Content uploaded by Yuanxi Yang
Author content
All content in this area was uploaded by Yuanxi Yang on Apr 16, 2023
Content may be subject to copyright.
•REVIEW•https://doi.org/10.1007/s11430-022-1069-7
...........................................................................................................
Development trends of the national secure PNT system
based on BDS
Yuanxi YANG1,2*, Xia REN1,2, Xiaolin JIA1,2 & Bijiao SUN1,2
1State Key Laboratory of Geo-Information Engineering, Xi’an 710054, China;
2Xi’an Research Institute of Surveying and Mapping, Xi’an 710054, China
Received April 10, 2022; revised November 24, 2022; accepted January 19, 2023; published online April 6, 2023
Abstract Satellite navigation systems are vulnerable. To guarantee the positioning, navigation and timing (PNT) safety of core
infrastructure, it is necessary to establish a secure PNT system with hybrid physical principles. In this paper, the augmentations of
the BeiDou satellite system (BDS) itself are analysed, namely augmentations through the BDS inter-satellite link, BDS geos-
tationary orbit (GEO) and inclined geostationary orbit (IGSO) satellites, and BDS PNT services supported by low earth orbit
(LEO) satellites. Then, taking BDS as the core component, the comprehensive PNT infrastructure seamlessly covering deep
space and deep ocean is described, consisting of the deep space PNT constellation, the sea-floor PNT sonar beacon network, and
the ground-based low frequency and very low frequency (VLF) long wave radio stations. Moreover, the key technologies of
resilient PNT application matching comprehensive PNT and various autonomous perception PNT information are discussed,
such as resilient PNT sensor integration, the resilient PNT functional model and the resilient stochastic model. As a future
development direction, the key factors of intelligent PNT services are analysed, including the intelligent perception of PNT
application scenes, the intelligent optimization of PNT functional and stochastic models and the intelligent fusion of multisource
PNT information.
Keywords Secure PNT system, Comprehensive PNT infrastructure, Resilient PNT application, Intelligent PNT application
Citation: Yang Y, Ren X, Jia X, Sun B. 2023. Development trends of the national secure PNT system based on BDS. Science China Earth Sciences, 66, https://
doi.org/10.1007/s11430-022-1069-7
1. Introduction
The accomplishment and provision of the BeiDou Global
Satellite Navigation System (BDS) indicate that significant
progress has been made in national space-based positioning,
navigation and timing (PNT) service infrastructure, and in-
dependent and controllable PNT information can be provided
to major national infrastructure operations and economic
construction (Yang Y F et al., 2020,2021). However, similar
to other Global Navigation Satellite Systems (GNSS), there
is inherent vulnerability in the services provided by the BDS.
Due to the low landing power and poor penetrating ability,
the signal of the Radio Navigation Satellite System (RNSS)
cannot serve users in nonexposed spaces (e.g., underground,
underwater, and indoors). Even the obstruction of tall
buildings, trees, and the interference of other electronic de-
vices might lead to the invalidation of space-based PNT
services. In addition, the service performance of GNSS in the
North and South Poles is relatively poor due to the limitation
of the GNSS constellation (Yang and Xu, 2016).
U.S. policy-makers regard PNT as the cornerstone and
significant infrastructure concerning the U.S. national
economy and national defence security and worry about the
vulnerability and security of GPS. Thus, changing the PNT
application rules is advocated, as well as the construction of
new PNT infrastructures (Mcneff, 2010). The U.S. Depart-
ment of Defense and Department of Transportation origin-
© Science China Press 2023 earth.scichina.com link.springer.com
SCIENCE CHINA
Earth Sciences
* Corresponding author (email: yuanxi_yang@163.com)
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
ally proposed constructing a new national PNT system in
2010, planned to be accomplished by 2025 (U.S. Department
of Transportation and Department of Defense, 2010). The U.
S. Defense Advanced Research Projects Agency initiated the
micro-PNT project in 2010, emphasizing the development of
micro-PNT components with micromachinery devices, and
the micro-PNT terminals featuring small size, low power
consumption and better performance (Dalal, 2012). Pro-
fessor Bradford Parkinson proposed the concept of Protect,
Toughen and Augment (PTA) for GPS applications (Par-
kinson, 2015,2017). The U.S. government issued multiple
acts concerning national resilience and secure PNT from
2015 to 2018, emphasizing the reconstruction of the ground-
based PNT system as the backup and complement of GPS (U.
S. Senate, 2015,2017,2018).
Before the completion of BDS-3, Chinese scholars rea-
lized the structural risks in a single satellite navigation sys-
tem, and started exploring the development direction of the
Chinese PNT system. In 2016, the concept of comprehensive
PNT was proposed by designing a seamless PNT source
system from deep space to deep ocean with different physical
principles (Yang, 2016). However, the comprehensive PNT
only provides multiple PNT information sources, while the
fusion of multiple pieces of information should also be rea-
lized to support the safe operation of major infrastructures
and the users with high continuity requirements. In 2018, the
resilient PNT conceptual framework was proposed to realize
the resilient integration of multisource PNT sensors, the re-
silient modification of functional models and stochastic
models, and the resilient data fusion of multiple PNT in-
formation sources in diverse environments, and guarantee
the high continuity and availability of PNT service (Yang,
2018).
The proposal of a resilient PNT conceptual framework
gave rise to research on resilient PNT theory and technology.
In 2020, President Trump signed the Executive Order on
Strengthening National Resilience Through Responsible Use
of Positioning, Navigation and Timing Services, aiming to
protect national infrastructure from the interruption of PNT
service through the responsible use of PNT services by both
the government and local organizations (Executive Office of
the President, 2020). Many other scholars have also char-
acterized resilient PNT systems from a macro perspective
(Aresta, 2017;Scholz, 2020).
As we know, the normal operation of almost all major
national infrastructures, such as transportation, finance,
power and communication, have high demands for PNT
service. As the GNSS cannot always provide continuous,
reliable or stable PNT service, a high security PNT system
should be constructed from the following four aspects. First,
the GNSS constellation should be augmented to improve the
PNT performance, and reduce the service interruption caused
by system failure. Second, a comprehensive PNT infra-
structure with hybrid physical principles should be con-
structed to tap complementary PNT information source, and
expand the service coverage from deep space to deep ocean
and from the outdoors to indoors. Third, resilient PNT
terminals should be developed with multisource awareness,
enabling the autonomous perception and selection of avail-
able PNT sources in complex environments and ensuring the
continuity and reliability of PNT services. Last, an intelligent
PNT application system should be developed to intelligently
integrate reliable and available PNT sources according to
different scenes, and realize intelligent fusion of multiple
PNT sources.
This paper describes the secure PNT system architecture
from the aspects of space-based PNT augmentation, com-
prehensive infrastructure augmentation and user terminal
resilient augmentation, analyses the related key technologies
and predicts the future development.
2. Space-based PNT augmentation
Spaced-based PNT remains the core infrastructure of global
PNT services. The GNSS constellations are generally de-
ployed at 27000 km altitude, and the Geostationary Orbit
(GEO) satellites and Inclined Geosynchronous Orbit (IGSO)
satellites of BDS are operated at 36000 km altitude. Ac-
cording to World Radio Communication Conference Re-
solution 609, the landing power flux density (PFD) of the
RNSS signal in the 1164–1215 MHz band transmitted by any
single satellite should be lower than –129 dBW m–2 MHz–1,
and the aggregate equivalent power flux density (AEPFD)
produced by all space stations of all RNSS systems should be
below –121.5 dBW m–2 MHz–1. With such a low landing
power, the GNSS signal is easily sheltered, interfered with
and spoofed, regardless of whether it is encrypted or not. In
addition, ground tracking and operational control systems are
indispensable for GNSS to update parameters such as sa-
tellite orbits and clock errors. Thus, to improve the perfor-
mance of space-based PNT services, BDS could be
augmented with the inter-satellite links (ISLs), high earth
orbit satellites and Low Earth Orbit (LEO) satellites (as
shown in Figure 1).
2.1 Inter-satellite link augmentation
The roles of the ISLs may be enlarged. The distances pro-
vided by the Ka-band phased array ISL and laser ISL of BDS
are important measurements to improve the global PNT
performance of the BDS. First, the ISLs establish the com-
munication channel between satellites, making it possible to
transfer information from BDS Global Short Message
Communication Service (GSMCS) and BDS Medium Earth
Orbit Search and Rescue (MEOSAR) within the constella-
2Yang Y, et al. Sci China Earth Sci
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
tion. Second, the ISLs help to realize ranging and time
synchronization between satellites to autonomously maintain
orbit determination within a short time (Yang and Ren, 2018;
Ren et al., 2017,2019;Yang Y F et al., 2021). Third, ISL
measurements can better constrain the constellation geo-
metry and achieve precise orbit determination of the whole
arc with only partial ground tracking observations (Yang et
al., 2018;Yang Y F et al., 2020,2021). In addition, with
stable and precise Hydrogen Maser clocks mounted on some
of the BeiDou satellites, the ISLs may help to maintain the
unified time system of the satellite constellation by using the
quasi-stable adjustment among the satellite clocks to im-
prove the time keeping accuracy (Yang Y X et al., 2021a;
Yang Y F et al., 2021).
2.2 BDS high orbit constellation augmentation
The IGSO satellites may help GEO satellites provide similar
featured services. GEO satellites are important for perfor-
mance improvement and service function expansion of the
BDS (Yang Y X et al., 2020b,2021a). On the one hand, GEO
satellites provide uninterrupted signals for users in the Asia-
Pacific region to improve the availability of regional PNT
services. Even if only one Medium Earth Orbit (MEO) sa-
tellite is visible, the standard PNT service is still available
with the support of the three GEOs. On the other hand, GEO
satellites provide BeiDou Satellite-Based Augmentation
System (BDSBAS) and BDS-3 Precise Point Positioning
(PPP-B2b) services for users in China and the surrounding
areas by transmitting precise satellite orbit and satellite clock
offset correction parameters. The featured services of Re-
gional Short Message Communication Service (RSMCS)
and Two-Way Satellite Time and Frequency Transfer
(TWSTFT) service, which were developed at BDS-1 and
BDS-2, are still retained.
The BDS IGSO satellites are significant information
sources of the augmented regional PNT service, and can
reduce the ‘South Wall’ impact of the GEO satellites, which
means that the IGSO satellites could replace the GEOs to
provide the same services if the GEOs in the south are in-
visible to the users. On the one hand, the combination of the
IGSOs can increase the number of visible satellites and de-
crease the value of Position Dilution of Precision (PDOP).
On the other hand, IGSOs can provide an RSMCS with
onboard RSMC devices, and join the services of BDSBAS
and PPP-B2b, if parameters such as SBAS parameters, pre-
cise ephemeris and Differential Code Bias (DCB) are up-
loaded (Yang Y X et al., 2020b,2022). Actually, the Japanese
QZSS has realized regional PPP service.
2.3 LEO constellation augmentation
The LEO constellation is an important means to augment
GNSS services. With a relatively shorter transmission dis-
tance, the LEO signals usually have larger landing power,
better penetrability and anti-interference capability com-
pared with those of the BDS-3 satellites, and the signal
power is easier to enhance. The orbit determination (OD)
accuracy of BDS-3 satellites can be improved by combined
OD with LEOs, indirectly augmenting the BDS PNT per-
formance (Zhao et al., 2017;Yang Y F et al., 2020). In
addition, with a higher speed of the satellites relative to the
ground users, the Doppler effect of LEOs is much more
remarkable, and a better velocity measurement accuracy can
be achieved. Similarly, with a quickly changing geometry
structure, the correlation between the LEO constellation
measurements of each epoch is relatively low, which can
improve the resolution of PNT parameters and reduce the
convergence time of carrier-phase ambiguity parameter es-
timation (Zhang and Ma, 2019).
The LEOs could contribute to the PPP service. On the one
hand, with the constellation consisting of LEOs and BDS
satellites, the number of available satellites for PNT service
is increased, the DOP value is reduced, and the PNT per-
formance is improved. Even without extra ground-tracking
stations, the point positioning performance can be improved
as well. On the other hand, the LEOs can broadcast the
precise orbits and clock error parameters of BDS-3 globally,
and extend the current service scope of BDS PPP-B2b with
the support of a global tracking system.
3. Comprehensive PNT infrastructure
A comprehensive PNT system is the integration of various
PNT information sources based on various physical princi-
Figure 1 Space-based PNT augmentation.
3
Yang Y, et al. Sci China Earth Sci
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
ples (Yang, 2016), including the comprehensive PNT infra-
structure and comprehensive PNT application system. The
comprehensive PNT infrastructure refers to the large-scale
PNT information sources built artificially, such as the La-
grange point navigation satellite constellation, medium and
high earth orbit satellite constellations, LEO augmentation
constellations, ground-based augmentation station nets, in-
door positioning beacon networks, sea surface positioning
buoy networks, and seafloor sonar beacon networks. The
comprehensive PNT application system is the integration of
available PNT sensors, including natural PNT information
sensors such as Pulsar signals, gravity field and magnetic
field, and traditional sensors such as inertial navigation
systems (INSs) and atomic clocks. A comprehensive PNT
system features the diversity of PNT physical principles, the
ubiquity of PNT sources and the unity of space-time datum.
The comprehensive PNT system has superiority in over-
coming the limited coverage of a single PNT information
source and improving the availability of PNT service
seamlessly covering from deep space to deep ocean; redu-
cing the interruption risk caused by the breakdown of a
single PNT system, especially the interference and deception
of the GNSS and realizing the continuity of PNT service in
complex situations; compensating for the possible systematic
errors and improving the accuracy of PNT results; fusing
multisource PNT information and improving the reliability
and security of PNT services. The infrastructures of the
comprehensive PNT system are mainly described in Figure
2.
3.1 The deep space PNT infrastructure
The deep space PNT infrastructure mainly consists of a sa-
tellite navigation constellation in deep space. The location
and time information are indispensable to secure navigation
for environment detection and scientific research in deep
space. The PNT service in deep space mainly relies on as-
tronomy navigation, ground-based very long baseline inter-
ferometry (VLBI), and pulsar navigation which may be
available in the future (Shuai et al., 2006;Mao et al., 2009).
A pulsar is a compact object outside the Milky Way, with a
radius of approximately 10 km and a mass 1.44 to 3.2 times
heavier than that of the sun, exhibiting the second largest
density after a black hole. A pulsar has better period stability
which can be used to measure the time difference and the
range between the two points. However, a pulsar is not an
infrastructure. To construct a controllable infrastructure in
deep space, navigation satellite constellations at the Earth-
Moon Lagrange points and Sun-Moon Lagrange points
should be taken into consideration. The deep space con-
stellation should adopt the same signal frequency and mod-
ulation as those of the BDS or be designed as having a
compatible and interoperable signal with the BDS. In addi-
tion, to realize an integrated service together with the BDS,
the Earth-Moon and Sun-Moon Lagrange point navigation
constellations should be equipped with downwards antenna
broadcasting BDS navigation messages to provide better
PNT service for near-Earth users, and upwards antennas
providing PNT service for deep space users together with the
BDS sidelobe signals.
3.2 The deep sea PNT infrastructure
The deep sea infrastructure is mainly composed of seafloor
control stations, namely, seafloor sonar beacon networks.
The construction of sonar beacon networks involves the
development of seafloor shelters, the selection of beacon
locations, the precise positioning strategy of the beacons, and
acoustic observation models and data processing strategies in
different ocean environments (Yang et al., 2017;Yang Y X et
al., 2020a;Yang and Qin, 2021;Qin et al., 2022,2023). In
recent years, Chinese scholars have made great contributions
to underwater PNT research, and deployed a seafloor
acoustic positioning network at depths over 3000 m in the
South China Sea. The precision of the seafloor station is
about 0.3 m by combining circular and cross observation
configurations (Yang Y X et al., 2020a;Zeng et al., 2021). A
centimeter-level positioning precision is achieved by using
the ocean sound speed model and equivalent sound speed
model supported by a neural network learning algorithm
(Xin et al., 2018;Wang et al., 2020a). Resilient acoustic
observation models considering ocean environments have
been established (Yang and Qin, 2021;Qin et al., 2022).
Stochastic models of acoustic observations have also been
built, and a series of underwater navigation and positioning
algorithms have been developed, such as the robust adaptive
Unscented Kalman filter (UKF) (Wang et al., 2020b;Qin et
al., 2023), and the differential positioning algorithm for deep
ocean control stations with depth difference constraints and
horizontal distance constraints (Sun et al., 2019). To achieve
meter level or 10 m level calibration of the INS accumulated
error of underwater carriers, the relayed seafloor acoustic
networks should be built in the future, that is, a set of sonar
beacon networks should be built every hundred nautical
miles to control the accumulated errors of INS, improving
the precise PNT capability for long-term sailing.
3.3 The ground-based PNT infrastructure
As an important component of the national comprehensive
PNT system, the ground-based radio PNT infrastructure
should be strengthened. The ground-based radio PNT infra-
structure refers to the long-range radio navigation system
and the thriving 5G ground-based communication base sta-
tion network, other than the ground-based augmentation
system (GBAS). The reason is that the GBAS only improves
4Yang Y, et al. Sci China Earth Sci
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
the service precision within the coverage of the ground
monitoring stations. It cannot strengthen the observation
geometry or the signal power, not to mention replacing the
GNSS.
Compared with space-based PNT, the ground-based low
and very low frequency PNT signals have better anti-in-
ference and anti-deception capabilities. In the middle of the
last century, long-range radio navigation (Loran) technology
was substantially developed and widely used. The U.S. in-
itiated research on long-range navigation technology and
formed a complete ground-based navigation system with
global coverage. Many similar ground-based radio naviga-
tion systems in the low and very low frequency bands have
been established, such as Omega, Chayka and Alpha
(Fuentes, 1987;Hu, 2018). Loran-C and Chayka transmit
pulse signals at low frequencies and provide regional land-
based remote radio PNT services within a radius of 2000 km.
Omega and Alpha transmit pulse signals at very low fre-
quencies and provide global PNT service with a radius larger
than 5000 km. The ground-based PNT system at low and
very low frequencies strongly supported global navigation
and aviation at that time. With the development of GNSS, the
ground-based radio PNT has gradually lost its superiority,
leading to the shutdown of most systems. Being aware of the
vulnerability of GNSS, developed countries have rebuilt
their ground-based radio PNT systems and taken them as the
backup of GNSS. The U.S. restarted Loran-C, and Russia
used Scorpius to replace Chayka. Scorpius has better cov-
erage and can control the whole network with only one
control station. In addition, it has advantages in the auto
maintenance of transmit signal parameters and restraint of
the residual radio pulses. Compared with the systems in the
middle of the last century, the rebuilt or modified ground-
based radio PNT infrastructure greatly advanced the cover-
age and performance and can meet the PNT requirement of
major infrastructure operation as well as flight route navi-
gation, terminal area navigation, nonprecision approaching,
ship navigation and secure entry into the harbor under low
visibility.
China has built several low frequency navigation stations,
namely the “Changhe” system (Wang et al., 2011). In addi-
tion to the modification of existing longwave navigation
stations, new low and very low frequency station networks
with better distributions should be built in the future to form
a radio PNT system complementary to the BDS with almost
global coverage. However, more efforts should be made to
miniaturize and minimize the power consumption of the low-
frequency and very low frequency PNT terminals.
The ground-based cellular radio communication (espe-
cially 5G) base station network could be used as an important
national PNT infrastructure. The cellular radio base station
network could provide PNT services for users within the
coverage (Liu et al., 2020;Zhang et al., 2022), and at least
can be an important compensation for major domestic PNT
infrastructures.
4. Resilient PNT application mode
Resilient PNT is the key to the flexible integration and usage
Figure 2 Comprehensive PNT infrastructures.
5
Yang Y, et al. Sci China Earth Sci
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
of PNT information. Although comprehensive PNT provides
multisource PNT information, it is necessary to formulate
resilient PNT using strategy, develop resilient PNT terminals
and form the resilient PNT application mode to realize the
secure and reliable use of PNT. Resilient PNT is similar to
Flexible PNT, and adaptive PNT (Yang et al., 2001;Yang
and Gao, 2006). The early adaptive PNT focuses on the
stochastic model of multiple kinds of PNT information to
optimally balance their contributions optimally. Resilient
PNT includes the resilient integration of PNT sensors and the
resilient modification of observation models and dynamic
models, and finally realizes the resilient fusion of PNT in-
formation (Yang and Gao, 2004).
4.1 Resilient PNT integration
Resilient PNT integration is the basis of multisource PNT
information application. Aiming to realize the mutual in-
formation compensation, support and replacement, resilient
PNT needs redundant PNT signal sources; otherwise, the
resilient selection is impossible. Obviously, comprehensive
PNT is the basis of the resilient PNT. In the circumstances
without any interference, deception or shelter, users in deep
space, near space and on the ground could preferentially
integrate radio PNT signals, including the GNSS signals,
space-based and ground-based GNSS augmentation signals
and Lagrange point constellation PNT signals and ground-
based low and very low frequency radio PNT signals as well
as cellular signals. When the radio signals are blocked, in-
terfered with or deceived, autonomous PNT information
such as inertial navigation information (Li et al., 2004), as-
tronomy navigation information (including pulsar signals),
optical navigation information and quantum perception in-
formation (Zou, 2014) can be used as resilient PNT in-
formation sources.
Different integration strategies should be used in different
application scenes (as shown in Figure 3). For example, deep
space users can integrate astronomy navigation information
(including pulsar information), VLBI and GNSS sidelobe
signals. Underwater users can use the sonar information
provided by acoustic networks on the seafloor or buoys
floating on the ocean surface and matching navigation in-
formation such as the Earth’s gravitational field, magnetic
field and the seabed terrain. Indoor users can use GNSS
pseudolite signals, the ultra-wideband (UWB) signals (Pang
et al., 2005), WiFi, Bluetooth (Wang et al., 2011), acoustic,
optical, radar and visual information (El-Sheim and Li,
2021), etc.
In addition, the velocity and acceleration information
measured by inertial navigation sensors (Li et al., 2004;El-
Sheim and Youssef, 2020), quantum perception information
(including quantum clock information, physical field in-
formation from quantum sensors, and quantum inertial in-
formation), and micro clocks are all available information for
resilient PNT.
Obviously, the resilient PNT information integration in-
cludes the information provided by the comprehensive PNT
infrastructure and sensed by various sensors in the nature.
Resilient PNT integration emphasizes microminiaturization,
low power consumption, availability and reliability. Thus,
micro-PNT is the development direction of resilient PNT
(Yang and Li, 2017).
4.2 Resilient PNT function model
A resilient functional model is the basis of resilient PNT
information fusion. In complex environment, the sensitivity
of PNT information is different in different backgrounds.
Resilient functional model modification focuses on the au-
tonomous compensation of various systematic errors through
PNT observation models (Yang, 2018). Usually, the ob-
servation model for a certain kind of PNT observation is
fixed and the correction for systematic error (if there is any)
is applied to simplify the observation model. However, the
resilient PNT functional model modification attaches re-
silient correction items to the frequently-used observation
models with parameters estimated together with the PNT
parameters while fusing multisource information or with the
support of a priori information. Certainly, the non-PNT es-
timated parameters could be eliminated through the reduc-
tion method.
The core of resilient functional model modification is
systematic error compensation. It is known that any ob-
servation model or dynamic model is approximate, and there
is basically no completely accurate functional model. Either
the linearization of the nonlinear model or the reduction of
the dynamic model will result in residual model errors. In
complex urban environments, mountain areas, deep space,
deep sea, and indoor and underground areas, certain sys-
tematic errors exist in all kinds of PNT information that vary
with time. The fusion of multisource PNT information sup-
ports the study of observation systematic errors, because the
sensitivity of observations based on different physical prin-
ciples to the environment is different, as are the error char-
acteristics. These errors could be compensatory, and as long
as appropriate resilient error correction items are set in the
functional model, the systematic error terms can be obtained
according to the data fusion criterion. The deep sea resilient
acoustic observation model is a meaningful experiment
(Yang and Qin, 2021;Qin et al., 2022).
4.3 Resilient PNT stochastic model
The resilient stochastic model refers to the measurement
model of the uncertainty evaluation of each kind of PNT
observation. The uncertainties of various PNT sources are
6Yang Y, et al. Sci China Earth Sci
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
different, especially in different environments. The resilient
PNT stochastic model focuses on the self-adaptiveness of the
actual uncertainty between the stochastic models and the
corresponding observations. Significant achievements have
been made in resilient stochastic models, among which
covariance estimation is a widely used adjusting algorithm
(Yang and Xu, 2003). Factually the adaptive factors in
adaptive Kalman filtering (Yang et al., 2001;Yang and Gao,
2006) are mainly used to adjust the stochastic models of
different PNT information.
The resilient stochastic model is mainly based on the re-
liable functional model. Any bias or outlier in the functional
model will affect the stochastic model, and thus the overall
modification of the functional model and stochastic model
must be solved. To control the effect of outliers on stochastic
models, the stochastic model modification based on the ro-
bust estimation should be a focus in the resilient stochastic
model studies.
The purpose of resilient PNT information integration, re-
silient PNT function model modification and resilient sto-
chastic model adjustment is to realize resilient PNT
information fusion. The study of resilient PNT has just
started, and further research should be conducted on the
theory, algorithm and mode.
5. Intelligent PNT application mode
Intelligent PNT includes intelligent PNT systems and in-
telligent PNT applications. In this section, only the intelligent
PNT application mode is discussed. For both comprehensive
PNT and resilient PNT, the core is to guarantee the security,
continuity, accuracy and reliability of PNT service. Due to the
ever-changing PNT application environment, the resilient
PNT application mode should be intelligent. Therefore, an
intelligent PNT application mode should be developed (Yang
Y X et al., 2021c). Intelligent PNT application includes the
intelligent perception of users, intelligent multisource in-
formation integration and intelligent modification of the
functional model and stochastic model (as shown in Figure 4).
Figure 3 Resilient information integration for typical scenes.
7
Yang Y, et al. Sci China Earth Sci
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
5.1 Intelligent perception of PNT users and intelligent
multisource information integration
Intelligent perception of users is the premise of intelligent
PNT applications. In certain environments, the availability,
reliability and accuracy of PNT information sources are
different. Meanwhile, users have particular requirements for
the efficiency, security and precision of PNT information.
Thus, intelligent PNT application should initially distinguish
the environment and understand the requirements of the
users, then intelligently select and integrate the PNT in-
formation sources, and finally provide the PNT service
matching the user requirements and environment. For ex-
ample, GNSS signals are unavailable to underwater users,
and information integration should be intelligently switched
to INS navigation, acoustic positioning, or magnetic, gravity
or terrain matching navigation modes. Once the user vehicle
surfaces, the ground-based radio navigation/GNSS/INS
mode is a reasonable choice.
It should be noted that intelligent PNT must be resilient
and integrated, while resilient integration may not be in-
telligent. Therefore, resilient integration does not necessarily
contain PNT intelligence. In addition, resilient PNT should
obtain a priori user requirements and background informa-
tion in advance. If the intelligent recognition, perception and
mining of user requirements and the environment are in-
cluded in the resilient PNT integration, the resilient PNT
information integration can be regarded as intelligent.
5.2 PNT functional model and intelligent modification
of the stochastic model
An intelligent PNT model is the basis for improving the
intelligence of PNT applications. Intelligent learning could
modify the PNT observation model and stochastic model in
real-time or near real-time, making the observation model
and stochastic model adapt to the observation environment
and the observation uncertainty, respectively. Any model
improvement with a priori information belongs to supervised
learning, and that without a priori information belongs to
unsupervised learning. For example, the determination of
GNSS multipath signals and non-line-of-sight signals based
on the supervised learning method can improve GNSS ob-
servation models and control the effect of multipath errors
(Zhu et al., 2021). Model improvement is an intelligent
learning belongs to the intelligent PNT model modification.
It is noted that the intelligent modification of functional
model includes the resilient modification of the functional
model, but the resilient modification does not have to be
intelligent. Similarly, the intelligent adjustment of the sto-
chastic model includes the resilient adjustment of stochastic
model, but the resilient adjustment does not have to be in-
telligent. Intelligent modification will be achieved if in-
telligent elements are integrated into the resilient
modification of the function model and stochastic model.
Intelligent PNT model modification is the basis of the
intelligent fusion of multisource PNT information. Multi-
source PNT fusion is not a simple weighted average, but
intelligent fusion comprehensively balances the reliability
(including continuity), accuracy, and uncertainty of each
kind of PNT information. Without intelligent learning and
determination of the PNT observation model errors, a reli-
able and modified observation model cannot be built; with-
out intelligent tracking and evaluation of the uncertainty of
the PNT information, the intelligent adjustment of PNT
stochastic model cannot be realized; without an intelligent
PNT functional model and stochastic model, intelligent fu-
sion of multiple PNT sources will not be achieved, not to
mention the intelligent PNT applications. In most PNT ap-
plications, the systematic biases of PNT observation models
can be obtained through intelligent learning and then com-
pensated by the intelligent functional model. Similarly, the
suitability and biases of each kind of PNT stochastic model
can be sensed to support the optimization of PNT stochastic
model (Yang Y X et al., 2021c).
6. Conclusion
Now and in the future, GNSS is the core of PNT systems with
Figure 4 Key factors of intelligent PNT.
8Yang Y, et al. Sci China Earth Sci
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
unparalleled coverage, performance, popularity and flex-
ibility. However, PNT signals are vulnerable in terms of the
signal intensity, signal penetration, anti-interference and
anti-spoofing capbility. To improve the security, robustness
and coverage, a seamless PNT system covering from deep
space to deep ocean, that is, a comprehensive PNT infra-
structure should be built first, and then the corresponding
resilient and intelligent PNT application modes must be
followed.
(1) As the core of the PNT service system, the BDS ISLs
must be used to improve the capability of space-time datum
maintenance based on the BDS. Furthermore, the BDS
GEOs and IGSOs should be fully used to improve the BDS
satellite-based augmentation and satellite-based precise
point positioning. Meanwhile, the LEO constellation should
be constructed to improve the anti-inference and anti-de-
ception of BDS PNT service.
(2) As the core of the secure operation of major infra-
structures, a national PNT system with powerful functions
and rich information sources must be constructed. A deep
space navigation constellation and deep ocean acoustic net-
work should be constructed to provide PNT services for deep
space and deep sea users, respectively. Ground-based low
and very low frequency longwave PNT stations should be
constructed as important backups of the BDS. Ground-based
5G stations could be used as a backup PNT system for major
domestic infrastructures.
(3) With the support of the comprehensive PNT infra-
structure and other autonomous PNT perception information,
a resilient PNT application mode adaptive to various com-
plex environments should be constructed, as well as a PNT
application system with solid theoretical basis, high avail-
ability, high continuity, high stability, high reliability and
high security from the aspects of resilient PNT sensor in-
tegration, resilient functional model, resilient stochastic
model and resilient data fusion.
(4) The intelligent perception of PNT users, intelligent
modification of the functional model and stochastic model
and intelligent data fusion are the preconditions for in-
telligent PNT applications. It is certain that the intelligent
PNT service system and application mode will be an im-
portant development direction in the near future.
(5) With the BDS as the core, the secure PNT system is the
inevitable development trend of the PNT system, and com-
prehensive PNT, resilient PNT and intelligent PNT are im-
portant ways to realize secure PNT. Comprehensive PNT
provides all-domain, seamless and multiple information
sources, resilient PNT provides the resilient application
mode for users, and intelligent PNT is the further whose core
is to apply knowledge learning, intelligent perception and
intelligent modification to every step of PNT application.
Acknowledgements This work was supported by the Key Program of
National Natural Science Foundation of China (Grant No. 41931076), the
Laoshan Laboratory(Grant No. LSKJ202205101), the National Natural
Science Foundation of China for Young Scholar (Grant No. 41904042), and
the National Key Research and Development Program of China (Grant No.
2020YFB0505800).
References
Aresta C. 2017. Resilience of the PNT Systems: A Portuguese case study
[EB/OL]. [2021-07-28]. https://comum. rcaap.pt/bitstream/10400.26/
21053/1/ASPOF%20Catarina%20Matos%20Aresta%20-%20Resilience
%20of%20the%20PNT%20systems%20-%20A%20portugueses%
20case%20study.pdf
Dalal M. 2012. Low noise, low power interface circuits and systems for
high frequency resonant Micro-Gyroscopes. Doctoral Dissertation.
Atlanta: Georgia Institute of Technology
El-Sheimy N, Youssef A. 2020. Inertial sensors technologies for navigation
applications: State of the art and future trends. Satell Navig, 1: 9
El-Sheimy N, Li Y. 2021. Indoor navigation: State of the art and future
trends. Satell Navig, 2: 88–110
Executive Office of the President. Strengthening national resilience through
responsible use of positioning, navigation, and timing services [EB/
OL]. (2020-02-18). [2022-07-07] https://www.federalregister.gov/
documents/2020/02/18/2020-03337/strengthening-national-resilience-
through-responsible-use-of-positioning-navigation-and-timing
Fuentes A F. 1987. LORAN-C in the 21st century. IEEE Aerospace and
Electronic Systems Magazine, 2: 8–10
Hu A P. 2018. Research on the development of land-based ultra-long-range
radio navigation (in Chinese). Navigation Positioning and Timing, 5: 1–6
Li R, Zheng S Y, Wang E, Chen J P, Feng S J, Wang D, Dai L. 2020.
Advances in BeiDou Navigation Satellite System (BDS) and satellite
navigation augmentation technologies. Satell Navig, 1: 126–148
Li R B, Liu J Y, Zeng Q H, Hua B. 2004. Evaluation of MEMs based micro
inertial navigation system (in Chinese). J Chin Inert Technol, 12: 88–94
Liu J N, Gao K F, Guo W F, Cui J S, Guo C. 2020. Role, path, and vision of
“5G + BDS/GNSS”. Satell Navig, 1: 23
Mao Y, Song X Y, Feng L P. 2009. Visibility analysis of X-ray pulsar
navigation (in Chinese). Geomat Inform Sci Wuhan Univ, 34: 222–225
Mcneff J. 2010. Changing the game changer, The way ahead for military
PNT. (2010-10-25). [2022-6-20]. https://insidegnss.com/military-pnt-
the-way-ahead
Pang Y, Zhang L J, Chen C J. 2005. An improved algorithm for UWB
precision positioning based on time averaging (in Chinese). J Beijing
Jiaotong Univ, 29: 60–63
Parkinson B. 2015. Assured PNT strengths and synergies [EB/OL]. [2022-
01-31]. https://www.gps.gov/governance/advisory/meetings/2015-06/
parkinson2.pdf
Parkinson B. 2017. Assuring PNT for all [EB/OL]. [2022-01-31]. https://
www.gps.gov/governance/advisory/meetings/2017-11/parkinson.pdf
Qin X, Yang Y, Sun B. 2022. The refined resilient model for underwater
acoustic positioning. Ocean Eng, 266: 112795
Qin X, Yang Y, Sun B. 2023. A robust method to estimate the coordinates
of seafloor stations by direct-path ranging. Mar Geodesy, 46: 83–98
Ren X, Yang Y, Zhu J, Xu T. 2017. Orbit determination of the Next-
Generation Beidou satellites with intersatellite link measurements and a
priori orbit constraints. Adv Space Res, 60: 2155–2165
Ren X, Yang Y, Zhu J, Xu T. 2019. Comparing satellite orbit determination
by batch processing and extended Kalman filtering using inter-satellite
link measurements of the next-generation BeiDou satellites. GPS Solut,
23: 25
Scholz A. 2020. Resilient PNT system concepts for critical infrastructure
[EB/OL]. [2021-07-28]. https://www.gps.gov/cgsic/meetings/2020/
scholz.pdf
Shuai P, Chen S L, Wu Y F, Zhang C P, Li P. 2006. X-ray pulsar navigation
technology and the development (in Chinese). Aerospace China, 10:
27–32
9
Yang Y, et al. Sci China Earth Sci
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
Sun W Z, Yin X D, Zeng A M, Bao J Y. 2019. Differential positioning
algorithm for deep-sea control points on constraint of depth difference
and horizontal distance constraint (in Chinese). Acta Geodaet Cartogra
Sin, 48: 1190–1196
The World Radiocommunication Conference Resolution 609. 2007. Pro-
tection of aeronautical radionavigation service systems from the
equivalent power flux-density produced by radionavigation-satellite
service networks and systems in the 1164-1215MHz frequency band.
[2020-06-20]. https://www.itu.int/en/ITU-R/space/Res609%20CM%
20Documents/RES-609_e.pdf
U.S. Department of Transportation. What is positioning, navigation and
timing (PNT)? [EB/OL]. (2017-06-13). [2020-06-20]. https://www.
transportation.gov/pnt/what-positioning-navigation-and-timing-pnt
U.S. Department of Transportation and Department of Defense. 2010.
National positioning, navigation, and timing architecture implementa-
tion plan. (2010-07-28). [2020-06-20]. https://rosap.ntl.bts.gov/view/
dot/18293
U.S. Senate. 2015. National positioning, navigation, and timing resilience
and security act of 2015 [EB/OL]. (2017-12-12). [2020-06-20]. https://
www.congress.gov/bill/114th-congress/house-bill/1678/text?r=6&s=5
U.S. Senate. 2017. National timing resilience and security act of 2017[EB/
OL]. (2017-12-12). [2020-06-20]. https://www.congress.gov/bill/115th-
congress/senate-bill/2220/text?q=%7B%22search%22%3A%22timing
+resilience%22%7D&r=2&s=6
U.S. Senate. 2018. National timing resilience and security act of 2018[EB/
OL]. (2018-11-4). [2020-06-20]. https://rntfnd.org/wp-content/uploads/
National-Timing-Security-and-Resilience-Act-of-2018.pdf
Wang J, Xu T, Nie W, Yu X. 2020a. The construction of sound speed field
based on back propagation neural network in the global ocean. Mar
Geodesy, 43: 621–642
Wang J, Xu T, Wang Z. 2020b. Adaptive robust unscented Kalman filter for
AUV acoustic navigation. Sensors, 20: 60
Wang R, Zhao F, Peng J H, Luo H Y, Lu B, Lu T. 2011. Combination of Wi-
Fi and Bluetooth for indoor localization (in Chinese). J Comput Res
Develop, 48(Suppl): 28–33
Wang Z, Yan J H, Zhang H Y. 2011. Changhe 2 navigation system and its
technolighy update (in Chinese). Digital Commun World, 78: 86–87
Xin M, Yang F, Wang F, Shi B, Zhang K, Liu H. 2018. A TOA/AOA
underwater acoustic positioning system based on the equivalent sound
speed. J Navigation, 71: 1431–1440
Xin M Z, Yang F L, Xue S Q, Wang Z J, Han Y F. 2020. A constant
gradient sound ray tracing underwater positioning algorithm consider-
ing incident beam angle (in Chinese). Acta Geodaet Cartograph Sin, 49:
1535–1542
Yang Y F, Yang Y X, Hu X, Tang C, Guo R, Zhou S, Xu J, Pan J, Su M.
2021. BeiDou-3 broadcast clock estimation by integration of observa-
tions of regional tracking stations and inter-satellite links. GPS Solut,
25: 57
Yang Y F, Yang Y X, Xu JY, Xu Y Y, Zhao A. 2020. Navigation satellites
orbit determination with the enhancement of low earth orbit satellites
(in Chinese). Geomat Inform Sci Wuhan Univ, 45: 46–52
Yang Y X. 2016. Concepts of comprehensive PNT and related key tech-
nologies (in Chinese). Acta Geodaet Cartograph Sin, 45: 505–510
Yang Y X. 2018. Resilient PNT concept frame (in Chinese). Acta Geodaet
Cartograph Sin, 47: 893–898
Yang Y X, Ding Q, Gao W G, Li J L, Xu Y Y, Sun B J. 2022. Principle and
performance of BDSBAS and PPP-B2b of BDS-3. Satell Navig, 3: 1–9
Yang Y X, Gao W G. 2004. Integrated navigation based on robust esti-
mation outputs of multi-sensor measurements and adaptive weights of
dynamic model information (in Chinese). Geomat Inform Sci Wuhan
Univ, 29: 885–888
Yang Y X, Gao W G. 2006. An optimal adaptive Kalman filter. J Geodesy,
80: 177–183
Yang Y X, Guo H R, He H B. 2021b. Principle of satellite navigation and
positioning (in Chinese). Beijing: National Defense Industry Press
Yang Y, He H, Xu G. 2001. Adaptively robust filtering for kinematic
geodetic positioning. J Geodesy, 75: 109–116
Yang Y X, Li X Y. 2017. Micro-PNT and comprehensive PNT (in Chinese).
Acta Geodaet Cartograph Sin, 46: 1249–1254
Yang Y X, Liu L, Li J L, Yang Y F, Zhang T Q, Mao Y, Sun B J, Ren X.
2021a. Featured services and performance of BDS-3. Chin Sci Bull, 66:
2135–2143
Yang Y X, Liu Y X, Sun D J, Xu T, Xue S Q, Han Y F, Zeng A M. 2020a.
Seafloor geodetic network establishment and key technologies. Sci
China Earth Sci, 63: 1188–1198
Yang Y X, Mao Y, Sun B J. 2020b. Basic performance and future devel-
opments of BeiDou global navigation satellite system. Satell Navig, 1:
1–8
Yang Y X, Qin X P. 2021. Resilient observation models for seafloor
geodetic positioning. J Geod, 95: 79
Yang Y X, Ren X. 2018. Maintenance of space datum for autonomous
satellite navigation (in Chinese). Geomat Inform Sci Wuhan Univ, 43:
1780–1787
Yang Y X, Xu J Y. 2016. Navigation performance of BeiDou in polar area
(in Chinese). Geomat Inform Sci Wuhan Univ, 41: 15–20
Yang Y X, Xu T H. 2003. An adaptive Kalman filter combining variance
component estimation with covariance matrix estimation based on
moving window (in Chinese). Geomat Inform Sci Wuhan Univ, 28:
714–718
Yang Y X, Xu T H, Xue S Q. 2017. Progresses and prospects in developing
marine geodetic datum and marine navigation of China (in Chinese).
Acta Geodaet Cartograph Sin, 46: 1–8
Yang Y X, Xu Y Y, Li J L Y C. 2018. Progress and performance evaluation
of BeiDou global navigation satellite system: Data analysis based on
BDS-3 demonstration system. Sci China Earth Sci, 61: 614–624
Yang Y X, Yang C, Ren X. 2021c. PNT intelligent services (in Chinese).
Acta Geodaet Cartograph Sin, 50: 1006–1012
Zeng A M, Yang Y X, Ming F, Ma Y Y. 2021. Positioning model and
analysis of the sailing circle mode of seafloor geodetic datum points (in
Chinese). Acta Geodaet Cartograph Sin, 50: 939–952
Zhang W, Yang Y X, Zeng A M, Xu Y Y. 2022. A GNSS/5G integrated
three-dimensional positioning scheme based on D2D communication.
Remote Sens, 14: 1517–1536
Zhang X H, Ma F J. 2019. Review of the development of LEO navigation-
augmented GNSS (in Chinese). Acta Geodaet Cartograph Sin, 48:
1073–1087
Zhao Q L, Wang C, Guo J, Yang G L, Liao M, Ma H Y, Liu J N. 2017.
Enhanced orbit determination for BeiDou satellites with FengYun-3C
onboard GNSS data. GPS Solut, 21: 1179–1190
Zhen W M, Ding C C. 2019. Development status and trend of land-based
radio navigation system (in Chinese). GNSS World China, 44: 10–15
Zhu B, Yang C, Liu Y. 2021. Analysis and comparison of three un-
supervised learning clustering methods for GNSS multipath signals (in
Chinese). Acta Geodaet Cartograph Sin, 50: 1762–1771
Zou H X. 2014. The inertial navigation technology of next generation—
Quantum navigation (in Chinese). Nat Defense Sci Tech, 35: 19–24
(Responsible editor: Xiong XIONG)
10 Yang Y, et al. Sci China Earth Sci
https://engine.scichina.com/doi/10.1007/s11430-022-1069-7
