A COMPAROTIVE STUDY OF GPS P3 AND GPS L3

Transcription

1 A COMPAROTIVE STUDY OF GPS P3 AND GPS L3 Gun Li 1 2 Bian Li 1 2 Shao-Wu DONG 1 1 Natal Time Service Center Chinese Academy o Sciences 3 Eastern Road o Lintong Reg Xi an Shaanxi CHINA 2 Graduate University o Chinese Academy o Sciences 19Yuquan Rd. (ia) Shiingshan Beiing CHINA Abstract-More recently GPS carrier phase standard internatal reerences or requency and time measurement using geodetic receivers is widely Internatal Atomic Time (TAI) and UTC which is conducted or accurate time and requency equal in rate to TAI but adusted by an integer number to comparisons. Time transer experiments were seconds to account or variats in the rotat o the conducted using the Ashtech Z12T receivers whose internal requency is provided by an external 20 MHz cloc and an external 1 pps (pulse per second) o a reerence cloc. The results o time comparison has shown that the stability o time transer using carrier phase is better than that o GPS common-view time transer using GPS C/A or P3. The sel-developed processing sotware was used to the cycle slip detect and ambiguity resolut determinat. The obective o this paper is to compare the GPS P3 and GPS L3 time transer. The GPS P3 and L3 data have been obtained using an Ashtech Z12T at NTSC and an Ashtech ZIIX3 at NICT. The main results and analysis are given in Sect 3 and Sect 4. For a long baseline comparison the RMS o L3 is better than that o P3. The stability o requency o L3 is earth. [12] Each time laboratory () contributing to the determinat o TAI maintains a local realizat o UTC called UTC(). Most time lins used or TAI have based on GPS common view technique and on Ku-band Two-way time transer using geo-statary satellites (TWSTFT) [23]. GPS code-based common view or carrier phase common view are carrier out by having two ground stats observe the same satellite at the same time These techniques have the advantage o removing the GPS satellite cloc error atmosphere delay broadcast orbit and broadcast osphere corrects. It is very important to mitigate the delay o osphere in time transer. In the days o single-requency receivers a osphere delay correct model was requires so that mitigat the propagat delay due to osphere. Lots o old receivers use the broadcast Klbuchar model or the better than that o P3 or short term τ < 6d ) correct since the irst use o IGS (Internatal GPS however or τ > 6d the stability o L3 and P3 is almost at the same level. service) osphere TEC-map (total electron content) as one useul method in July 1999 the corrects are now I. INTRODUCTION applied to all TAI GPS common-view lins [4] which has improve the precis o osphere mitigat The Bureau Internatal des Poids et Measures (BIPM) in France is charged with providing the time standard UTC (Coordinated Universal Time). The BIPM collects data via GPS or TWSTFT rom more than 200 atomic clocs and a ew primary absolute requency standards rom more than 50 time laboratories around the world. Once a month BIPM uses these data to produce the signiicantly. Recently GPS carrier phase measurement using geodetic-lie receivers Ashtech Z12T is widely conducted or accurate time and requency comparisons which have been perormed or determining cloc dierences at the level o a ew hundred pico-seconds. [56] The precis o GPS carrier phase time transer is approximately 10 times better than GPS common view /05/$ IEEE. 672

2 time transer. [6] The potential capability and applicat using GPS carrier phase rather than C/A code with the common view technique to transer precise time and requency has been recognized described and discussed by many researchers in the reerence [6~14]. The one main reason is dual requency carrier phase measurement provide osphere-ree technique L3. Since 2003 GPS P code measurements obtained with calibrated geodetic processing technique have widely used or P3 measurement which provide another useul method or correcting the osphere delays in time transer. While geodetic-quality carrier phase receivers have demonstrated high quality requency transer results the Ashtech Z12T receiver used or precise time transer requires nowledge o the hardware delays at each site. In the experiment the receiver at NICT has calibrated in The receiver at NTSC has also calibrated by BIPM s traveling receiver but the calibrat has something wrong thereore the obective o this paper is to compare the stability o GPS P3 and L3 techniques. In contrast to osphere delays we estimate the stability level that may be achieved by GPS P3 and L3 by these two methods. In this paper P 1 will represent the precise pseudorange on requency 1 P 2 will represent the precise pseudorange on requency 2 and the P 3 osphere-ree linear combinat o P 1 and P 2. liewise L1 denotes the carrier phase measurement on 1 L2 carrier phase measurement on 2 L3 osphere-ree linear combinat o L 1 and L2. Precise orbital data and the modeling o troposphere should be used in order to obtain high precis corrects to the time transer measurements. The precise data obtained rom IGS website. The data used at these experiments are the RINEX iles generated by the geodetic GPS receiver. 2.1 GPS observable equats In this sect we represent the principle o P3 and L3 briely which have already introduced in details in Reerence [6]. The GPS pseudorange observables i or a given satellite receiver i and requency can be written as ollows: Where = P+ c( δ δ ) + δ i r s trop i + δ δ + ε m (1) is the pseudorange o receiver i and the th GPS satellite the geometric range P is the dierence in the satellite posit at the time o transmiss and the receiver posit at the time o recept c is the speed o light in a vacuum δ and δ are the receiver and satellite cloc r s osets with respect to GPS time respectively and δ δ are the propagat delay trop due to the troposphere and the osphere respectively δ is the multipath error ε represents m un-modeled errors and receiver noise. Analogously the typical model o GPS carrier phase observables can be written as φ λ = P+ c( δ δ ) + δ i r s trop + δ + δ + λ b + ε φ φ φ m i Where φ i is the carrier phase measurement o the (2) The main results and analysis are given in Sect 3 and Sect 4. For a long baseline comparison the RMS o L3 is better than that o P3. The stability o requency o L3 is better than that o P3 or short termτ < 6d ) however or τ > 6d the stability o L3 and P3 is almost at the same level. Further study will conduct in the near uture. II. GPS P3 AND L3 MEASUREMENTS 673 receiver i and the thgps satellite andλ is the GPS carrier wavelength the phase biases b i is deined by b = n + δφ δφ (3) i i i Where n is an integer number o cycles δφ i is the

3 un-calibrated delay o the receiver and associated equipment and δφ is the un-calibrated delay originating in the satellite. Both o the Eq. (1) and Eq.(2) or a dual requency GPS receiver the ospheric delay can be eectively removed by using appropriate linear combinat o the L1 and L2 phase data. The troposphere delay can be determined by modelling the troposphere Saastamoinen model which required nowledge o the standard atmosphere values or temperature atmospheric pressure and vapor pressure coeicients. 2.2 GPS osphere-ree observables The osphere is a dispersive medium at GPS requencies (L1= MHzL2= MHz). This causes the group velocity to be delayed by an amount equal in magnitude but opposite to the phase velocity.in other words φ = (4) Furthermore to irst order the osphere delay is proportal to 1 2. This allows construct o the osphere-ree o the pseudorange observable i3 (5) 1 2 i3 = i1 i =2.55i1-1.55i2 δ δ δ δ ε 3 3 i3 = P+ c( r s) + trop + + (6) multi Analogously the osphere-ree GPS carrier phase observable L3 can now be written as where the new phase bias term B b b 1 2 i3 = λ 1 i1 λ 2 i The time dierences between the two reerence sources were estimated by perorming the carrier phase single dierence measurements which are the dierences between the carrier phase measurements simultaneously observed by a pair o receivers rom the same satellite. The carrier and code measurements data in RINEX data ile are used or the single-dierence measurements. It is very important to detect and the repair o cycle slip in the carrier phase measurements. We use a new algorithm or detecting cycle slips in dual requency GPS rather than single requency which including two steps. We also use an ambiguity resolut algorithm in data processing. III. DATA PROCESSING AND ANALYSIS Time transer data or each o the two methods have been studied or the NTSC-NICT in this paper. For comparison the TWSTFT data have been obtained rom regular internatal time transers which tae place three days per wee (Tuesday and Friday). GPS P3 and L3 data have been obtained using Geodetic-lie receiver Ashtech ZIIXT and ZIIX3T at NTSC and NICT. The GPS C/A data have been obtained using a single requency multi-channel VP3 receiver at NTSC a dual requency multi-channel Euro-80 receiver at NICT or comparison. The sel-developed processing sotware was used to the cycle slip detect and ambiguity resolut determinat. The precise orbit data rom IGS is used. The troposphere model used in the processing is the standard Hopield s model. And the multi-path error and un-modeled error was deleted in EQ. (1) and (2) Figure 1 shows the data processing loatchart simply. (9) L 1 2 i3 = φ i1λ1 φ i2 φi1λ = 2.55Φ 1.55Φ i1 i2 L λ = P+ c( δ δ ) + δ i3 3 RT st trop + δ + λ B + ε Φ3 Φ3 multi 1 i3 (8) (7) To urther evaluate the accuracy o time comparisons reduced rom dierent method we propose to use RMS errors o comparison results. Figure 2 and 3 are the time transer Results using GPS carrier phase L3 and P-code P3 common view techniques between NTSC and NICT rom MJD=53490~53538 the RMS is 2.6ns and 1.5ns 674

4 respectively. The receiver at NTSC is un-calibrated so we cannot mae calibrated time transer measurements. Thereore The Modiied Allan variance o TWSTFT GPS C/A GPS P3 and L3 are also represent in order to compare the requency stability o each method.. Figure 4 shows the results o the stability o requency o P3 TWTFT and C/A. The analysis indicates better behavior o TWSTFT or all and 5 represent the results o the stability o requency by two transer methods. The stability by GPS L3 much better than can be achieved with the GPS P3 or τ < 6d whereas orτ > 6d the stability o L3 and P3 is almost at the same level. Fig.3. Data or the GPS L3 [UTC(NTSC)-UTC(NICT)]. Stat A Observat ile GGTTS ile Precise orbit ile Cycle slip detect Phase ambiguity Troposphere delay Ionosphere delay Stat A Observat ile GGTTS ile Precise orbit ile L1L2P1P2 Combinat P1&P2L1&L2 Comparison End Figure 1. The loatchart o data processing. Fig.5. Frequency stability o UTC(NTSC)-UTC(NICT)] by GPS P3 GPS C/A and TWSTFT. Fig.2. Data or the GPS P3 [UTC(NTSC)-UTC(NICT)]. Fig.4. Frequency stability o [UTC(NTSC)-UTC(NICT)] by GPS P3GPS L3. IV. CONCLUSIONS Time transer experiments were conducted using the Ashtech Z12T receivers whose internal requency is provided by an external 20 MHz cloc and an external 1 pps o a reerence cloc. The results o time comparison 675

5 has shown that the stability o time transer using carrier phase is better than that o GPS common-view time transer using GPS C/A or P3. The urther investigat has provided valuable inormat about the stability o carrier phase measurements the 2004 internatal symposium on GNSS/GPS p123~130 December [6] W.Lewandowsi and J.AzoubibTime transer and GPS P3 and L3. The experiments have demonstrated that TAI IEEE Internatal Frequency Control the stability o carrier phase-based is better than GPS Code-based C/A or P3. For a long baseline comparison Symposium p586~ [7] John F.Plumb carrier phase time transer using the the RMS o L3 is better than that o P3. The stability o global positing system. Ph.D dissertat. requency o L3 is better than that o P3 or short term( τ < 6d ) however or τ > 6d the stability o L3 University o Colorado December [8] Petit G. P. Moussay and C. Thomas GPS time and P3 is almost at the same level. The sel-developed transer using carrier phase and P-code processing sotware was used to the cycle slip detect measurements Proc. 10th European Frequency and and ambiguity resolut determinat. We are Time Forum undertaing a number o experiments to improve the stability o GPS P3 and GPS carrier phase time transer. As mented above the sotware or cycle slip detect and ambiguity resolut determinat are more urgently. In addit to obtain calibrat measurement we will re-correct the calibrat results o the Astech Z12T receiver at NTSC. It is important to tae into account the eect o earth solid tide. In the near uture we will [9] Larson K.M. and J. Levine Time transer using GPS carrier phase methods Proc. 29th Precise Time and Time Interval Meeting [10] Petit G. Frequency comparison using GPS carrier phase: Some experimental results Proc. 11th European Frequency and Time Forum [11] Bruyninx C. P. Deraigne J-M. Sleewaegen and P. consider reducing the eect in distant time transer. Paquet Frequency transer using GPS: A ACKNOWLEDGEMENTS The authors wish to than Pro. Yonghui Hu and Pro. Kaixian Shen or valuable help in data processing. Thans are also due to Ms. Wei Liang o the Institute o Soil and Water ConservatChinese Academy o Sciences. REFERENCES [1] Thomas E. Parer and Demetrios Matsais Time and requency disseminat advance in GPS transer techniques. GPS world. p32~37. November [2] Bian Li Zhengming Wang Comparisons o GPS P3 GPS C/A and TWSTT time lins ATF2004 p258~261 Beiing China. comparative study o code and carrier phase analysis results Proc. 30th Precise Time and Time Interval Meeting in press [12] Larson K. and J. Levine Time transer using the phase o the GPS carrier IEEE Trans. on Ultrasonics Ferroelectrics & Freq. Control 45(3) [13] Larson K. L. Nelson J. Levine T. Parer and E.D. Powers A long-term comparison between GPS carrier-phase and two-way satellite time transer Proc. 30th Precise Time and Time Interval Meeting in press [14] Petit G. C. Thomas Z. Jiang P. Uhrich and F. Taris Use o GPS Ashtech Z12T receivers or accurate time and requency comparisons Proc. [3] Gerard Petit Stability and accuracy o geodetic GPS 1998 IEEE Internatal Frequency Control time lins compared to Two way time transer Symposium ATF2004 p1~7beiing China. [4] P. Deraigne G. Petit Use o geodetic receivers or TAI Metrologia Vol.40 No.2 p184~ [5] ChangBoLeeSungHoonYangand YoungouLee internatal time/requency comparisons using GPS 676