WO2000030272A2

WO2000030272A2 - Method and device for precise measurement of return loss over a wide dynamic range

Info

Publication number: WO2000030272A2
Application number: PCT/DK1999/000632
Authority: WO
Inventors: Hans Erik Iversen
Original assignee: Polar Lab
Priority date: 1998-11-13
Filing date: 1999-11-15
Publication date: 2000-05-25
Also published as: WO2000030272A3; DK199801471A; AU3790000A

Abstract

The invention describes a method and an apparatus for extending the dynamic range when measuring return loss on antennas, e.g. on base stations for cellular phone. The invention has only one common variable attenuation/gain element (102) and one common detector (103). This one branch is time multiplexed between FWD (21) and REV (11) sensing part of the directional coupler (DC) by switch means (101). The output of the common detector (103) is time multiplexed to storing means (14, 24) by switch means (104). FWD DC-value (23) is compared to reference value (71) in regulation means (72) and control signal (73) is lead to common variable gain/attenuation means (102). Switch control (105) operate the multiplex switches (101 and 104). Calculating means (31) calculate return loss from FWD and REV DC-values (23 and 13). Only one common branch for both FWD and REV signals is used. This one branch has exactly the same attenuation/gain, frequency characteristic and temperature profile for REV signal as for FWD signal. In principle infinity dynamic range can be obtained.

Description

METHOD AND DEVICE FOR PRECISE MEASUREMENT OF RETURN LOSS OVER A WIDE DYNAMIC RANGE

The invention relates to a method for extending the dynamic range when measuring return loss on a coaxial cabel via a directional coupler. E.g. when one or more transmitters are coupled to an antenna via a feeder coax cable and the return loss of the antenna is measured on the feeder cable . There is a need for monitoring the condition of the antenna on transmitters and base stations for celular telephone. In the GSM mobile phone system big variations in the output power occour due to regulation of power level in each time slot and the number of used frequen- cies. It is desirable to monitor the antenna condition indepently of the actual transmitted power level.

The prior art for measuring returnloss via a directional coupler (DC) is illustrated in figure 1. A part of the transmitted power is coupeld to the forward sensing part of the directional coupler (DC) . This part (21) is designated FWD in this document. Due to mismatch, the antenna reflect a part of the transmitted power. A part of the reflected power is coupled to the reverse sensing part of the directional coupler (DC) . This part (11) is designated REV in this document. The forward detector (22) converts the high frequent FWD signal (21) to a DC signal (23) proportional to the high frequent signal. This signal (23) is designated FWD-value in this document. The reverse detector (12) converts the high frequent REV signal (11) to a DC signal (13) proportional to the high frequent signal. This signal (13) is designated REV-value in this document.

The two DC signals (13) and (23) are lead to calculating means (31), wich e.g. can be an analog logaritmic circuit. Logaritm of the reverse DC signal (13) is subtracted from Logaritm of the forward DC signal (23) . This diference signal (32) is proportional to the antenna return loss in dB. In this method the dynamic range is limited to that of the detectors. If diode detectors are used in the s.c. squarelaw region, the total dynamic range is aproximately 35 dB . For the reverse detector this dynamic range must cover both forward power level range and return loss range, which can be difficult as mentioned above . The PCT patent application no. WO 97/2654 describe a method for extending the dynamic range. This method is illustrated in figure 2. In this method the dynamic range of the reverse detector (12) is extended by regulating the incident power to the reverse detector (42) in response to the detected forward power (21) . In this method the forward power dynamic range is limited by the forward detector (22) dynamic range (aprox. 35 dB) . Patent application WO 97/2654 describe several variations of this method, including methods using 2 voltage vari- able attenuation/amplification means (51 and 61) and 2 detectors (12 and 22) - one set in the forward branch and one set in the reverse branch. Figure 3 illustrates one variation of these methods which require two equal voltage variable attenuation/amplification means (51 and 61) and two equal detectors (12 and 22) . Both voltage variable attenuation/amplification means (51 and 61) are regulated by a feedback loop in the forward branch. The feedback loop concists of a reference voltage (71) , a comparision/integration means (72) a voltage variable attenuation/amplification means (61) and a forward detector (22) . Thereby the output of the forward detector (23) is held at the same level as the reference voltage (71) . The required dynamic range for the detectors is thereby reduced to the return loss range. If the two set of voltage variable attenuation/amplification means (51) and (61) and detectors (12 and 22) are equal, high dynamic range can be obtained. However, in pratice it is very dificult to obtain good tracking between forward and reverse branch. The higher frequency, the more dificult it is to obtain equal attenuation/amplification characteristics, equal frequency characteristics and equal temperature profile of the two branches. Further more, the higher required dynamic range, the more dificult it is to obtain good tracking between the two branches.

The invention describes a method for extending the dynamic range without having concern about tracking error, which is the case when using 2 voltage variable attenuators and 2 detectors. Instead of having two different branches, wich lead to the mentioned tracking error, forward and reverse signal is time multiplexed through the same branch. Thereby obtaining exact tracking between forward and reverse signal. When measuring on a TDMA signal, the multiplexing frequency must be chosen asynchronous from the measured TDMA frequency. In principal infinity dynamic range can be obtained with this method. The practical limitations is set by the maximum power reatings and noise floor of the used components .

This is obtained by means illustrated in figure 4, C h a r a c t e r i z e d in that the invention only has one common variable attenuation/gain element (102) and one common detector (103) . This one branch is time multiplexed between the FWD (21) and the REV (11) sensing parts of the directional coupler (DC) by switch means (101). The output of the common detector (103) is time multiplexed to storage means (14 and 24) by switch means (104) . The FWD DC value (23) is compared to the reference value (71) in the regulation means (72) and the control signal (73) is lead to the common variable gain/attenuation means (102) . The switch control (105) operates the mulitplex switches (101 and 104) . Calculating means (31) calculate return loss from FWD and REV DC signals (23 and 13) .

The said common variable gain/attenuation means (102) can by way of example, but not necessarily, be implemented as a PIN diode attenuator. The said common detector (103) can by way of example, but not necessarily, be implemented as a diode detector. The said switch means (101) can by way of example, but not necessarily, be implemented as a PIN diode switch. The said storage means (14 and 24) can by way of example, but not necessarily, be implemen ted as capasitors. The said switch means (104) can by way of example, but not necessarily, be implemented as a c-mos switch. The said reference value (71) can by way of example, but not necessarily, be implemented as- or be derived from a voltage regulator. The said regulation means (72) can by way of example, but not necessarily, be implemented as an integrating operational amplifier. The said switch control (105) can by way of example, but not necessarily, be implemented as discrete logic. The said calculating means (31) can by way of example, but not necessarily, be implemented as an analog logaritmic circuit.

By using only one common branch for both FWD and REV sig- nal, tracking error between the 2 branches is avoided.

This one branch has exact same attenuaiton/gain, frequen-

SUBSTΓΓUTE SHEET ⁽RULE 26) cy characteristic and temperature profile for REV signal as for FWD signal. The higher required dynamic range, the more important this is, since the voltage variable attenuation/gain means normally has very high sensitivity at high attenuation. Further it is very difficult to obtain exact same temperature profile and frequency characteristics for 2 branches. The higher frequency, the more important this is. Minor production spread, ageing or temperature changes will affect the balance between 2 branches. Modern GSM telephone systems operate in the region 900MHz to 2 Ghz . In this region small changes make a big difference.

Switch control (105) operates switches (101 and 104) synchronous. In one time period, the FWD signal (21) is lead through the switch (101) and the voltage variable attenuation/gain means (102) , converted to a DC value in the common detector (103), and lead through the switch means (104) to storage means (24) . In another time period the REV signal (11) is lead through the switch (101) , and the voltage variable attenuation/gain means (102) , converted to a DC value in common detector (103) and lead through the switch means (104) to storage means (14) . The common voltage variable attenuation/gain means (102) is controlled from the regulation means (72) which compares the FWD DC value (23) to the referece value (71) . In this way, the FWD value (23) is kept at same level as the reference value (71) . The necessary dynamic range of the common detector (103) is thereby limited to that of the return loss range. The required dynamic range for the common detector (103) can be limited further, if the power balance of the incident FWD signal (21) and incident REV signal (11) is displaced in favour of the incident REV signal (11) . Calculation means (31) calculates a value (32) proportional to the measured return loss from the FWD and REV DC signals (23 and 13) .

Different way of implementing the invention is mentioned below.

As mentioned in claim 2 and 8, instead of comparing the reference value (71) to the the FWD value (23) in the regulation means, the reference value (71) can as well be compared to the reverse value (13) or a combination of the forward (23) and the reverse (13) value.

As mentioned in claim 3 and 9, the invention can also be useful when more than 2 radio frequent signals are compared over a wide dynamic range. In this case, one storage means for each radio frequent signal is required, and the switch means must have as many poles as the number of radio frequent signals to be measured.

As mentioned in claim 5, gain means can be incorporated e.g. in common attenuation/gain means (102) for improving sensitivity, or attenuation means can be inserted e.g. in the FWD (21) and the REV (11) sensing branch for improving input return loss .

As mentioned in claim 6, instead of having an analog switch (104) placed after the common detector (103), an analog to digital converter can be insert in its place (201) figure 5. In this way, the analog switch can be replaced by a digital switch or a software switch in a u-computer (202) figure 5. Storing means (14 and 24) can e.g. be analog capacitors or digital memory. Regulation means (72) can be analog comparison/integration means or digitalized regulation means i.e. u-computer (202) figure 5.

As mentioned in claim 7, calculation means (31) can be analog logaritmic circuit or digitalized calculation means (202) figure 5. Output can be return loss, standing wave ratio, reflection coeficient or other. Calculation means (31) can also be omitted. Since the forward value (23) is locked to the reference value, consequently the reverse value (13) contains enough information in itself.

High integration can be obtained when using a u-computer. An examble is illustrated in figure 5.

Below is an overwiev of the figures.

Figure 1 showing traditional prior art technique.

Figure 2 showing a prior art technique for extending dynamic range from patent application WO 97/2654.

Figure 3 showing an other prior art technique for exten- ding dynamic range, also from patent app. WO 97/2654.

Figure 4 showing the invention.

Figure 5 showing an other way of carrying out the invention.

Figure 6 showing attemuation- and afplification means wich can be incorporated in the construction figure 4 or 5

Below the figures is futher explained.

Figure 1 showing prior art technique. One detector in each branch - forward (22) and reverse (12) . Dynamic range is limited to that of the detectors.

Figure 2 showing a prior art technique for extending dynamic range from PCT patent app. WO 97/2654. Adjustable attenuation/amplification means in the reverse branch (41) is controlled from the forward detector (22) .

Figure 3 showing another prior art technique for exten- ding dynamic range, also from PCT patent app. WO 97/2654. Adjustable attenuators i both the forward (61) and the reverse (51) branch are controlled from the forward detector (22) .

Figure 4 showing the invention. A high dynamic range is obtained by having only one branch for attenuation/gain (102) and detection (103), thereby obtaining exact tracking between the forward (21) and the reverse (11) signal. The forward (21) and the reverse (11) signal is time multiplexed through the same branch. Switch control (105) operates switches (101 and 104) synchronous. In one time period, the FWD signal (21) is lead through the switch (101) and the voltage variable attenuation/gain means (102) , converted to a DC value in the common detector (103) , and lead through the switch means (104) to storage means (24) . In another time period the REV signal (11) is lead through the switch (101) , and the voltage variable attenuation/gain means (102) , converted to a DC value in common detector (103) and lead through the switch means (104) to storage means (14) .

The common voltage variable attenuation/gain means (102) is controlled from the regulation means (72) which compares the FWD DC value (23) to the referece value (71) . In this way, the FWD value (23) is kept at same level as the reference value (71) . The necessary dynamic range of the common detector (103) is thereby limited to that of the return loss range. The required dynamic range for the common detector (103) can be limited further, if the power balance of the incident FWD signal (21) and incident REV signal (11) is displaced in favour of the incident REV signal (11) . Calculation means (31) calculates a value (32) proportional to the measured return loss from the FWD and REV DC signals (23 and 13) .

Figure 5 showing an other way of carrying out the invention, where as much as posible is digitalized. Switch control (105) , Switch means (104) , storage means (14 and 24) , calculation means (31) , reference value (71) and regulation means (72) in figure 4, is all replaced in figure 5 by analog to digital converter (201) , u-proces sor (202) and digital to analog converter (203) . Many functions is thereby replaced by software. Output is also digitalized, and communication with main computer can e.g. be a serial data connection.

Figure 6 showing attemuation- and amplification means wich can be incorporated in the construction figure 4 or 5. Amplification means can be incorporated e.g. in common attenuation/gain means (102) for improving sensitivity, or attenuation means can be inserted e.g. in the FWD (21) and the REV (11) sensing branch for improving input return loss.

SUBSTTΓUTΈ SHEET (RULE 26)

Claims

C l a i m s

1. A method for determing the ratio between to radio frequent signals which e.g., but not exclucively, can be a forward sensing signal (21) and reverse sensing signal (11) from a directional coupler (DC) measuring the forward and the reverse power on a feeder cable (F) from a transmitting base station (B) to an antenna (A) , c h a r a c t e r i z e d in that only one common attenu- ation/gain means (102) and one common detector (103) are used for both forward signal (21) and reverse signal (11) wich is time multiplexed via first switch means (101) through the said one common attenuation/gain means (102) and converted to DC values (106) in the said one common detector (103) , and the said DC values (106) are time demultiplexed in switch means (104), and stored in the respective storage means (14 and 24) , thereby return loss (32) can be calculated by calculation means (31) on the basis of stored values (13 and 23) in the said storing means (14 and 24), and the said common attenuation/gain means (102) is controled from comparison/integration means (72) witch compare stored forward value (23) with reference value (71) .

2. A method as defined in claim 1, c h a r a c t e r i z e d in that the said regulation means (72) compare the said reference value (71) with the said stored reverse value (13) or a combination of the said stored forward value (23) and the said stored reverse value (13) . The combination can i.e. be the mean value of the two said values (23) and (13), or an other weigthed combination of the said values (23) and (13) .

3. A method as defined in one or more of claim 1 and 2 c h a r a c t e r i z e d in that it compares more than 2 radio frequent signals. In this case is used one said storage means (14, 24....) for each radio frequent signal that shoud be compared and the switches (101) and (104) must have as many poles as the number of radio frequent signals that shoud be compared.

4. An apparatus for determing the ratio between to radio frequent signals which e.g., but not exclucively, can be a forward sensing signal (21) and reverse sensing signal (11) from a directional coupler (DC) measuring the forward and the reverse power on a feeder cable (F) from a transmitting base station (B) to an antenna (A) , c h a r a c t e r i z e d in that only one common attenuation/gain means (102) and one common detector (103) are used for both forward signal (21) and reverse signal (11) wich is time multiplexed via first switch means (101) through the said one common attenuation/gain means (102) and converted to DC values (106) in the said one common detector (103), and the said DC values (106) are time demultiplexed in switch means (104) , and stored in the respective storage means (14 and 24) , thereby return loss (32) can be calculated by calculation means (31) on the basis of stored values (13 and 23) in the said storing means (14 and 24), and the said common attenuation/gain means (102) is controled from comparison/integration means (72) witch compare stored forward value (23) with reference value (71) .

5. An apparatus as defined in claim 4, c h a r a c t e r i z e d in that further attenuation and/or amplifi- cation means is incorporated in any branch of the struc ture.

6. An apparatus as defined in one or more of claim 4 and 5, c h a r a c t e r i z e d in that the said switch (104) , the said storing means (14) and (24) and the said regulation means (72) can be of any combination of the type: Analog, digital or software.

7. An apparatus as defined in one or more of claim 4, 5 and 6, c h a r a c t e r i z e d in that the said calculation means (31) can be of any type: analog, digital or software - resulting in any kind of output type i.e.: return loss, standing wave ratio, reflection coeficient, or without calculation means at all.

8. An apparatus as defined in one or more of claim 4, 5, 6 and 7 c h a r a c t e r i z e d in that the said regulation means (72) compare the said reference value (71) with the said stored reverse value (13) or a combi nation of the said stored forward value (23) and the said stored reverse value (13). The combination can i.e. be the mean value of the two said values (23) and (13), or an other weigthed combination of the said values (23) and (13) .

9. An apparatus as defined in one or more of claim 4, 5, 6, 7 and 8, c h a r a c t e r i z e d in that it compares more than 2 radio frequent signals. In this case is used one said storage means (14, 24....) for each radio frequent signal that shoud be compared and the switches (101) and (104) must have as many poles as the number of radio frequent signals that shoud be compared.