, such as frequency, ports, and power.

18. The method as claimed in claim 1, wherein a measured PIM fault pattern of the located PIM fault source and corresponding PIM signal feature vectors are retained in the PIM fault fingerprint database as new fingerprints.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The present invention will now be described by way of example with reference to the accompanying drawings. In the drawings:

(2) FIG. 1 shows the mathematic relationship of PIM generation.

(3) FIG. 2 shows the structure picture of traditional PIM testing method.

(4) FIG. 3 shows a block diagram of an exemplary system for locating passive intermodulation faults with two-step process.

(5) FIG. 4 shows an exemplary system for locating passive intermodulation faults.

(6) FIG. 5 shows a flow chat of searching and inferring the PIM fault source.

DETAILED DESCRIPTION OF THE INVENTION

(7) The following description is presented to enable any person skilled in the art to make and use the invention, and is provided in the context of a particular application. Various modifications to the disclosed embodiments will be readily apparent to those skilled in the art.

(8) The general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the present invention. Thus, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

(9) Hereinafter, the present invention will be further described in detail with reference to the accompanying drawings. The invention is described in connection with wireless communications and more particularly relates to measuring passive intermodulation (PIM) and identifying the location of a source of PIM fault, but the invention is not limited to any embodiment. The scope of the invention is limited only by the claims and the invention encompasses numerous alternatives, modifications and equivalents.

(10) In wireless communication systems, passive intermodulation (PIM) is a form of intermodulation distortion that occurs in components normally thought of as linear, such as cables, connectors and antennas. However, when these non-linear, passive components are subjected to the RF power levels found in cellular systems, they behave like a mixer, generating new frequencies that are mathematical combinations of the downlink frequencies present at the site, as shown in FIG. 1 at 101 and labelled by PIM. PIM generated products will affect received band (UL) by increasing the received noise floor thus reducing base station sites coverage and capacity.

(11) Although in the network design, RF engineers will stringently control the PIM to guarantee the network coverage and capacity, after the network is installed, and the PIM sources can start to fault due to the corrosion and oxidation of devices, loose and dirt at the connectors, and environment, such as cold climate, moisture, and wind. Then, the network performance, such as coverage and capacity, will degrade gradually, and the user experience will be worse increasingly. The situation will be more complex in multi-operator, multi-technology and multi-band wireless systems. Therefore, locating the PIM fault sources is very important for today's wireless networks for both indoor and outdoor scenario.

(12) The present invention provides a method for locating passive intermodulation (PIM) fault based on PIM fingerprint in a radio frequency (RF) transmit/receive system, comprises a two-step process. As a first step, a training phase is performed, which includes modelling a wireless network and creating a fingerprint database with one or multi-PIM sources faults. In the second phase, when the source has PIM fault, testing the PIM signal and matching the signal with the fingerprints in the database. The location of PIM fault source or further test guidance is given in the second phase.

(13) In radio communications, passive sources have a passive intermodulation (PIM) value. When the PIM value of a source increases, the radio communication service may be negatively affected, or even dysfunctional. This occurs when, as referred to herein, PIM sources get fault. In other words, a source may be said to develop a PIM fault when the PIM value of that source increases above a threshold level such that a radio communication service is negatively affected.

(14) FIG. 3 shows the block diagram of components of the embodiment of the present invention using fingerprint as a means of determining the location of the PIM fault sources. The solution includes two phases, which are training phase 301 and locating phase 302.

(15) In the training phase 301, by using the distributed antenna system (DAS) planning and optimization tool, such as iBuildNet, the communication network and environment can be modelled 303 in detail, that may include building structure and materials, outdoor terrain and clutter, network and connection network external environment, and so on. FIG. 4 shows an indoor DAS system, where multi-technology and multi-band signals share the same network. In the network, the placed devices' features (loss, gain, PIM etc.) will be configured as well as their connections with different cables (i.e. material, length, loss per metre). Such a model shall also be able to give the PIM result of each port for the given scenario.

(16) Based on the link calculation, the link signal transmission and reception can be calculated and simulated 303, so the characteristics of the PIM 304 fault result for each port can be collected, such as noise floor, power level, spectrum, and interference. Such kind of information will be abstracted into the feature vectors to serve as unique fault signatures, and stored in a database as fingerprints 305 of the corresponding trouble case.

(17) Assuming there are K passive sources in the network, and are marked into {1, 2, . . . , K}. For each source, the signal point at the port can be defined where the PIM-related result of source can be tested, so there will be K signal points, as shown in FIG. 4, the signal point is located at the input port of each passive source, and there are total 20 signal points in this case. The uplink Rx noise floor is f.sub.p,k(t) after the uplink signal passes the passive source, where t is the time variable. Based on a specific sample time the sampled noise floor can be denoted into
f.sub.p,k=[f.sub.p,k(1),f.sub.p,k(2), . . . ,f.sub.p,k(N)]
where p is the fault pattern of passive sources, N is the sample points, and k is the kth signal point. If one or multi passive sources are faulty, the fault pattern p can be denoted into:
p=[0.sub.1, . . . ,1.sub.i, . . . ,0.sub.K]
i.e. if the i.sup.th source has the PIM faults, it is marked as 1, and the others without PIM faults are marked as 0. If there are two sources, i.sup.th and j.sup.th passive sources, have the PIM faults, the corresponding location will be marked as 1, and the others will be marked as 0, and the pattern is p=[0.sub.1, . . . , 1.sub.i, . . . , 1.sub.j, . . . , 0.sub.K]. Therefore, the relationship between PIM fault source and corresponding uplink Rx noise floor at k.sup.th signal point can be written as:
.sub.p,k={(p,f.sub.p,k)|[0.sub.1, . . . ,1.sub.i, . . . ,0.sub.K],[f.sub.p,k(1),f.sub.p,k(2), . . . ,f.sub.p,k(N)]}

(18) If the i.sup.th passive source gets the fault, i.e. PIM is fault, and the PIM value will increase, that makes the PIM-related value is larger than a fix threshold, where the threshold can be configurated into 120 dBm, the noise floor feature vectors at each signal point can be abstracted as f.sub.p,k(t). Therefore, the p.sup.th fingerprint can be stored as (PIM fault pattern, feature vectors), i.e.

(19) $p_{}=\left\{{\left[0_1,\mathrm{.Math.},1_i,\mathrm{.Math.},0_K\right]}_p,\left[\begin{array}{cccc}{f}_{p,1}\left(1\right),& f_{p,1}\left(2\right),& \mathrm{.Math.},& f_{p,1}\left(N\right)\\ f_{p,2}\left(1\right),& f_{p,2}\left(2\right),& \mathrm{.Math.},& f_{p,2}\left(N\right)\\ \mathrm{.Math.},& \mathrm{.Math.},& ,& \mathrm{.Math.}\\ f_{p,K}\left(1\right),& f_{p,K}\left(2\right),& \mathrm{.Math.},& f_{p,K}\left(N\right)\end{array}\right]\right\}$

(20) In the fingerprint database generation, the faults of all passive sources, i.e. connections and devices, should be calculated and the fingerprint should be stored. Assuming there are total P passive source fault patterns, based on all fault pattern P, the fingerprint database is
={.sub.p|p=1, . . . ,P}

(21) At the same time, based on the frequent faults of some passive sources, some fault patterns 306 can be created and saved in the database to constrain the inference of PIM fault sources. For example, in the network, connectors' fault will be up to 60.97%, and jumper cable will be up to 32.26%, and so on. Based on the statistic of PIM faults, the pattern can be used to define the searching and inferring method in the locating phase.

(22) In the locating phase 302, when network performance, such as coverage and capacity, is degraded due to the PIM fault, i.e. the increasing noise floor reduces the network coverage and capacity, a PIM testing 307 for the network will be done, as also shown in 501, and record the signal variation, such as noise floor, interference, and so on. Assuming the tested noise floor at the k.sup.th signal point is s.sub.k(t), the N sampled values at the k.sup.th signal point can be written as:
s.sub.k=[s.sub.k(1),s.sub.k(2), . . . ,s.sub.k(N)]
If K (KK) signal points are measured, the measured vectors can be denoted as

(23) 0$s=\left[\begin{array}{cccc}{s}_{b_1}\left(1\right),& s_{b_1}\left(2\right),& \mathrm{.Math.},& s_{b_1}\left(N\right)\\ s_{b_2}\left(1\right),& s_{b_2}\left(2\right),& \mathrm{.Math.},& s_{b_2}\left(N\right)\\ \mathrm{.Math.},& \mathrm{.Math.},& ,& \mathrm{.Math.}\\ s_{b_{K^{}}}\left(1\right),& s_{b_{K^{}}}\left(2\right),& \mathrm{.Math.},& s_{b_{K^{}}}\left(N\right)\end{array}\right]$
where b.sub.k denotes the k.sup.th signal points, which can be mapped to one of the total K signal points, i.e. b.sub.k =k,k 1, . . . , K.

(24) When searching and matching 310 the PIM testing results with the fingerprint in the database, a closest case(s) is(are) found, as shown in 502. The closest distance algorithm can be used to locate the PIM source, where the closest distance may be the Euclidean distance. If the minimum distance between the measured vector and the fingerprint is the fault pattern based on one source case, then the PIM fault source 311 can be found; otherwise, based on the fault pattern with minimum distance, a further test guideline 312 can be given to test the specific sources, until the final PIM fault source can be found, as shown in FIG. 5 at 505.

(25) The closest distance of signal space is denoted as Dist(.) function, which can be the Euclidean distance. Therefore, calculate the closest distance between the test vector and the fingerprint as follows:

(26) $\hat{p}=\arg \underset{p}{\min }\mathrm{Dist}\left(p_{},s\right)$
where {circumflex over (p)} is the estimated fault pattern, as shown in 503. In the closest distance calculation, the Euclidean distance method can be expressed as

(27) $\mathrm{Dist}\left(p_{},s\right)=\frac{1}{K^{}}\underset{k^{}=1}{\overset{K^{}}{.Math.}}\left(\frac{1}{N}\sqrt{\underset{i=1}{\overset{N}{.Math.}}{\left({p,b_{k^{}}}_{}-s_{k^{}}\right)}^2}\right)$

(28) If the test points are properly selected, the PIM fault source could be located, as shown in 505, otherwise, further test guide is given to position the fault source, as shown in 506, including the guideline (i.e. test procedures) to further test the potential ports (i.e. different powers, frequencies) to minimise the total testing effort, or the test procedure (i.e. test sequence and cases) to exactly locate the fault device and connection. This is as shown by 504. For example, after the matching algorithm, a faults pattern {0.sub.1, . . . , 1.sub.i, . . . , 1.sub.j, . . . , 0.sub.K} is obtained, as shown in FIG. 4, if i and j signal points are 5 and 10 at the first floor, respectively, the further signal points C, D, and E need to be measured to abstract the new measured vectors, and test sequence is (C, D, E). Correspondingly, the measured vector is

(29) $s=\left[\begin{array}{c}{s}_C\left(1\right),s_C\left(2\right),\mathrm{.Math.},s_C\left(N\right)\\ s_D\left(1\right),s_D\left(2\right),\mathrm{.Math.},s_D\left(N\right)\\ s_E\left(1\right),s_E\left(2\right),\mathrm{.Math.},s_E\left(N\right)\end{array}\right]$
and further closest distance of signal space will be calculated, until the final fault source is inferred. The flow chart of searching and inferring the PIM fault source is shown in FIG. 5.

(30) As shown in FIG. 5, the searching an inferring the PIM fault source comprises the following steps. Step 501 shows starting of the PIM testing. Step 502 shows searching the PIM testing results with the fingerprint in the database to search and find the closest case(s). Step 503 shows calculating the estimated fault pattern of those closest case(s). Step 504 shows deciding whether the fault sources can be located based on the calculated fault pattern. Step 505 shows that if the fault sources can be located that the fault sources are reported. Step 506 shows that if the fault sources cannot be located then a further test guideline can be generated based on analysing the generated fault pattern from step 503.

(31) Alternatively, in the network operation & maintenance, user equipment feedback information can also be used to analyse the PIM fault instead of the PIM testing, so the mobile record (MR) data 308 can be used to abstract PIM fault feature vectors by uplink interference data analysis 309, where the uplink SINR (signal to interference plus noise ratio) will significantly increase if the source PIM fault, and the increasing of interference will result in the increasing of noise floor.

(32) Once is finished, the method will be able to determine whether to retain these result as new fingerprints into the database. Next time if similar symptom is met, the method will be able to provide the solution immediately.

(33) A detailed description of the preferred embodiment of the present invention specific or more. It should be understood that one of ordinary skill in the art without creative work to many modifications and variations may be made according to the teachings of the present invention. Therefore, all those skilled in the art under this inventive concept on the basis of prior art technical solutions through logical analysis, reasoning or limited experiments could be obtained, are to be made within the scope of the claims determined.

(34) The applicant hereby discloses in isolation each individual feature described herein and any combination of two or more such features, to the extent that such features or combinations are capable of being carried out based on the present specification as a whole in the light of the common general knowledge of a person skilled in the art, irrespective of whether such features or combinations of features solve any problems disclosed herein, and without limitation to the scope of the claims. The applicant indicates that aspects of the present invention may consist of any such individual feature or combination of features. In view of the foregoing description it will be evident to a person skilled in the art that various modifications may be made within the scope of the invention.