Prerpints.org logo
Preprint
Article

A Blockchain Based Framework for Remote Sensing Data Management

Altmetrics

Downloads

185

Views

80

Comments

0

A peer-reviewed article of this preprint also exists.

This version is not peer-reviewed

Submitted:

01 August 2023

Posted:

02 August 2023

You are already at the latest version

Alerts
Abstract
A large amount of raw data collected by satellites is processed by the production chain to obtain a large amount of product data, of which the secure exchange and storage is of interest to researchers in the field of remote sensing information science. Authentic, secure data is a critical foundation for data analysis and decision-making. And traditional centralized cloud computing systems are vulnerable to attacks, and once the central server is successfully attacked, all data will be lost. Distributed Ledger Technology (DLT) is an innovative computer technology that can ensure information security, traceability and tamper-proof, and can be applied to the field of remote sensing. Although there are many advantages to using DLT in remote sensing applications, there are some obstacles and limitations to its application. Remote sensing data has the characteristics of large data volume, spatiotemporal nature, global and so on, and it is difficult to store and interconnect remote sensing data in the blockchain. To address these issues, this paper proposes a trustworthy and decentralized system using blockchain. The novelty of this paper is to propose a multi-level blockchain architecture in which the system collects remote sensing data and stores it in the Interplanetary File System (IPFS) network, after generating the IPFS hash, the network rehashes the value again and uploads it on the Ethereum chain for public query. Distributed data storage improves data security, supports the secure exchange of information, and improves the efficiency of data management.
Keywords: 
Subject: Computer Science and Mathematics  -   Artificial Intelligence and Machine Learning

1. Introduction

The application field and direction of remote sensing are more and more extensive. With the development of remote sensing technology and the rich popularization of remote sensing data, remote sensing has developed from a scientific research technology to global and popular technology. In 2014, the National Integrated Earth Observation Data Sharing Platform, the core technical facility of China's Global Earth Observation System of Systems (GEOSS), realized the interconnection and unified data service of 8 satellite centers for the first time [1]. This indicates that the earth observation data sharing in China has entered the stage of social sharing.
In the traditional satellite remote sensing industry chain, multiple upstream and downstream data centers often form independent "data islands" because they cannot trust each other, which leads to a low degree of trust in data sharing. At the same time, more and more massive data collected by proximity sensors (Drones, Internet of Things sensor networks, open government data, etc.) are being used to assist in more real-time and accurate earth observation. However, the providers of such data are often the owners of imperfectly trusted research institutions or start-up companies, and may even be the general public. So, the use of such data could also be damaging. Therefore, it is urgent to solve the problems of mutual trust, data authenticity, and confidentiality among users and institutions in the remote sensing industry chain in remote sensing image data distribution and storage, to strengthen the sharing and interconnection of remote sensing data.
Data security sharing is the first major challenge of collaborative network systems. The main problem is that not all participants of the system are willing to share their raw data. To promote earth science data sharing, many countries and organizations have made exploration and attempts. Early through privately negotiated to realize data sharing, remote sensing data at present earth observation group (GEO), world data system (WDS) and international data of science and technology commission (revealed) and other international institutions rely mainly on data policy to help the earth observation data sharing, organizations have also set up some centralized data-sharing platforms to provide data services. For example, the geospatial data cloud platform of the Network Center of the Chinese Academy of Sciences, the Natural Resources Remote Sensing Cloud service platform of Land Satellite Remote Sensing Application Center of the Ministry of Natural Resources, Remote Sensing Market cloud platform of Remote Sensing Group of Chinese Academy of Sciences [2], the Natural Resources Remote Sensing Cloud service platform of Land Satellite Remote Sensing Application Center of the Ministry of Natural Resources[3] and Sentinel data of European Space Agency [4]. Although cloud computing has evolved over the years, data security and trusted computing remain major challenges in current cloud computing applications. To solve these problems, many scholars have done a lot of research and put forward many models including data integrity testing and secure multi-party computing. However, most solutions face problems such as excessive computational complexity or lack of scalability. And cloud computing is centralized computing, once the central server goes down or other errors occur, the data will be inaccessible, untraceable, and face a great risk of tampering.
With the success of Bitcoin, the blockchain technology, as its underlying core technology, has gradually attracted the attention of a large number of scholars, and the blockchain technology has been developing continuously. Blockchain [5] is an emerging data sharing and application technology. It uses a peer-to-peer network, consensus mechanism, smart contract, cryptography, and other technologies to avoid the traditional centralized system of highly dependent on the central authority, high trust cost, poor reliability, and low-security credibility problems. The application of blockchain technology has been extended to digital finance, the Internet of things, intelligent manufacturing, supply chain management, medical care, people's livelihood, and other fields, and has a good solution strategy in data security, data sharing, and data rights confirmation.
At present, blockchain technology has been applied to many fields. As remote sensing application has become indispensable support for many industries in the world, trusted remote sensing information service has become a new user demand in a considerable number of countries around the world. Blockchain technology also brings feasible solutions for remote sensing information services. The decentralized, traceable, and unchangeable characteristics of blockchain can well solve the problems existing in traditional remote sensing data information services, such as single point of failure, unauthorized tampering, data leakage, data traceability, and so on. At present, the Open Geospatial Consortium [6] , the European Space Agency [7], and NASA [8] have all started to introduce blockchain technology into the study of geospatial data. Some domestic institutions and enterprises have also started to develop blockchain platforms for remote sensing data, such as Spatial Information Industry Blockchain Alliance, Beidou Aerospace Group and Digital China Research Institute, Oracle and Yungeng Agriculture, and Internet of Things Industry Research Institute of Zhejiang University. At present, blockchain technology is still in the development stage, and its application in many fields has not been fully verified. Its performance, operation, and maintenance cost, functional integrity, and many other aspects need to be improved. The application of blockchain technology in the field of geosciences is in the preliminary exploration. At present, there has been preliminary theoretical exploration and application research in such fields as map data sharing [9], water resources management [10], air pollution index measurement [11], cadastral data management and update [12] and property rights protection of geological resources [13]. Although these studies have tried data recording, real-time tracking, product traceability, and other aspects of the experiment, there are still many problems unresolved. These features put forward new requirements for blockchain. Currently, the blockchain still has the problem of difficult storage of remote sensing data in the blockchain, which is also the difficulty in the combination of blockchain technology and remote sensing data sharing.
To sum up, this paper intends to make full use of the advantages of blockchain to research remote sensing data sharing methods by taking advantage of blockchain's decentralized, secure and reliable features and preventing tampering. Because of the question, this article on the chain architecture and the block of data structure, on the one hand, by improving the existing blockchain model, according to the business functions of the internal need to build more chain structure, on the one hand, design and build data interoperability information model, the research content of the traceability of production process, on the other hand, the block storage model, Form efficient and distributed remote sensing data blockchain storage scheme to ensure data security and efficient access.
Decentralized, secure storage and sharing of remote sensing data based on blockchain will facilitate reliable sharing of remote sensing data around the world, help more people to participate in and access geospatial information services, and help spread geospatial information more widely. This will lay the foundation for the reconstruction of spatial data facilities and provide data and technical support.
Section 2 analyzes the relevant issues and work. The specific remote sensing data interoperability system architecture and the block data structure applied to the remote sensing production chain are introduced in Section 3. Section 4 introduces the realization of the prototype system and related experiments, and analyzes experimental results. Finally, conclusions and discussions of future work are presented in Section 5.

2. Problem description

2.1. Characteristics of remote sensing data

Remote sensing data refers to the remote sensing image obtained by various remote sensing techniques, which is multi-level, multi-angle, multi-spectral, multi-temporal, and multi-dimensional data with temporal, spatial and spectral dimensions. Based on the above definition, the remote sensing data are abstracted into the following types, . The is the hash value generated by the hashing operation of , is the spatial attribute, and is the temporal attribute, represents the metadata of this image itself, such as image type, size, scale, etc. And means remote sensing information, includes satellite and sensor information [14].
At the same time, remote sensing data also faces the characteristics of big data, with massive remote sensing datasets as the main data set and comprehensive other auxiliary data from multiple sources, and using big data thinking and means to focus on the theory, method, technology and activities of obtaining valuable information from massive remote sensing datasets of multiple sources, multiple media, multiple frequency bands and multiple resolutions.
At present, the main challenges facing the management and production of remote sensing big data are as follows.
  • Data credibility problem
From satellite data acquisition to product visualization for public service, the production process of remote sensing data faces various factors that cause data unreliability: a. The source of auxiliary data such as proximity sensors is unreliable. b. Starting from satellite data reception, remote sensing data will go through multiple pre-processing steps such as geometric correction, radiometric correction, atmospheric correction, etc., and then various value-added products will be obtained through various algorithms. Throughout the life cycle, the complex process brings the possibility of errors [15]. It is difficult to assess the accuracy of remote sensing data processing steps in the process, which leads to unreliable final decision knowledge. c. There is a risk of falsification in processing and circulation, and the data is recorded in a centralized platform, then there is a possibility of data being attacked and tampered with, both of which can cause unreliable data in circulation. These factors then cause the quality of the data that users eventually obtain cannot be guaranteed. This is one of the main challenges of remote sensing data production and management.
2.
The issue of data traceability and confirmation of rights
The traceability of remote sensing data to confirm the right refers to determining the ownership attribution and using right attribution of remote sensing product data. This issue and the credibility issue are complementary to each other. If we can successfully trace the source and clarify the data attribution issue, then this problem data can also find the source and provide the basis for the credibility of the data.
The current production process of remote sensing products cannot be authentically traced because of the lack of reliable accounting books for data processing, although the process of data processing can be recorded throughout the data-information-knowledge life cycle. The ownership of product data is often unclear. The traditional means of validation uses the model of submitting proof of ownership and expert review, which lacks technical credibility and has uncontrollable factors such as potential tampering. Improving the transparency of the process of realizing the value of remote sensing data and providing users with authentic and reliable datasets are other major issues of remote sensing big data management.
3.
Data sharing problem
In the current context of the big data era, using multiple remote sensing data sources for information extraction has become a major trend. In the current remote sensing data production process, it is almost impossible to produce multi-source remote sensing products under the premise of ensuring the privacy of non-public data. At present, multi-data centers use data aggregation mode for data sharing and trading among units, and the data sharing process requires intermediate institutions for aggregation, which has problems such as complex business processes, data leakage, and non-traceability. In addition, the willingness of remote sensing production workers to share product data is low, and the driving force is insufficient at present; the lack of audit or effective evaluation of the shared data makes it difficult to guarantee the quality of the shared data; the problem of data validation after sharing, and the inability to validate the rights will affect the motivation of data sharing. These are other important challenges faced by remote sensing products.
In summary, facing remote sensing data, blockchain still has certain problems in data sharing and supervision mechanism, trusted data sharing mode and data access control. In order to achieve safe and efficient sharing of remote sensing data, the data-sharing business has the following needs: (1) the sharing of remote sensing data requires the administrator to authenticate and manage the members involved in data sharing to prevent malicious members from participating in data sharing; (2) due to the sensitivity of remote sensing data and the characteristics of big data, it is necessary to adopt a storage method with high security to guarantee the safety of remote sensing data storage; (3) the adoption of a trusted data sharing mode to get rid of the complicated process and inefficiency caused by the need for intermediate organizations to summarize the traditional sharing mode; (4) data resources and data sharing records are traceable to increase the possibility of recourse for data misuse.

2.2. Research on data storage architecture of blockchain

Compared with traditional databases, blockchain also has many shortcomings in terms of performance. In terms of data storage, since all nodes in a blockchain system keep a copy of the blockchain, the increasing growth of data makes storage more and more difficult. To address this problem, researchers have proposed several expansion schemes, such as modifying the block capacity [16], converting the originally single-chain data structure to a directed acyclic graph DAG [17], or shifting complex computations and high-frequency transactions to be performed off-chain and storing the final results on the blockchain. The modifications of block capacity and directed acyclic graphs are mainly improved from the blockchain architecture to enhance the whole network's transaction processing capacity. However, this approach is mostly limited by the number of nodes and transactions in the network, with limited performance improvement, and it tends to exacerbate the risks of blockchain centralization and security attacks. One solution is to move computation and high-frequency transactions off-chain. This solution separates off-chain computation from on-chain storage. It can effectively improve blockchain performance while guaranteeing a certain degree of decentralization. However, off-chain scaling faces problems such as off-chain computation security and difficulty in supporting complex operations (e.g., smart contracts). Therefore, in-depth research on decentralization, security and performance needs to be conducted in the specific study concerning the characteristics of remote sensing data and applications.
The traditional blockchain represented by Bitcoin introduces a single-layer chain structure in terms of blocks, which is used to guarantee the tamper-evident nature of data on the chain. The block header and the block body together constitute a complete block. The block header stores the antecedent parent block hash, the transaction set Merkle root, etc., and the transaction data is stored in the block body. Currently, most of the blockchain systems are using single-chain architecture, such as the PoC project of the Japanese network bank and the common ledger model of the European Central Bank.
With the increasing data stored on the blockchain, research found that the single-chain architecture has corresponding bottlenecks in terms of performance, capacity, privacy and scalability, which are insufficient to manage the data, and a double-chain structure blockchain was invented inspired by the DNA structure. The double-chain structure blockchain means that two chains can compute in parallel and each chain can use different hardware; the double-chain structure not only maintains the characteristics of the original blockchain, but also improves the computing speed and scalability. In domestic and international research according to the needs of each application, scholars have proposed their double-chain structure. To solve the characteristics of single-chain structure without supervision and legal protection [18], Guangfu Wu et al. [19] proposed a double-chain structure of supervisory blockchain and transaction blockchain to separate supervision and transaction. C. Xie et al. [20] proposed a dual-chain structure of transaction blockchain and data blockchain for blockchain-based tracking of agricultural products, which solved the security problem and ensured that the data would not be maliciously tampered with or destroyed. W. Ren et al. [21] used a combination of public chain Ether and agricultural sample data chain to improve the security level. s. Chen et al. [22] proposed user blockchain and fog aggregation blockchain, again to enhance security. To alleviate the block data storage inflation problem, Shengqiao Gao et al. [23] established a double-chain structure of main and summary chains, which allows users to choose the appropriate storage mode according to their needs, which is also a more mainstream double-chain structure.
In some research, the constructed multiple blockchains are called multi-chain structure, which relies on the multi-chain structure to solve the problems that cannot be solved by one blockchain, that is, scaling, data sharing, security, etc. Polkadot [24] is a famous heterogeneous multi-chain framework, which is a system of a relay chain connecting multiple parallel chains, where the relay chain is responsible for consistency and the parallel chain is responsible for validity. Multichain [25] implements multiple private chains running simultaneously, but the chains are isolated and independent from each other. Parallel blockchain [26] proposed by Yuan Yong and Feiyue Wang is also a multi-chain structure that relies on the approach of parallel intelligence theory combined with blockchain technology to add practical computing power and more decision-making capability to the blockchain. Tsai et al. [27] proposed a new blockchain "Beihang Chain" that combines account blockchain (ABC) and transaction blockchain (TBC). Liang Hao [28] and others proposed user blockchain, transaction blockchain and agricultural information chain. The architectural form of blockchain has become more and more flexible and more suitable for each application. However, the existing multi-chain structure is not very good in terms of consensus efficiency and transaction throughput, and it is not suitable for the storage of large data.
As the number of file storage increases, the storage cost will also become more and more expensive. Moreover, the traditional blockchain combined with centralized cloud storage platform data storage scheme has the risk of overstepping and tampering with the cloud storage platform. In this context, the Interplanetary File System (IPFS), a peer-to-peer network for storing and sharing data in a distributed file system, was born. The network uses content addressing to uniquely identify each file, a feature that is well suited for distributed data storage. Therefore, many studies [29,30,31] use IPFS as the underlying data storage in combination with blockchain to achieve the effect of storing shared big data in blockchain to reduce the storage burden of blockchain.
At present, in blockchain-based data management, most of the researches mainly focus on single-chain structure, which is difficult to reach an effective balance between consensus efficiency, security and cost. In particular, the research in the field of remote sensing does not make full use of the features of public, federated and private chains according to the characteristics of remote sensing industry. Meanwhile, the data throughput of single-chain structure is low and far from sufficient to support diverse remote sensing applications. Therefore, it is necessary to break through the blockchain performance constraint and propose a new blockchain structure to support the on-chain demand of diversified remote sensing industries.
In summary, the application of blockchain technology in the field of geosciences is still in the nascent stage, mainly focusing on the exploration of applications in data copyright and data encryption, etc. These explorations show that the introduction of blockchain technology in the process of geosciences data sharing to solve certain sharing problems has good application prospects. However, the research on using blockchain technology for remote sensing data sharing is still extremely limited and most of them are still in the exploration stage. In view of the above challenges, this study proposes to use blockchain technology to overcome the challenges in data tampering detection, process documentation, process accuracy assessment, and privacy protection.

3. System

3.1. Trusted Interoperability Service of Remote Sensing Data

The remote sensing information from imaging to remote sensing product output roughly goes through the following processes: remote sensing data acquisition, data processing, image processing, information extraction/fusion/integration, and product output, in which various errors are introduced due to various influences such as operation and environment, and these errors will eventually produce uncertainties in remote sensing products, including uncertainties in remote sensing imaging mechanism, uncertainties in processing process and the uncertainty of information product evaluation.
The original intention of the remote sensing data trusted interoperability technology is to build a complete trust chain from data acquisition to product production in the entire remote sensing data processing system. Due to the non-objectivity of the actual operation, the remote sensing service itself may cause uncertainty and untrustworthiness of the results due to different algorithms selection and process selections. Only by sorting out its scientific nature, the origin and process of a remote sensing product image on the chain can be explained, and it can be mathematically deduced strictly and scientifically processed to obtain credible results.
After remote sensing images are acquired from satellite sensors, they are shared, collaborated, and fused between data storage centers. The products are dependent on various parameters in the production process. Facing the same target, the results obtained by different processing operations have different reliability. Therefore, remote sensing data products should be combined with the complete parameters on their own chains to promote and ensure the scientificity, traceability, and high reliability of production results, and to provide more reliable information for user decision-making.
To address the above problems, in order to realize the requirements of traceability of the whole course of remote sensing data changes and information tamper-proof, this paper design and build an interoperable meta-information model including the characteristics of the Earth observation data sources and the service records of the processing process, as shown in Figure 1.

3.2. Architecture of remote sensing data-sharing model based on multi-level blockchain

For a single-chain structure in the current blockchain-based data management, it is difficult to reach an effective balance between consensus efficiency, security, and cost. And it is not suitable for big data storage. In this paper, we store the original data, data summary, and transaction data separately, study the multi-chain model suitable for remote sensing big data storage, and design multiple chains with parallel execution to achieve high performance as well as low protection cost while ensuring data security and privacy security. In this paper, we study the combination method of public and federated chains, further conduct in-depth research on data storage methods in terms of security and performance, focus on the characteristics of remote sensing industry chains, and design remote sensing blockchain with high concurrency of transactions.
Since the storage capacity of blockchain is limited and remote sensing data is large, one data needs to occupy hundreds to gigabytes of storage space, it is not very realistic to store all remote sensing data on the blockchain and could introduce a risk of confidentiality, in order to provide better on-chain data storage, therefore, this paper proposes to store remote sensing data under the chain and build a data-sharing chain between remote sensing data owners to store remote sensing data under the chain In order to provide better on-chain data storage, this paper proposes to store remote sensing data under the chain and build a data-sharing chain among remote sensing data owners, which stores the hash value of remote sensing data under the chain for on-chain storage and ensures the consistency of the hash value of remote sensing data stored by each remote sensing data owner based on the consensus mechanism of blockchain to realize remote sensing sharing among remote sensing data owners.
Combining the characteristics of remote sensing applications, this paper proposes a remote sensing data storage model based on a multi-chain structure combining public chain and alliance chain and designs a three-layer blockchain storage architecture, including IPFS storage layer, remote sensing industry alliance chain layer, and remote sensing open data public chain layer. Using the IPFS storage layer to store remote sensing raw data and remote sensing product data after processing is convenient for the management and processing of big data, and not easy to be attacked by others. Since remote sensing data is highly specialized compared with general data, in addition to remote sensing image information, it also needs to save its descriptive information of remote sensing image data itself, including spatial, spectral, and temporal dimension information, etc. It needs to save metadata information, especially for easy searching. The information hash value is transmitted through blockchain, and data transmission and sharing are realized through IPFS to achieve secure data sharing. Its architecture diagram is shown in Figure 2.
The remote sensing industry chain needs to do a parallel cut of data according to the needs of remote sensing application business functions in order to achieve the business requirements and security requirements of data isolation. Since each full node in the single chain in the current blockchain technology system owns all data in the whole network, it cannot meet such application requirements. This paper designs a 1+N multi-chain structure based on remote sensing application research, including a public chain and multiple independent alliance chains, which partition business logic and data processing, support parallel processing of multiple transactions, and asynchronously write to the public chain transaction ledger after the transaction is completed.
Specifically, the Remote Sensing Open Data Chain (hereinafter referred to as RSOC), provides content-based search for the public, provides remote sensing products produced in the remote sensing alliance chain such as flood monitoring maps to the public, and the public provides open data based on incentive mechanisms and accesses data based on smart contracts.
Build the Remote Sensing Industry Consortium Chain (hereinafter referred to as RSIC) on the basis of the public chain. Put the verified and game-tested remote sensing trust services on the basic public chain, and build each alliance chain on the basic public chain. There may be some private data and private deployment on the alliance chain, while there is open data on the public chain, and the data is mainly used for deposition and basic trust services.
The problem of trustworthy data interaction between different agency nodes and between the general public and remote sensing agencies must be considered when designing a multi-chain structure combining public and coalition chains. The specific idea is that the alliance chain combines related remote sensing data centers according to remote sensing applications, shares remote sensing raw data among each other, realizes the shared alliance among different remote sensing industry institutions, and nodes on the same chain maintain the same data, thus promoting the collaborative development of upstream and downstream users in the remote sensing industry chain. All organizations among alliance chains share remote sensing metadata, transaction data as well as consensus protocols, and smart contracts, but do not share raw data between chains in order to protect data privacy. Each alliance chain ensures the integrity and traceability of a remote sensing industry chain, and alliance chains interact with each other through public chains.
Each chain consists of different nodes, and the nodes on the same chain maintain the same data, while the nodes on different chains maintain different data, so that the data on different chains are isolated from each other. At the same time, the nodes in the system can join different chains according to their own working needs, and that node only synchronizes the data on the relevant chain. Such a storage method can reduce to a greater extent the amount of data that nodes in the system need to synchronize and store, shorten the time consumption required for data synchronization, and reduce the storage space required for data storage. The data on different federated chains are isolated from each other, which can realize data concurrent operation of different chains and improve system concurrency.
IPFS can assist different blockchain networks to transfer information and files. The multi-chain structure combining public chain and federated chain can break through the function and performance constraints of a single chain and has good performance of high concurrent transactions while taking into account the isolation and protection of privacy data to meet the diversified business needs in the remote sensing field.

3.3. Block data structure serving large-scale remote sensing applications

For different remote sensing data providers with different requirements on data security, suitable blockchain data structures are designed for multi-chain structures respectively. This study constructs a new data structure in the blockchain as a prerequisite basis for remote sensing data storage and sharing.
In view of the current problems such as large differences in the quality of Earth observation information products and no unified evaluation index of information products for domain applications, we study the block storage model based on the open and recordable features of access to data and operation process information under the interoperable environment and form an efficient and distributed blockchain storage scheme for remote sensing data to ensure the security and efficient access to data. Blockchain technology can make the data in the blockchain more valuable. By multiple nodes participating in data recording together and verifying the accuracy and validity of their information with each other, it can achieve all-around confirmation and verification of information, thus eliminating uncertainty and reaching consensus.
The blockchain stores the data storage records of different users and the data storage addresses on IPFS and the stored records are guaranteed not to be tampered with by generating unique hash values through hash functions. This study analyzes the remote sensing data storage aspect and proposes a generic block data structure to store the related data with state attributes and remote sensing attributes and query them based on the uniqueness of the path.
The raw remote sensing data is stored on IPFS, and when the file is uploaded to the IPFS node for storage, the node will organize the file in the Merkle Directed Acyclic Graph (Merkle DAG) format for block storage, and after the storage is completed, the file will be represented by the root hash number of the Merkle DAG, and the user can build the network from IPFS in the way of Distributed Hash Table (DHT) Get the file. By hashing the recorded data blocks and storing the hash values in the blockchain, there is no need for nodes to have massive storage space. Once the content is modified, the corresponding hash value will also change and cannot be matched with the hash value in the blockchain, ensuring the unmodifiability of the content. The hash value ensures that the data is trustworthy and has not been tampered with, so accurate backtracking can be achieved and the property rights of the data are guaranteed.
Since remote sensing data has strong professionalism compared with general data, in addition to remote sensing image information, it also needs to save its descriptive information of remote sensing image data itself, including spatial, spectral, and temporal dimensional information, etc. It needs to save metadata information, especially for easy finding, through metadata we can easily query the remote sensing data we want to access without exposing confidential data at the same time. Therefore, we divide the remote sensing raw data into two files, remote sensing image data, and metadata, and perform a hash calculation on them separately.
In the multi-chain structure designed in subsection 3.2, the public chain is used as the main chain to store the remote sensing open data and the summary of the federated chain blocks. Considering the different identities of data providers, the remote sensing data center should provide a data-sharing platform between them to facilitate collaborative operations, while the open data providers should ensure data trustworthiness, and data privacy protection to solve the problems of mutual trust, and data authenticity and confidentiality. As shown in Figure 3, this paper divides the public chain blocks into two types, one is to record the various transaction data in the alliance chain, which is recorded as Class A blocks. At the same time, because it is a public chain, the data are all public, so it is convenient for transactions between various alliance chains through the public chain, which is obtained by the synchronization of alliance chains; the other type is the class B block, which is obtained by the general public after submitting open data.
The data storage structure of this blockchain system is based on the design idea of Ethernet. The remote sensing information blocks form a chain structure in a linear way, consisting of a block header and a block body.
The data block of the public chain is referred to as the main chain block, and the main chain block body includes the ledger transaction data and the IPFS hash value of the telemetry metadata. The public searchable data recorded by the transaction data with storage incidentally stores the information such as the serial number of the coalition chain where the corresponding telemetry data is located, the block height, and the block hash value of the block where the data is located in the coalition chain.
The federation chain maintains the actual transactions in each chain, and the data block is called the slave chain block. The IPFS hash value in the block body contains the hash value of the remote sensing data block and the hash value of the remote sensing metadata.

4. Experiment

In this section, we provide experimental simulations for the system proposed in this paper. The evaluation was performed on an Ubuntu 20.04 system with a Core i5 CPU and 8GB RAM. The network speed is 50 Mbps. Access IPFS through the browser front-end, upload remote sensing pictures, and transfer the hash generated by IPFS to the blockchain through the web and node.js servers, and the hash data is stored in the blockchain through smart contracts.
The solution proposed in this paper uses RSIC, RSOC blockchain and IPFS network. The specific blockchain in this article is built using geth (go-ethereum). The parts related to smart contracts are written by solidity.

4.1. Data upload efficiency

In the prototype system designed in this paper, the remote sensing data is successfully uploaded to IPFS and the blockchain. The data upload time to IPFS storage is shown in the Figure 4.
The main chain in multiple blockchains is Ethereum, which provides content-based public search. The secondary chain is built for a specific remote sensing data production application system. Firstly, a large amount of remote sensing data is transmitted to IPFS. The IPFS network stores the entire data received while redirecting the data to the consortium chain: RSOC. Secondly, the RSOC chain stores hashes and metadata in blocks and generates RSOC block hashes. Finally, the block hash that can be made public is uploaded to the Ethereum main chain and stored in the block.

4.2. Data Provenance

Another important goal of this system is to complete the traceability of data. From the metadata stored in the block, the remote sensing image data on IPFS can be obtained. In this section, we show the results of using the Remix IDE [32] to achieve provenance. Remix is an IDE that allows to create, compile and deploy smart contracts. It also provides an environment for executing transactions and interacting with smart contracts. First, we use Remix to create and deploy our smart contract. To interact with this smart contract, we use the Web3 Provider to connect Remix with our blockchain. In Figure 5, we display the produced remote sensing product information using Remix.

4.3. Security and Availability

The remote sensing data trusted and traceable storage system designed in this paper is more secure than the traditional system. As mentioned above, traditional remote sensing data storage is a centralized technology based on cloud computing or cluster computing. Once the central server is destroyed, the data in it will be tampered with and destroyed. Although the public chain is a completely decentralized technology, its upload and download performance is slow, making it difficult to meet the throughput of large-scale data in global monitoring applications. At the same time, participation in the consensus process by people unrelated to the production process of a particular remote sensing product is redundant. Although the upload and download speed of the alliance chain is greatly improved compared with the public chain, incomplete decentralization makes it constrained by the central node, and the network also has hidden dangers such as witch attacks.
Therefore, this paper considers a hybrid chain design method, and the multi-chain structure, IPFS storage network, and Ethereum technology jointly build a more reliable remote sensing data sharing system. This is mainly because: (1) The availability of the prototype system benefits from the ability to store various remote sensing data. Remote sensing images are stored separately on IPFS, and their stored hash values ​​are stored in blocks together with metadata, which not only protects the data privacy but also ensures user interaction with the system. (2) The designed RSIC and RSOC hybrid chain transmits data through IPFS and is applied to the remote sensing data sharing network. The hash value transmitted by IPFS is stored in RSIC, and the public product data hash block on RSIC is uploaded to RSOC, and the security is verified by comparing the hash value. (3) All transactions are recorded and stored in the immutable ledger of the blockchain, thus providing a high degree of transparency and traceability for the production of each data in a secure, trusted, reliable and efficient manner. (4) The "IPFS + blockchain" method not only retains the advantages of decentralization and security of the blockchain, but also solves the long-standing problem of scalability, and can store remote sensing big data.

5. Conclusion

In this paper, we propose a data storage security solution for remote sensing products based on multiple blockchains, that provides an accountable and distributed storage environment for storage participants. Firstly, the storage structure of the block is proposed as a trusted representation of shared data. Then, a data model for sharing data requests and services based on blockchain is proposed, which makes data sharing more credible. Finally, through the Ethereum blockchain platform, the prototype system is designed and simulated. Multi-chain solutions with IPFS storage solves the problems of storage centralization, centralized verification of data integrity and denial of service in cloud storage scenarios. It offers a higher level of security than traditional cloud storage and Ethereum. Our implementation and evaluation of the prototype system showed that it is a highly viable solution in terms of data storage speed, traceability, security and availability. The storage scenario in this paper still has a few challenges, for example, data could be encrypted for further security.

Author Contributions

Conceptualization, Q.Z. and Z.B.; Methodology, Q.Z. and W.Y.; Software, Q.Z. and Z.B.; Validation, W.Y.; Formal Analysis, Q.Z. and W.Y.; Investigation, Q.Z.; Resources, Q.Z. and Z.B.; Data Curation, W.Y. and Q.Z.; Writing—Original Draft Preparation, Q.Z.; Writing—Review & Editing, Z.B. and W.Y.; Visualization, Z.B.; Supervision, Z.B. and W.Y.; Project Administration, Q.Z. and Z.B.; Funding Acquisition, W.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the International Partnership Program of Chinese Academy of Sciences, grant number 131211KYSB20180002.

Data Availability Statement

The data for this study is available within the article or its supplementary material. Data used in this study were derived from the following resources that are available in the public domain: [Geospatial data cloud, Computer Network Information Center, Chinese Academy of Sciences: https://www.gscloud.cn/sources].

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

References

  1. Li, G. Q., H. Y. Zhang, L. C. Zhang, Y. Y. Wang, and C. Z. Tian. “Development and trend of Earth observation data sharing.” Journal of Remote Sensing 2016, 20, 979–990. [CrossRef]
  2. Gscloud, 2023. “Geospatial data cloud,” accessed in 23 May 2023. http://www.gscloud.cn.
  3. Sasclouds, 2023. “Natural resources satellite remote sensing cloud service platform”, accessed in 23 May 2023. http://sasclouds.com/chinese/home.
  4. Eurpean Space Agency(ESA), 2023. “Copernicus open access hub”, accessed in 23 May 2023. https://scihub.copernicus.eu/.
  5. Nakamoto, S. 2008. Bitcoin: A Peer-to-Peer Electronic Cash System. https://bitcoin.org/bitcoin.pdf.
  6. OGC DWG, 2023. accessed in 23 May 2023. https://www.ogc.org/projects/groups/bdltdwg.
  7. Eurpean Space Agency(ESA), 2019. Blockchain and Earth Observation. https://eo4society.esa.int/wp-content/uploads/2019/04/Blockchain-and-Earth-Observation_White-Paper-April-2019.pdf.
  8. Mandl, D, 2017. “Bitcoin, Blockchains and Efficient Distributed Spacecraft Mission Control.” Information Science & Technology (IS&T) Colloquium 1-36. https://ntrs.nasa.gov/api/citations/20170009470/downloads/20170009470.pdf.
  9. FOAM, 2023. FOAM –The Consensus Driven Map of the World. https://foam.space/publicAssets/FOAM_Whitepaper.pdf.
  10. Khatri, Y, 2019. IBM Blockchain Assists Groundwater Pilot in Drought-prone California. https://www.coindesk.com/ibm-blockchain-assists-groundwater-pilot-in-drought-prone-california.
  11. Yohan, H., P. Byungjun, and J. Jongpil. “A Novel Architecture of Air Pollution Measurement Platform Using 5G and Blockchain for Industrial IoT Applications.” Procedia Computer Science 2019, 155, 728–733. [CrossRef]
  12. Torun, A. 2017. “Hierarchical Blockchain Architecture for a Relaxed Hegemony on Cadastre Data Management and Update: A Case Study for Turkey.” Uctea International Geographical Information Systems Congress. https://www.researchgate.net/publication/321485252_Hierarchical_Blockchain_Architecture_for_a_Relaxed_Hegemony_on_Cadastre_Data_Management_and_Update_A_Case_Study_for_Turkey.
  13. Zhang M. D., Z. J. Gao. “A discussion of blockchain application in the intellectual property protection of geological big data.” China Mining Magazine 2019, 28, 9–14.
  14. Zou, Q., G. Q. Li, W. and Y. Yu. “MapReduce functions to remote sensing distributed data processing—Global vegetation drought monitoring as example.” Software- Practice and Experience 2018, 48, 1352–1367. [CrossRef]
  15. Jing G., F. , “Analysis on Remote Sensing Credibility and Trusted System interoperability. ” Remote Sensing Information 2020, 35. [Google Scholar] [CrossRef]
  16. BIP, 2023. “Bitcoin improvement proposals,” accessed in 23 May 2023. https://github.com/bitcoin/bips.
  17. COTI, 2023. “Currency Of The Internet”, accessed in 23 May 2023. https://github.com/coti-io.
  18. Wasim, M. U. A. Z. A. Ibrahim, P. Bouvry, et al. 2017. “Law as a service (LaaS): Enabling legal protection over a blockchain network.” 2017 14th International Conference on Smart Cities: Improving Quality of Life Using ICT & IoT. [CrossRef]
  19. Wu G. F., P. Yu, K. K. Wang. “Transaction regulatory research on double-chain blockchain.” Computer Engineering and Applications 2020, 56, 116–123. [CrossRef]
  20. Xie, C., Y. Sun, and H. Luo, “Secured data storage scheme based on block chain for agricultural products tracking.” 2017 3rd International Conference on Big Data Computing and Communications 2017. 45-50. [CrossRef]
  21. Ren, W., Wan, X. T., Gan, P. C. “A double-blockchain solution for agricultural sampled data security in Internet of Things network.” Future Generation Computer Systems 2021, 117, 453–461. [CrossRef]
  22. Chen, S., L. Yang, C. Zhao, V. Varadarajan, and K. Wang. Varadarajan, and K. Wang. “Double-blockchain Assisted Secure and Anonymous Data Aggregation for Fog-enabled Smart Grid.” Engineering 2020, 8. [CrossRef]
  23. Gao, S. Q., X. L. Liu, and Y. P. Gao. “Optimized model of dual-chain storage of food supply chain data based on blockchain.”. Food&Machinery 2020, 36, 63–70. [CrossRef]
  24. Wood, G. “Polkadot: Vision for a heterogeneous multi-chain framework. ” White Paper 2016, 1–21, 2327–4662. [Google Scholar]
  25. Multichain, 2023, accessed in . http://www.multichain.com/.
  26. Yuan, Y., F. Y. Wang. “Parallel blockchain: concept, methods and issues”. Acta Automatica Sinica 2017, 43, 1703–1712. [Google Scholar] [CrossRef]
  27. Tsai, W. T. Blower, Y. Zhu, L. Yu. “A system view of financial blockchains.” 2016 IEEE Symposium on Service-Oriented System Engineering (SOSE) 2016. 450-457. [CrossRef]
  28. Liang, H., S. Liu, Y. Zhang, and K. Lv. “Multi-blockchain application technology for agricultural products transaction.” Smart Agriculture 2019, 1, 72–82. [Google Scholar] [CrossRef]
  29. Hoffman, A. Becerril-Blas, K. Moreno, and Y. Kim. 2020. “Decentralized security bounty management on blockchain and IPFS.” 2020 10th Annual Computing and Communication Workshop and Conference. [CrossRef]
  30. Reen, G., M. Mohandas, and S. Venkatesan. 2019. “Decentralized patient centric e-Health record management system using blockchain and IPFS.” 2019 IEEE Conference on Information and Communication Technology 1-7. [CrossRef]
  31. Tiwari, A. , and U. Batra. 2020. “IPFS enabled blockchain for smart cities.” International Journal of Information Technology 13(BJIT). [CrossRef]
  32. Remix, 2023. “Remix – Ethereum IDE”, accessed in 23 May 2023. https://remix.ethereum.org.
Preprints 81209 g001
Figure 1. The interoperable meta-information model.
Figure 1. The interoperable meta-information model.
Preprints 81209 g001
Preprints 81209 g002
Figure 2. Multi-chain structure blockchain architecture.
Figure 2. Multi-chain structure blockchain architecture.
Preprints 81209 g002
Preprints 81209 g003
Figure 3. Block data structure.
Figure 3. Block data structure.
Preprints 81209 g003
Preprints 81209 g004
Figure 4. The upload time of remote sensing data to IPFS.
Figure 4. The upload time of remote sensing data to IPFS.
Preprints 81209 g004
Preprints 81209 g005
Figure 5. Read remote sensing data information using Remix.
Figure 5. Read remote sensing data information using Remix.
Preprints 81209 g005
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

facebook logotwitter logolinkedin logo
MDPI Initiatives
Important Links
Subscribe

Disclaimer

Terms of Use

Privacy Policy

© 2024 MDPI (Basel, Switzerland) unless otherwise stated