bitpro가 원문에서 글자만 모은 내용이에요. 표·그림·수식은 빠지거나 깨질 수 있어요. 전체 3쪽 중 2쪽
ed for POA), a tvalidation value
of 20 ms is initially proposed and can be changed as an algorithm parameter by
the GeoDAO. The validator nodes are selected to be proximal to the miners, which
can be achieved by geographic location or by simply minimizing communication
latency (i.e., “ping”) times.
Note that the validated location check does not validate the physical location of
the miner node, but rather the location of its antenna; thus, a data retransmission
is considered valid for the original antenna location and does not negatively affect
the quality of the network, as long as the latency is not excessive.
The validation message checks are combined with differential pseudorange and
carrier phase analysis as part of the POL protocol. This approach limits observation
errors if a sophisticated attack based on the security code and estimation replay is
attempted (Psiaki et al., 2016). While this class of attack can be successfully per -
formed with latencies < 1 ms using custom hardware/software, the costs will out -
weigh the mining rewards. Any attacker generating invalid data for any reason will
be detected based on differential analysis with nearby miner nodes, performed by
validation nodes, including spoofing attacks against the network.
5.4 Proof of Accuracy (POA)
Validated GNSS miners are rewarded in proportion to the following formula:
Q = BWMNTSL
Q: POA metric
B: Base reward set by the token supply generation rate
W: 0–1.0 discount factor accounting for sharing rewards with neighbors within a
20-km square, with heavier weighting for earlier miners
M: Reward multiplier based on the number of independent bands observed divided
by 6 (such that M = 1 for 6 tracked bands)
N: 0–1.0 linear scaling factor corresponding to the bounded range of 10–30 satel -
lites with C/N0 > 40 dB-Hz
T: 0–1.0 exponential scaling factor corresponding to the bounded range of 50%–
100% uptime
FIGURE 3 POL validation messages
HORTON et al.
S: 0–1.0 linear scaling factor corresponding to the bounded range of 2-10cm of esti-
mated position shift
L: 0–1.0 linear scaling factor corresponding to a bounded range of 1-5cm of
latency
These parameters are computed by a validator node, which is described later,
with the results stored publicly on the GEODNET blockchain. The parameters may
be updated from time to time as the network grows, and additional metrics may
be added to the POA rubric. Miners within 10 m of each other are treated as a
single miner and are rewarded as such. Incentive programs may increase the base
rewards in a specific region to attract miners to low-density areas.
POA is actively used to reward GEODNET miners. Deposit transactions occur
daily and can be viewed at the GEODNET Polygon network SmartContract address:
0xAC0F66379A6d7801D7726d5a943356A172549Adb.
Finally, GEODNET uses a PPP processing pipeline to validate and maintain each
miner’s coordinates with third-party services such as NRCAN as well as internally
developed tools.
5.5 Genesis Miners
To bootstrap the global development of the GEODNET network, a global net -
work of more than 40 survey-grade GNSS stations has been installed, with a tar -
get of 100 stations prior to the end of 2022. Figure 4 shows the installation as of
September 2021. The locations of these genesis miner stations are evenly spread
over the surface of the Earth, providing a baseline network capability to produce
the data products necessary for PPP and PPP-AR algorithms. PPP’s global capabil-
ity bootstraps the functionality of the network at a global level.
6 NETWORK ARCHITECTURE
The GEODNET network consists of space weather miners, validator nodes,
and service provider nodes, as shown in Figure 5. Service provider nodes face the
end-user and utilize the blockchain to select which miner stream or combination
of streams to utilize.
Validators are internet-connected instances that are staked with the GEOD
token and participate in a POS consensus protocol to validate GEODNET block -
chain transactions. In the initial phase, the number of validator nodes will be
limited to 64, with a minimum of 20 to launch the main net. When a new space
FIGURE 4 Genesis miners
HORTON et al.
weather miner first connects, it will communicate with a validator boot node. The
validator boot node will assign the miner to a primary validator. If the assigned
validator leaves the network, the process repeats. After the validator confirms
the miner’s security certificate, the miner begins signed data transmission and
the initial validation process. After the initial validation is complete, the valida -
tor continues to receive POL validation messages and requests observation data
from the miner at random intervals to compute the POA. The consensus protocol
determines which node generates a new block at a particular time point. If a
miner provides fake or inaccurate data, it will be detected, and the proof will be
included in the block.
6.1 Proof of Stake (POS)
POS is a consensus protocol that has been widely used in recent blockchain proj-
ects as an alternative to POW. POS is more energy efficient than POW and can
provide improved security. Token owners put up tokens as collateral in exchange
for the opportunity to participate in blockchain validation operations, for which
additional tokens may be earned as rewards. In GEODNET, validator nodes stake
tokens to participate in system consensus, which is required to add new blocks to
the blockchain. Each validator node will have an opportunity to produce blocks
based on the Boneh–Lynn–Shacham (BLS) signature verification algorithm,
weighted by the staked token proportion.
GEODNET uses the BLS algorithm for fast and reliable Byzantine-fault-tolerant-like
consensus (Sousa et al., 2018). Multiple parties use a threshold signature to reach
an agreement on a blockchain proposal. The staking will also improve economic
security by punishing any misbehaving validators through a loss of staked tokens.
Anyone who owns GEODNET tokens can bond (or delegate) their GEOD tokens
and become a validator, making the validator set open and permissionless.
6.2 Service Provider Node Software Stack
The service provider runs a GeoDAO-provided stack written for the node, includ-
ing several libraries and application programming interfaces (APIs) to interact with
the blockchain and receive real-time data streams from miners. These libraries and
APIs are extendable by the operator of a service provider node and include hooks
to connect code written in other programming languages. As of September 2022,
this stack is centralized and run by the GeoDAO Foundation. During 2023, the
stack is planned to be decentralized via validator nodes, with significant portions
FIGURE 5 GEODNET architecture: Miners, validators, and service providers
HORTON et al.
open-sourced. The service provider nodes may output a variety of data products
including the following:
• Reliable daily and hourly receiver independent exchange format (RINEX) files
for select stations
• Network RTK corrections (RTCM) for real-time RTK via NTRIP
• Precise clock, orbit, and localized ionospheric corrections for PPP-AR via
NTRIP
Testing of proposed network data products has already begun, and current per -
formance results were presented by the GeoDAO Foundation at the ION GNSS+
2022 conference (Horton et al., 2022). Presented data included the following:
• Comparison of GPS orbit accuracy with IGS data, where data were processed
with Geoscience Australia GINAN software using exclusively GEODNET
station data
• Local RTK results between various neighboring GEODNET stations processed
with RTKlib 2.4.3
• Accuracy results obtained by NRCAN and TEQC analysis of station quality
While the GEODNET network has not yet entered full operational function,
early results show that its performance is consistent with that of traditional cen -
tralized networks.
6.3 Validator Node Software Stack
The validator will run an open-source GeoDAO-maintained stack that will fully
decentralize the operation of the blockchain. The GEODNET blockchain provides
benefits over existing POW and POS protocols alone.
Most POW blockchain systems only offer probabilistic finality. Probabilistic
finality refers to the fact that, theoretically, some nodes with overwhelmingly high
computation power can build an alternative branch and eventually replace the
original branch. Additionally, POW consumes a high level of energy to perform a
brute-force search for the required hash function input (Xiao et al., 2019).
POS is a consensus protocol that uses stakes as the criteria to vote on new block
generation. Unlike POW, POS is a closed system, which means tokens are created
without any external entropy. Thus, the POS system will likely suffer from cen -
tralization in the initial token distribution and reward system. Another drawback
of the POS consensus protocol is that the POS system itself can be easily cloned,
resulting in an unlimited token supply (Nguyen et al., 2019).
GEODNET takes an approach that merges the desired features of POW and POS.
External resource consumption (external entropy) is critical to the system’s security
and decentralization. However, instead of using unbounded energy consumption
as in Bitcoin or most other POW protocols, GEODNET uses geospatial information
as the unique and unforgeable resource to include in the consensus process. The
GEODNET consensus protocol uses both geospatial data input (POL, POA) and
staked tokens (POS) to efficiently determine block generation.
In this way, GEODNET will be the first and only network to utilize limited Earth
surface as a resource, which provides enormous value for the network. Additional
services such as localization, high-definition mapping, and autonomous naviga -
tion are simply natural derivatives of this network.
HORTON et al.
7 GEODAO AND CURRENT STATUS
The GeoDAO Foundation is the DAO that governs the GEODNET network.
7.1 Organization
A DAO is best described as an effective and safe way to operate a project with
a set of individuals around the globe. The GeoDAO Foundation is an organiza -
tion dedicated to a decentralized blockchain-based network of environmental and
space weather collection stations for utilization by the geospatial community and
industry.
GeoDAO is organized around an ERC20 SmartContract (Ethereum, 2023), a
piece of code that describes the creation and distribution of native GEOD tokens.
The GEODNET SmartContract is registered on the Polygon public blockchain, peg-
ging the GeoDAO and the token distribution logic into a well-established public
chain that is currently used by companies such as Starbucks. As the GEODNET
software stack matures, operations will be transitioned onto a separate blockchain
to more fully decentralize the network.
Decisions regarding the GeoDAO can be executed by the blockchain
SmartContracts, based on token holder “votes.” The GEODNET Polygon
SmartContract address is 0xAC0F66379A6d7801D7726d5a943356A172549Adb.
7.2 Token Supply
The total token supply growth rate halves every two years. The dependence of
the token supply over time t can be written as follows:
M(t) = M0 + ∫ I0 ∗ exp[−ln2 ∗ (t/T)]dt + ∫ (m(t) + Δ(t))dt
Here, M0 is the initial token supply. The first integration term is the staking
reward in terms of time t, and the second integration term is the mining reward.
The number of base stations as a function of time is m(t), and Δ(t) is the expected
miner growth.
7.3 Software Development
GeoDAO developers will deliver a baseline set of open infrastructure necessary
to bootstrap the community. These deliverables include initial miner reference
designs, validator and service provider reference code, and user-facing tools.
User-facing tools include the GEODNET website, a user-facing web console
application, and a mobile GEOD wallet application with miner registration fea -
tures. GeoDAO will contribute other tools as the need arises.
7.4 Current Status
As of January 2023, i.e., less than one year since the launch of GEODNET, the
network has grown to over 1700 deployed stations, and approximately 2500 sta -
tions have been delivered for installation. By the end of 2022, it is anticipated
HORTON et al.
that there will be three certified GNSS manufacturers producing GEODNET
mining hardware. Community interest in this network has been consistent and
growing since the original presentation at the ION GNSS+ 2021 conference
(Horton et al., 2021). Real-world testing of the GNSS correction network for
agriculture machine control and civil survey is ongoing at several locations. A
real-time deployment map of GEODNET CORS stations can be found at https://
console.geodnet.com , and a snapshot image is shown in Figure 6. Recent test
results from GEODNET were presented at the ION GNSS+ 2022 conference
(Horton et al., 2022).
8 APPLICATIONS
A public, secure, dense global reference station network will transform numer -
ous scientific and commercial applications.
8.1 Climate Monitoring
High-accuracy geodetic position data from dense and diverse global locations
help scientists predict the effects of rising sea levels and changing precipitation
patterns. Monitoring of shoreline changes, forest fire threats, desertification, and
air quality have all utilized data from existing CORS networks (Yao et al., 2017;
Goncalves et al., 2017; Knight et al., 2020).
8.2 GNSS Spoofing and Jamming Detection
The European GNSS Agency estimates that the global installed base of GNSS
receivers is 6.5B units; however, commercial GNSS receivers are easily spoofed
and jammed. In the United States, 14 of 16 critical infrastructures depend on GPS
(USTTI Report, 2012). Current consumer receivers provide little to no protection
against well-documented attacks utilizing easily acquired low-cost hardware. A
dense network of secure GNSS receivers can provide an important alert and warn-
ing capability against localized GNSS attacks (GSA, 2019).
FIGURE 6 GEODNET station map, January 2023
HORTON et al.
8.3 Geodeformation Hazard Warning
According to the EU Commission on Science, more than one in three people live
in areas exposed to earthquakes. While GEODNET cannot prevent earthquakes or
other natural geohazards such as sink-holes and landslides, accurate real-time data
help in predicting disasters and identifying at-risk infrastructure (Lin et al., 2021;
Pesaresi et al., 2019).
8.4 Precise Positioning and Timing
Many of the existing networks of GNSS receivers are maintained to enable
precise positioning and timing services on rover devices. Trimble and numer -
ous other companies have supplied receivers used for survey, farming, and
other precise geographic information system applications for more than 20
years. Both the rover receiver and GNSS data correction services for these appli -
cations have traditionally been quite expensive; however, a great deal of enthu -
siasm has recently emerged for lower-cost rovers for autonomous automobiles,
drones, and even mobile phone applications. These applications are driving a
growing need for low-cost highly available correction streams. A large decen -
tralized network can meet this need with secured global data at a fraction of
the cost of alternatives.
9 NEXT STEPS AND ROADMAP
The GEODNET architecture was first introduced at the ION GNSS+ 2021 con -
ference (Horton et al., 2021), and the initial response has been favorable. The
GEODNET network is growing, with additional stations coming online every day.
The goals for initial public operational service in February 2023, fully qualified
data products by the end of 2023, and 100,000 online stations within three years
are highlighted in Table 1.
GEODNET ( https://geodnet.com) regularly posts new information regarding
project status and provides links to additional resources.
TABLE 1
Completed and Anticipated GEODNET Milestones
Date
Milestones
Event Description Miners
2022.1 Testnet launch Initial token distribution and test net running 100
2022.6 Mainnet
GeoDAO created,
SmartContract issued, token rewards
initiated, private tests of GEODNET data
products started
500
2023.2 Service provider launch GEODNET data products publicly available 2,500
2024.1 Mainnet 2 GEODNET fully operational 30,000
2025.1 Global RTK available RTK on-demand globally, largest single network worldwide 100,000
HORTON et al.
10 CONCLUSION
Decentralized infrastructure promises a scalable and rewarding way to enable
high-precision measurements at Earth scale. GEODNET has proposed and is
currently implementing an approach for building, running, and maintaining
this decentralized infrastructure by utilizing ideas and technologies successfully
demonstrated by innovative projects in the blockchain space over the past decade.
Compared with the current sources for GNSS measurement data, GEODNET will
allow participation in this valuable area by many more people (both as providers
and users), with a relative benefit over the current state, particularly in developing
and underserved areas. The GEODNET team has been excited to see the interest in
and early growth of the project to date and looks forward to continuing to grow and
improve the network with a global community.
references
Estey, H. L., & Meertens, C. M. (1999). TEQC: The multi-purpose toolkit for GPS/GLONASS data.
GPS Solutions, 3(1), 42–49. https://doi.org/10.1007/PL00012778
Ethereum Foundation. (2023). ERC-20 Token Standard. https://ethereum.org/en/developers/
docs/standards/tokens/erc-20/
Gakstatter, E. (2014). Finally, a list of public RTK base stations in the U.S. GPS World. https://www.
gpsworld.com/finally-a-list-of-public-rtk-base-stations-in-the-u-s/
Goncalves, R., & Wange, J. (2017). Three most widely used GNSS-based shoreline monitoring
methods to support integrated coastal zone management policies. Journal of Surveying
Engineering, 143(3), 05017003. https://doi.org/10.1061/(ASCE)SU.1943-5428.0000219
GSA. (2019). GNSS market report (issue 6). https://www.euspa.europa.eu/gnss-market-report-
issue-6-october-2019
Haleem, A., Allen, A., Thompson, A., Nijdam, M., & Garg, R. (2018). Helium: A decentralized
wireless network. Helium Systems, Inc. http://whitepaper.helium.com/
Horton, M., Wen, X., & Yi, Y. (2021). Blockchain for CORS networks. Proc. of the 34th International
Technical Meeting of the Satellite Division of the Institute of Navigation (ION GNSS+ 2021), St.
Louis, MO. 290–310. https://doi.org/10.33012/2021.17882
Horton, M., Chen, D., & Yi, Y. (2022). GEODNET: Secure geo-spatial infrastructure & service
for business and metaverse. Proc. of the 35th International Technical Meeting of the Satellite
Division of the Institute of Navigation (ION GNSS+ 2022), Denver, CO. https://www.ion.org/
gnss/virtual-abstract-view.cfm?paperID=11439
Jakowski, N., Mayer, C., Borries, C., & Wilken, V. (2009). Space weather monitoring by ground
and space based GNSS measurements. Proc. of the 2009 International Technical Meeting of the
Institute of Navigation , Anaheim, CA. 729–735. https://www.ion.org/publications/abstract.
cfm?articleID=8353
Knight, P., Plater, A., Bird, C., & Sinclair, A. (2020). A low-cost GNSS buoy platform for
measuring coastal sea levels. Ocean Engineering, 203, 107198. https://doi.org/10.1016/j.
oceaneng.2020.107198
Lin, J., Melgar, D., & Searcy, T. (2021). Early warning for great earthquakes from characterization
of crustal deformation patterns with deep learning. JGR Solid Earth, 126(10), e2021JB022703.
https://doi.org/10.1029/2021JB022703
Nguyen, C., Hoang, D., Nguyen, D., Niyato, D., Nguyen, H., & Dutkiewice, E. (2019). Proof-of-
stake consensus mechanisms for future blockchain networks: fundamentals, applications and
opportunities. IEEE Access, 7, 85727–85745. https://doi.org/10.1109/ACCESS.2019.2925010
Pesaresi, M., Ehrlich, D., Kemper, T., Siragusa, A., Florc프로젝트로 돌아가기