Satellite Autonomous Integrity Monitoring of BDS and Onboard Performance Evaluation

BIAN Lang,LIU Xiao,LIU Wenshan,YAN Tao,LEI Wenying,JIA Yizhe,MENG Yansong,ZHANG Lixin

China Academy of Space Technology (Xi'an),Xi'an 710100

Abstract:With the development of satellite navigation technology,the user demands for the integrity of Global Navigation Satellite System (GNSS) have increased more and more.A ground-based monitoring system can hardly report an alarm message to GNSS users during the valid alarming period due to the satellite-Earth propagation delay.It is beneficial to monitor abnormal events and report the corresponding alarms from orbit.Adopting this approach,which is an important feature for future GNSS integrity monitoring,the time needed to provide an alarm is shorter and the system integrity capability is strengthened.The BeiDou Navigation Satellite System (BDS) new generation satellites have the capabilities of satellite autonomous integrity monitoring (SAIM).This paper presents the technical scheme of SAIM,and proposes the integrity monitoring method of both navigation signals and the clocks onboard.The proposed method was verified through the onboard test on the BDS satellites.In addition,we analyzed the integrity telemetry data from the new generation of BDS satellite,including signal delay,power,carrier phase measurement,correlation peak,consistency of pseudo-code and carrier phase,clock phase and frequency step.The analysis results indicated that the quality of the data on orbit met the requirements,and SAIM could monitor effectively any abnormal change of satellite clocks and navigation signal,generate rapidly an alarm message,and transmit it to the user.The alarm time was less than 6 s through the message,and 2 s through non-standard code (NSC).Finally,we present future opportunities for improving the SAIM technology of BDS.

Key words:SAIM,navigation signal,satellite clock

1 INTRODUCTION

With the critical applications of Global Navigation Satellite Systems (GNSS) in military and civilian areas,navigation integrity is increasingly demanded,especially in the field of Safety-of-Life(SoL)[1],and the users are very strict regarding the requirements of integrity.Hence,integrity of a global navigation satellite system is a key requirement for upgrading the system.GPS III and GALILEO have both proposed the development goal of a global CAT I service[2-6].

Due to the limits of territory and other factors,it construct a global monitoring station network,because it was hard to implement a ground facility global integity monitoring solution alone.By relying on the ground monitoring stations alone,the alarm time would be very long,up to tens of minutes or several hours,so it would be difficult to meet the alarm time requirement of users in the field of SoL.So,SAIM would be an effective way to solve the global integrity monitoring of BDS where there would be limited global station distribution.

Any abnormality in the space segment,ground segment or user segment may lead to loss of integrity under abnormal events,so ensuring the security of a navigation satellite system is an extremely complex systematic operation.SAIM needs to ensure that the navigation signal of the satellite remains intact,and the alarm information can be sent out immediately after detecting any abnormal event.In previous work,China Academic of Space Technology (CAST),Xi'an,had carried out systematic research on SAIM technology,designed the scheme of SAIM[7-8],and conducted research on a joint space-ground integrity monitoring system.

The lastest generation of BDS navigation satellites are equipped with SAIM payloads.For the first time,the autonomous integrity monitoring of a navigation signal and satellite clock was carried out on orbit,and the performance evaluation and integrity alarm verification test were conducted.This paper presents the technical scheme of SAIM and the integrity monitoring method for navigation signals and satellite clocks,and analyzes the integrity telemetry data of the latest generation of BDS navigation satellites.The results show SAIM can monitor any abnormal jump of the satellite clocks and navigation signals while on orbit,and can quickly generate alarm information to broadcast to users.The time to alarm for messages is less than 6 s,and time to alarm with non-standard code (NSC) is less than 2 s.Finally the future research direction covering integrity technology development of subsequent BDS system is introduced.

2 MONITORING SCHEME FOR AUTONOMOUS INTEGRITY OF BDS SATELLITES

SAIM receives the navigation signals from the downlink transmit antenna front-end through a wired link,generates a pseudo range,carrier phase,signal power and correlation measurement,and monitors the time and frequency reference signals which generate navigation signals to determine if there is a satellite clock phase and frequency jump.Navigation signal integrity information is generated by integrating the satellite clock and navigation signal monitoring information.The signal measurement data can be further analyzed via telemetry data to support further data analysis and performance evaluation.

Figure 1 The overall scheme of SAIM for BDS

2.1 Selection of Monitoring Measurement

The SAIM monitoring measurement should be selected based upon the users' application requirements.User positioning accuracy can be defined by the formula (1)[12]:

Whereσuis user position error,GDOPis Geometric Dilution of Precision,andσUEREis user equivalent measurement error.

SAIM is independent ofGDOP,and it is only related to the ranging error introduced by the satellite,which is only part of total errors.

A user directly uses the pseudo-range and carrier phase to locate.An abnormal decrease in power usually means that the payload is abnormal on the satellite,and will cause the carrier to noise ratio (C/N0) to decrease.Non ideal characteristics of payloads,such as a temperature variation,AM/AM,AM/PM or other factors,will cause signal delay variation in the payloads,and the pseudo-range measurement system,the signal correlation peak deviation;plus the signal correlation peak distortion can bias the range to the user due to different RF front-end filtering and the correlator interval measurement[13];baseband clock and radio frequency local oscillator (LO) frequency coherence variation will make the pseudo-range and carrier phase measurements inconsistent or divergent,thus influencing the precise positioning information of the user.

The observational measurements of the navigation signals,and the relationship between the abnormal events and the impact on the user are shown in Figure 2.

Figure 2 The relationship between measurement,abnormal events and influence on receiver measurement error

Because the reference frequency which generates all the frequency signals of a navigation satellite is provided by the satellite clock,an abnormal jump of the satellite clock will cause all the navigation signals to be abnormal.

Therefore,the measurement integrity monitoring mainly includes:

1) Navigation signal:pseudo range,carrier phase,signal power and correlation value;

2) Satellite clock:phase jump and frequency jump.

2.2 Monitoring Process and Alarm Type

SAIM only generates an alarm flag for fast changing abnormal events.The slow change in abnormal events are mainly analyzed from the ground,so as to decide whether or not to issue an alarm flag.

An integrity monitoring alarm is mainly based upon an abnormal navigation signal and clock.If it is found that the satellite clock phase or frequency jumps abnormally,and exceeds a preset alarm threshold,warning signals are issued;if the delay or power of a branch signal abnormally jumps,the branch signal alarm will be issued.At the same time,the satellite will transmit the signal measurement information such as pseudo range,carrier phase,power and correlation value to the ground.The ground can further analyze and process the information,taking into account long-term monitoring data of code/carrier phase consistency,distortion evaluation of zero crossing of S curve through correlation peak distortion.

There are two alert methods:

1) Information alarm:this is where an embedded alarm is generated and inserted into the navigation message and broadcast to the user along with other navigation data.

2) Signal alarm:the navigation signal spread spectrum code is set to a non-standard code.

The alarm time of the information alarm is mainly limited to the broadcast cycle of the navigation message,while the non-standard code alarm time is shorter,within 2 s.

3 NAVIGATION SIGNAL MONITORING TECHNOLOGY

The navigation signal monitoring parameters include signal power,signal delay,signal correlation peak distortion,code/carrier phase inconsistency.

3.1 Signal Power Monitoring

A decrease of signal power can be caused by various reasons,such as through the digital DA or the amplifier reduces the C/N0 at the user receiver.In an ideal case,the signal power fluctuates around the mean within a certain range,usually less than 0.5 dB.If the signal power is abnormal,the power measurement value will be seen by an abnormal jump.And the power measurement value can be obtained from the following formula:

WherePWrepresents the signal power.Irepresents the correlation value of the same phase branch received by the signal.Qrepresents the correlation value of the quadrature branch received by the signal,ΔPWis a constant for the calibration of the power zero value.

If the power generates an abnormal jump,it will be noted immediately.And if the default threshold is exceeded,alarm information will be generated.

3.2 Signal Delay Monitoring

The signal time delay anomaly observed by users is a pseudo-range jump abnormal,which may be caused by a pseudo code fault,an instruction anomaly or other payload fault.There is no dynamic on-board monitor,so the pseudo range of the monitoring receiver in normal condition should be seen as a Gauss distribution around a certain range.Because the frequency source of the receiver is different from the satellite clock,the pseudo range measurement will show the phenomenon of a linear drift.Therefore,clock error compensation should be made in the process of signal delay measurement.The actual signal time delay measurement values are obtained from the following formula:

WhereTDelayis signal delay,PRis the pseudo range measurement,andΔTis the clock difference between satellite clock and local oscillator.

3.3 Code/Carrier Phase Inconsistency Monitoring

High precision users will incorporate the carrier phase measurements into the navigation solution.The code/carrier phase inconsistencies observed on the ground are mostly caused by the ionosphere.Code/carrier phase inconsistency is also possible to occur onboard.Although the PN code and carrier are traceable to the onboard atomic clocks,the frequency signals used to generate PN code and carrier phase are through different paths.Any abnormality will lead to pseudo code and carrier phase inconsistency.Usually,the code/carrier phase difference caused by the time &frequency system will be manifested by the divergence between the pseudo range and the carrier phase measurement value.In addition,the relative phase relationship between the PN code and the carrier will also change because of the abnormal situation.This kind of fault will affect the ambiguity solution in precise positioning applications.Therefore,with aspect of carrier/pseudo-code phase consistency monitoring,CAST Xi'an has designed two monitoring schemes for two types of fault modes.

(1) CCD (Code/Carrier Divergence)

A two-order filter is used to estimate the PN code and carrier divergence rate (equivalent to the series of two first order filters),so as to reduce the influence of code noise.The input to the filter is the original PR measurement value minus the carrier measurement value.The filter is represented as (Laplace domain)[10]:

The system is converted to the digital domain,using,at this time:

In the formula (5),z(k)=PRr(k)-φ(k),Tis the sampling interval of the code/carrier measurements,τd1andτd2are time constant.

Firstly,the difference between the pseudo range of the original code and the carrier pseudo range is calculated,andz(k) is obtained.Then two order digital filtering is performed to getd2(k),whered2(k) is the code carrier divergence rate.

(2) CCB (Code/Carrier Bias)

The purpose of CCB is to determine whether the increments of the pseudo-range and carrier phase are consistent in normal operation.

The difference between the increment of code and the carrier is CCB,as follows:

4 SATELLITE CLOCK MONITORING TECHNOLOGY

A high-stability crystal oscillator is the local reference of SAIM payload.The oscillator has good short-term stability,but worse long stability compared with the onboard atomic clocks.Thus SAIM only monitors the satellite clock phase or frequency jump abnormally,without long-term deterioration of performance monitoring.The deterioration of the long-term performance of the atomic clock can be given by the long-term observation in the ground monitoring network and from the inter-satellite link measurements.

The clock difference between the satellite clock and the local reference oscillator for three consecutive periods istk,tk-1,andtk-2respectively,where thekrepresents the current time.It is assumed that the phase clockxis generated at the present time.In order to remove the phase shift introduced by the frequency offset of the satellite clock and the local reference oscillator,we compare the adjacent measured values,as follows:

Δtfis the phase shift introduced by the frequency difference between the satellite clock and the local oscillator.As the phase shift introduced by the frequency offset of the local oscillator is very small in a 3 s period,it can be ignored.SoΔtf(k,k-1)andΔtf(k-1,k-2)can be approximately considered equal.

The difference betweenΔT1andΔT2is as follows:

The phase jump of the satellite clock is obtained from the above formula,and the phase offset is offset by the frequency offset of the local reference oscillator,so the phase jump variable of the satellite clock can be obtained.The frequency jump of the satellite clock can be derived from phase jump.

Figure 3 Evaluation Result of time PR and carrier phase measurement

5 PERFORMANCE EVALUATION RESULTS FROM IN ORBIT OPERATION

A new generation of BDS navigation satellites have been launched,tested and evaluated.The pseudo-range and carrier phase,signal power,satellite clock phase jump,satellite clock frequency jump,the original concept of measurement quality were statistical analyzed.The following figures show the results of the original observation of the latest generation MEO satellites.The pseudo range accuracy is less than 0.08 ns,carrier phase accuracy is less than 0.008 cycle,power monitoring accuracy is less than 0.2 dB,satellite clock phase jump and frequency jump are less than 0.01 ns and 0.1 mHz respectively.

The code/carrier phase consistency monitoring results are as follows:

In addition to the performance evaluation under normal operation,onboard fault injection and the satellite-ground closed loop alarm time test were conducted.The signal and satellite clock faults were generated from ground instructions,for example time delay,power,satellite clock phase and frequency adjustment.The results show that after the fault was generated,SAIM detected the fault in 1 s,and quickly issued an alert to the user.The alarm time was less than 6 s while using navigation message feature.If NSC warning mode was adopted,the alarm time was less than 2 s.Figure 7 shows the partial monitoring results during signal abnormally.

Figure 4 Result of signal power measurement

Figure 5 Result of phase and frequency step measurement

Figure 6 Result of code and carrier consistence monitoring

Figure 7 Result of B1c signal PR step measurement

6 CONCLUSION

At present,SAIM is being used to conduct an on orbit longterm monitoring experiment for monitoring on orbit abnormal events to accumulate integrity faults for navigation satellites.The current monitoring focuses on anomalies such as satellite clock and navigation signal,resulting in an abnormal jump seen through receiver measurements,which can identify the lack of integrity for users due to the navigation satellite's abnormality.The monitoring accuracy meets these requirements.The alarm time is less than 6 s by navigation message,and less than 2 s by NSC.

Looking at the historical operation of GPS,satellite clock and orbit anomalies are also the main types of in-orbit errors.The most of satellite errors is clock failures[11],and the slow failure of the satellite clock accounts for a considerable proportion.BDS satellites may also have similar integrity risk.We will research further into the monitoring scheme and algorithm of the orbit and gradual change in the clock characteristics.We will make a comprehensive analysis of main and backup clock differences and the inter-satellite link observation data,then evaluate the feasibility of autonomous monitoring on orbit.