GB/T 18314-2024Specifications for survey and measurement of global navigation satellite system (English PDF)

4 Basic requirements

4.1 GNSS survey results shall adopt the China Geodetic Coordinate System 2000; the definition of that system and its reference ellipsoid parameters are given in Annex A.

4.2 A unified time system shall be adopted when GNSS data are processed.

4.3 Instruments used for GNSS network survey of any class shall be verified or calibrated by an authorized verification and calibration body, and shall be used within the period of validity of that verification.

4.4 GNSS network survey of any class shall use the mean square error as the technical indicator of accuracy, and twice the mean square error as the limit error.

5 Class division and measurement accuracy

5.1 GNSS survey is divided into classes A, B, C, D and E according to its use and its measurement accuracy.

5.2.1 Class A GNSS survey is used to establish the national first-order geodetic control network and to carry out global geodynamic research, crustal deformation measurement and precise orbit determination.

5.2.2 Class B GNSS survey is used to establish the national second-order geodetic control network, to establish local or urban coordinate reference frames, and for regional geodynamic research, crustal deformation measurement, local deformation monitoring and precise engineering surveys.

5.2.3 Class C GNSS survey is used to establish the third-order geodetic control network and the basic control networks of regions, cities and engineering projects.

5.2.4 Class D and E GNSS survey is used to establish the fourth-order geodetic control network and the control surveys of medium and small cities and towns, and of mapping, cadastre, land information, real estate, geophysical prospecting, exploration and building construction; such work may be carried out by the network RTK method of the satellite navigation positioning reference station network and shall meet the requirements of the corresponding accuracy class of GB/T 39616.

5.3.1 The class A GNSS network is formed by satellite navigation positioning reference stations. According to Table 1 the mean square error of the annual variation rate of the coordinates shall not exceed 2 mm/a for the horizontal component and 3 mm/a for the vertical component, the relative accuracy shall not exceed 1 x 10 to the power minus 8, and the annual average mean square error of each geocentric coordinate component shall not exceed 0.5 mm.

5.3.2 According to Table 2, for class B the point position mean square error shall not exceed 5 mm horizontally and 10 mm vertically, the baseline component mean square error between adjacent points shall not exceed 5 mm horizontally and 10 mm vertically, and the average distance between adjacent points is 50 km; for class C the corresponding values are 10 mm, 15 mm, 10 mm, 20 mm and 15 km; for class D they are 15 mm, 30 mm, 20 mm, 40 mm and 5 km; for class E they are 15 mm, 30 mm, 20 mm, 40 mm and 2 km.

5.3.3 For GNSS survey used to establish the national second-order geodetic control network and the third and fourth-order geodetic control networks, and in addition to the class B, C and D accuracy of 5.3.2, the relative accuracy shall be not lower than 1 x 10 to the power minus 7, minus 6 and minus 5 respectively.

5.3.4 The accuracy of the GNSS geodetic height difference between adjacent points of a network of any class shall not exceed the requirement for the vertical component of the baseline between adjacent points laid down for that class in Table 1 and Table 2.

6 Layout of the GNSS network

6.1.1 A GNSS network should as a rule be laid out class by class; where the requirements for accuracy and density are met, a class may be skipped.

6.1.2 The layout shall follow the principle of optimum design, taking into account the aim of the project, the accuracy required, the material already available for the survey area, and its terrain and transport conditions. The layout of the class A network follows GB/T 28588.

6.1.3 According to Table 3 the number of sides of an independent loop of the class B, C, D and E GNSS networks shall not exceed 6, 6, 8 and 10 respectively.

6.2.1 Points of a network of any class shall be evenly distributed; the distance between adjacent points should not be longer than twice the average point spacing of that network, nor shorter than two thirds of it.

6.2.3 A newly laid out GNSS network shall be connected to the higher class national GNSS points nearby, and the number of connected points shall not be fewer than 3. Where there are satellite navigation positioning reference stations nearby, they should be used first.

6.2.5 Every point of a class A network shall be connected by levelling to a height. Every point of a class B network within 5 km of a national first or second-order levelling line shall be connected by levelling. Points of a class C network shall be connected according to the requirements of the refinement of the regional quasi-geoid, and points of class D and E networks may be connected as circumstances allow.

6.2.6 The accuracy of the height connection shall be not lower than that of second-order levelling for class A and B points, not lower than third-order levelling for class C points, and not lower than fourth-order levelling for class D and E points, or a method of equivalent accuracy may be used. The method and the technical requirements of the height connection follow GB/T 12897 or GB/T 12898.

6.3.1 The name of a GNSS point shall be taken from the place where the point stands; where names cannot be told apart, (one), (two) and so on may be added after the name.

9 Instruments

9.1 The choice of GNSS receiver for a class A network follows GB/T 28588; for class B, C, D and E networks it follows Table 4. According to that table a class B network uses multi-system multi-frequency receivers, at least 4 receivers observing simultaneously and a choke ring anti-interference antenna; a class C network uses multi-system multi-frequency or single-system multi-frequency receivers, at least 3 receivers observing simultaneously and a geodetic antenna; class D and E networks use single-system multi-frequency or multi-system single-frequency receivers, at least 2 receivers observing simultaneously and a geodetic antenna. In all three cases the observables are, as a minimum, carrier phase and pseudorange.

9.2.1.1 A newly purchased receiver, a receiver whose antenna has suffered a heavy impact, a receiver whose parts have been renewed and a receiver whose antenna pairing has been changed shall be fully checked as prescribed before use.

9.2.1.2 The content, method and technical requirements of the receiver check follow CH/T 8016.

9.2.1.3 When receivers of different types work together, a comparison test shall be carried out on a known baseline; a receiver exceeding the tolerance of the corresponding class shall not be used.

9.2.1.4 The circular level and the optical plummet of the antenna or of the tribrach, and the antenna height measuring rule, shall be checked at least once a month during the working period.

9.2.2 Where a ventilated psychrometer, an aneroid barometer or other auxiliary equipment is needed for the GNSS survey, it shall be sent periodically to the metrological verification department and used within the period of validity. The main technical requirements for the meteorological instruments follow Annex C.

9.3.7 When the antenna is placed on a roof, on a high signal tower or on the top of any other structure, it shall be made secure; in thunderstorms lightning protection shall be provided or observation shall be stopped.

9.3.8 While a receiver is stored indoors the room shall be aired periodically, and the receiver shall be switched on for a check every one to two months; the internal battery shall be kept fully charged and the external battery charged and discharged as required.

10 Observation

10.1.1 The technical requirements for the observation of a class A network follow GB/T 28588.

10.1.2 The basic technical rules for the observation of class B, C, D and E networks are given in Table 5. The satellite cut-off elevation angle is 15 degrees for all four classes, and at least 4 valid satellites of the same system shall be observed at the same time in all four classes. The total number of valid observed satellites is at least 20 for class B, at least 6 for class C and at least 4 for classes D and E. The number of observation sessions is at least 3 for class B, at least 2 for class C and at least 1.6 for classes D and E. The session length is at least 23 h for class B, at least 4 h for class C, at least 60 min for class D and at least 40 min for class E. The sampling interval is 30 s for class B, 15 s to 30 s for class C and 5 s to 15 s for classes D and E.

10.1.2 (footnotes to Table 5) The repeated satellites between sessions are deducted when the total number of valid satellites is counted. The session length runs from the start to the end of data recording. A session number of at least 1.6 refers to the network observation mode, in which each station is observed for at least one session and the number of points set up twice is not less than 60 % of the total number of points in the network. Where the point observation mode based on a satellite navigation positioning reference station is used, observation may be continuous, but the observation time shall not be less than the sum of the session times given in the table.

10.1.3 For class B, C, D and E networks the meteorological elements need not be observed; only the state of the weather is recorded.

10.1.4 The name of the observation data file shall carry the station name or station number, the date, the session number and the data type.

10.1.5 GNSS observation should not be carried out in thunder, lightning or storm.

10.2.1 Class B, C, D and E survey may be observed area by area according to the size of the survey area; where it is, adjacent areas shall have at least 4 points in common.

10.4.2 When the antenna is set on a tripod its centring error shall not exceed 1 mm. For class B survey the antenna orientation mark shall point to true north and, once the local magnetic declination has been taken into account, the orientation error shall not exceed +/- 5 degrees; where the orientation mark of the antenna is not clear a mark may be set in advance and the instrument set to it each time, and during simultaneous observation all antennas shall be constrained to be of the same model. The circular bubble on the antenna assembly shall be centred; where the antenna has no circular bubble, the levelling screws of the tribrach may be adjusted until the antenna heights measured in three directions 120 degrees apart differ by less than 3 mm.

10.5.6 The state of the weather, the real-time latitude, longitude and geodetic height and the position dilution of precision are recorded once at the beginning and once at the end of each session. Where the session is longer than 2 h, these items are recorded once more at every full hour of Coordinated Universal Time, relaxed to every 4 h at night.

10.5.7 The antenna height shall be measured once before and once after each session by the method of Annex D. The two values shall not differ by more than 3 mm and their average is taken as the final antenna height; where the difference exceeds the limit, the reason shall be found and the treatment recorded in the remarks column of the survey booklet.

10.5.10 During observation a radio set shall not be used within 50 m of the antenna, nor a walkie-talkie within 10 m.

10.5.12 During a session the receiver shall not be restarted, no self-test shall be run, the satellite cut-off elevation angle shall not be changed, the data sampling interval shall not be changed, the antenna shall not be moved, and the keys that close or delete files shall not be pressed.

12 Data processing

12.1.1 Baseline data of class A and B networks shall be processed with dedicated high-accuracy software; baselines of class C, D and E networks may be solved with the commercial software supplied with the receiver.

12.1.3 For the baseline solution, a class A network shall use as starting points not fewer than 5 evenly distributed IGS stations; a class B network shall use not fewer than 5 evenly distributed class A points or IGS stations; class C, D and E networks shall use not fewer than 3 evenly distributed points of a network at least one class higher than their own.

12.2.1 For class A and B networks the multipath value of the data of each station shall be less than 0.5 m and the data rejection rate shall be less than 20 %.

12.2.2 Where a few points have to be added to the network, the point observation mode based on a reference station may be used. In that mode the checks on repeated baselines and on synchronous and independent loops are not carried out between different points, but the baseline data of different sessions at the same point shall be checked according to 12.2.3 to 12.2.7.

12.2.4 For the field pre-processing of class B baselines and the baseline processing of class C, D and E networks, the difference in length between repeated baselines shall satisfy formula (1), in which the difference in length is expressed in millimetres and the baseline measurement mean square error is calculated according to 12.2.5.

12.2.5 The baseline measurement mean square error of class B, C, D and E networks is calculated by formula (2) from a fixed error in millimetres, a proportional error coefficient in millimetres per kilometre and the baseline length in kilometres, and is graded according to Table 6: the fixed error shall not exceed 3 mm, 5 mm, 10 mm and 10 mm and the proportional error coefficient shall not exceed 1 mm/km, 3 mm/km, 10 mm/km and 20 mm/km for classes B, C, D and E respectively.

12.2.6 The closure of a synchronous loop of a class B, C, D or E network shall not exceed the values of Annex F.

12.3.4 In the baseline vector solution, precise ephemerides shall be used for the fine processing of class A and B baselines, while broadcast ephemerides may be used for class C and lower networks. Tropospheric delay correction shall be applied to the observations of class B, C, D and E networks, and standard meteorological elements may be used directly in the correction model. The solution is carried out session by session; in the multi-baseline solution each session yields one set of independent baseline vectors with their full variance-covariance matrix, and in the single-baseline solution each baseline yields its components and their variance-covariance matrix. For class B and C networks the baseline may be solved as a double difference, single difference or undifferenced solution; for class D and E networks the processing model may be chosen according to the baseline length, but a baseline shorter than 15 km shall be solved as a double difference fixed solution, while for a baseline longer than 15 km the better of the double difference fixed solution and the double difference float solution may be chosen.

12.6.1.1 The overall adjustment of a class A or B network shall be carried out in the China Geodetic Coordinate System 2000 or in the International Terrestrial Reference Frame; where the epochs of the sub-networks differ, they shall be reduced to a common epoch with a plate motion model and a velocity field.

12.6.2.1.1 Once the baseline vectors have passed the checks, a free adjustment is carried out with the three-dimensional baseline vectors and their variance-covariance matrices as the observations and the three-dimensional coordinates of one point as the datum; it outputs the three-dimensional coordinates of each point, the baseline vectors, their corrections and their accuracy.

12.6.2.2.1 Using the baseline vectors from the free adjustment, a three-dimensional or two-dimensional constrained adjustment shall be chosen and carried out in the China Geodetic Coordinate System 2000.

13 Acceptance of the results and material to be handed over

13.1.1 Checking and acceptance of the results follow GB/T 24356. The results handed over for acceptance, including the storage media of the observation records and their backups, shall be complete in content and quantity and undamaged, and all annotations and finishing shall meet the requirements.

13.1.2 A report shall be compiled once checking and acceptance are finished, and the report is written according to GB/T 24356.

13.2 The material to be submitted comprises the survey task book or contract and the technical design; the point descriptions, sky view diagrams, the custody agreement for the survey mark or the land use agreement, and the point selection and monument setting material; the inspection records of the receivers, the meteorological instruments and the other instruments; the station information table for a class A network, or the field observation records, survey booklets and other records for class B, C, D and E networks; the files, material and result tables produced during data processing; the network point layout diagram; and the technical summary and the acceptance report.