Application of real-time data processing system of Internet of Things based on blockchain technology in the financial field of Yangtze River Delta urban agglomeration

After more than 40 years of long-term development, as a pioneer of China's reform and opening up, the Yangtze River Delta region has made tremendous economic progress and strong scientific and technological innovation strength, which has made the economy flourish and scientific and technological innovation strength strong. However, in the process of financial growth in the Yangtze River Delta, there are also financial structures that do not match the optimization of industrial structure, thus hindering the quality and efficiency of industrial structure optimization, thus affecting the strategic function of industrial development. For the development of the financial field, this paper applies the real-time data processing system of the Internet of Things based on blockchain technology to analyze the financial industry of the Yangtze River Delta urban agglomeration. The Internet of Things, based on the technology of real-time data processing system, can store different types of financial assets. These assets are a common component of supporting the application of the Internet of Things. They mix and build a powerful back-end process that can support the data processing application of the Internet of Things. Therefore, the Internet of Things data network and other information are stored in the data structure, so that data analysis and consolidation can be carried out with consensus, Thus, the processing efficiency of relevant financing data in the Yangtze River Delta region has been greatly improved, and the development process of the financial sector in the region has been effectively accelerated. Therefore, the research on this issue will enable the financial industry to give full play to its potential as a major component of national economic development, so as to maximize the improvement of the industrial structure and promote the development of blockchain and Internet of Things technologies.


Introduction
As for the Yangtze River Delta region that continuously adapts to the new natural conditions, it is essential to comprehensively assess their financial efficiency and explore ways to improve it for understanding and studying the comprehensive development of the Yangtze River Delta and seeking the economic transition to a high-quality development stage [1][2][3].By 2021, the total economic output of 27 major cities in the Yangtze River Delta region has exceeded 20 trillion yuan, accounting for 20.68% of the national GDP.In addition, the value of economic production in the Yangtze River Delta is also expanding year by year, the value of financial production is also increasing year by year, and the contribution to the national economy is also increasing [4].The significant impact of financial development on the national economic development and industrial structure is gradually emerging.With the prosperity of Internet of Things technology, intelligent terminal devices have also entered people's daily life, and personal data can also be transmitted to various device nodes [5].At the same time, the traditional three-tier model of the Internet of Things does not have a perfect security architecture, and the computing storage capacity of the terminal device is also quite limited.Therefore, the device does not have an intelligent privacy protection system, and the terminal device is vulnerable to attacks when transmitting data, leading to the disclosure of privacy data [6].However, blockchain technology provides a safe and reliable way to use the distributed shared information model to achieve transparency, security, privacy, auditability, access authentication (identity verification), data immobility and other functions.Blockchain technology can be used to deploy the Internet of Things system on unreliable networks [7].The real-time data processing system of the Internet of Things based on blockchain technology can effectively eliminate single points of failure while keeping the data of terminal devices still, establish trust between different terminal devices, eliminate intermediaries, reduce the cost of the system, and ensure the proprietary functions of devices and the integrity of data.Therefore, this paper applies this to the urban agglomeration economic sector of the Yangtze River Delta, with the blockchain layer as the independent layer [8].When the Internet of Things terminal equipment accesses the blockchain network through an intelligent gateway, it relies on network connectivity to ensure the security of financial information and achieve efficient data processing.

Related work
The literature points out that finance is a key factor for a country, so the development of financial economy has great potential, which can promote the optimization of industrial structure [9].At the same time, there is still much work to be done in economic development and industrial structure optimization.In order to promote the optimization of the industrial structure, the economic industry has enough influence to affect the allocation of resources.Therefore, it is necessary to support the sound development of the financial industry to promote the development of financial services and industrial services [10].It is also necessary to support the development of the real economy and promote the upgrading and optimization of the industrial structure.Based on the theoretical research of financial development, the literature puts forward the concept of sustainable financial development, dividing financial efficiency into partial financial efficiency, which is the basis of partial financial efficiency and general financial efficiency, and fully recognizes the important role of financial innovation in financial development [11].The literature takes 16 major cities in the Yangtze River Delta as research materials, with the help of panel data from 2010 to 2020, and through research and analysis, it is known that when considering that there is a long-term stable equilibrium relationship between financial industry development and industrial structure optimization, economic development plays a role of financial support for industrial structure optimization on the one hand, and financial support for industrial structure optimization on the other [12].The literature shows that in the whole data transmission process, the establishment of a complex Internet of Things network makes the security of Internet of Things applications extremely dangerous when its data is transmitted through different transmission methods.Therefore, this security issue needs to be considered in stages [13].Data is the most important part of the IoT system.In order to ensure the reliability of data, we need to achieve a reliable IoT vision under information security.To sum up, we need to consider the security of reliable IoT data from the entire life cycle of IoT data transmission, processing, storage, etc. to ensure data security [14].The literature points out that blockchain has many advantages.The first is decentralization.Blockchain is a group of point-to-point nodes, and counting can effectively avoid single point failure and other problems.The second is non convertibility.The blockchain connects all blocks with numbered hash functions, while retaining the main account data, so as to jointly maintain the main ledger data and ensure that the data is not tampered with.Thirdly, transparent blockchain provides a completely auditable audit record for recording transactions [15].The entire network is shared in unauthorized blockchains, and one group is shared in approved block sequences.The literature proposed a new structure of the hybrid Internet of Things system based on blockchain technology.The blockchain Internet of Things platform can be divided into device awareness layer, network public chain layer, intelligent network portal and application management layer [16].Among them, the perceptual processing layer of the device is directly related to the physical object, and supports the network connection between entities operating in the "decentralized" mode, and directly collects information through sensors.

Block Data Query Scheme
After the verification data is uploaded to the blockchain of the IoT node, the verification data cannot be tampered with; The current time of the block data cannot be changed.Therefore, the data of all blocks except the current block are determined, so dynamic modification is not required.Therefore, the hash computing server can generate a better performance Bloomberg filter based on the determined data volume.The criterion of Bloom filter with the best performance is that the possibility of error normality is the lowest and the memory consumption is the lowest.The BFi length of the Blunt filter built for this block is mi, and the normal probability of error can be calculated from the current situation as follows: Suppose that r spatial location attributes are written into the Bloon filter BFi, that is, ki hash functions are used to map elements to BFi.Each hash function maps data to a certain position in the Bloon filter 1.All hash functions are used to insert spatial location attributes into the Bloon filter BFi, which can be expressed as follows: Add all spatial location attributes in the block to the corresponding Blum filter BFi, as shown in the following formula: Input the location characteristics of all blocks into the corresponding block color, and the number of hash calculations required in the Bloom filter is: Check the Bloom filter BFi suitable for block Bi, and use all the hash functions in BFi to calculate the mapping position result of spatial position attribute Y, which can be expressed as follows: [] == ℎ  (), (0 ≤  ≤  − 1) (5)   To challenge the index, suppose that the block to be challenged is limited to B, the corresponding Bloom filter set is BF, and then the number of hash calculations required to challenge the index should be: Error rate in the first study: assuming that the quality is limited B, each BFi Bloon filter has a false natural probability, so the error probability of each study is:

Internet of Things transmission security model
When judging the blockchain authentication network, only after the chain authenticates the node and votes the identity information of the node to be connected to the network, can the network determine and vote to enter the contract chain of the network, and the device can join the blockchain authentication network.The device SDi requests the data on the chain through the ID of cloud platform P, obtains the identity key hash tail value and signature tail value Sig of cloud platform P, and performs signature verification.Once the signature verification is passed, the main key formula of the double ratchet algorithm is calculated through the X3DH algorithm as follows: Calculate the master key of the additional algorithm used to extend the Chebyshev ratchet  =   ( 2 )( 9) Cloud platform P computing And the extended increment key of Chebyshev ratchet, the formula is as follows  =   ( 1 )( 11)  =   ( 2 ) =   (  ())  =   ()( 12)  =   ( 1 ) =   (  ()) =   () (13) After obtaining the master key SK and the value-added key ST of the double ratchet algorithm, both parties calculate the master key S and the master key SK of the next double ratchet algorithm through the session key S and the KDF ratchet.The proposed formula is shown in the following formula.
( ′ , ) = (, ) (14) At this stage, the main negotiations between the two parties have been completed.At the same time, the cloud platform has achieved the control of the equipment and the equipment has reported data information to the cloud platform.Therefore, the two parties have completed the negotiations on the key, and used the drag control equipment through closed meetings and data information reports.
In order to avoid the connection between the account in the PoW consensus and the BFT consensus, this study uses verifiable random functions to realize offline quick selection of nodes that make up the block, and the winning node becomes the node that generates the block.
The verifiable random function is an encrypted hash of the public key version.Only the owner of the private key can calculate the hash value, but any participant who knows the public key can check the accuracy of the hash.The operation of the algorithm is as follows: The private key holder uses the private key SK and the public input data alpha to calculate the hash beta and evidence pi.
= _(, ℎ)(15)  = _2ℎℎ() (16) If it matches the provider, the hash is correct: In this paper, the consensus algorithm uses a consensus mechanism to divide the time into fixed length cycles and use it in each RF technology cycle to determine whether the current round harmonic node is selected as the node.VRF requires all participants to hold key pairs.The block out node election algorithm process in each round is as follows: First, calculate the alpha of the shared information in the current round, where t represents the current time and t represents the duration of the round: This node uses its personal key and shared information alpha to calculate the hash value and prove the pf value: Xk is treated as related k in the following way, where hashlen represents the bit length of the hash beta:

Algorithm optimization and parameter selection
Define the quality of data collected by node Si as: The pdf of dl can be expressed as follows For normal nodes, you can simplify the normal distribution and understand the data quality of data.Then, for common nodes, the formula can be rewritten as follows: The difference calculation is performed on the actual inventory data, which is the confidence interval between the upper and lower gray lines of the model.The value P is the coordinate that crosses the upper confidence interval of ACF for the first time, and q is the abscissa that crosses the upper confidence interval of PACF for the first time.
As shown in Table 1, pf and PACE show two obvious trailing forms.It can be noted that the transaction delay in the first phase has basically not changed, while the delay in the second phase is obviously a cyclical cycle, because the sensor data submission is regular, so the transactions on the website are basically regular.
When the gateway submits close to the next round of transactions, only a small transaction delay needs to be confirmed, otherwise a large transaction delay is required.

Figure 1 Transaction Delay Time Change Chart
The consensus algorithm used in the structure blockchain network built in this paper is KAFKA, which has been tested for different log levels of peer nodes.The log level includes debugging and error modes.The performance test of each mode's open operation is performed.The open test is used to perform account opening, writing and other activities.Among them, there are four rounds of opening test.The number of transactions in each round is set to 1000, and the correct transaction speed is set to 50tps, 100tps and 1SOtps200tps respectively.Table 2 shows the test results for this log level.Economic data analysis is based on the real-time observation data provided when the system realizes intelligent mining and decision-making.The commonly used data include the data of stock trading of listed companies, financial daily reports, stock market news, etc.The verification, extraction, cleaning, consolidation and synthesis shall be carried out in accordance with the screening rules and standards, and shall be presented in graphical form.In the design and implementation, according to the characteristics of the stock data obtained from the securities market, Sina Finance, Netease Finance, Bigquant and other industry quantitative analysis platforms and financial analysis websites were developed.The system formed a set of practical technical solutions, giving consideration to the convenience of rapid implementation of the system, and financial data analysis was carried out on financial data.Combining deep learning and machine learning methods to analyze and predict.When data processing and other specific algorithm processes are required, the "Larave" and "Vuei" frameworks are developed to separate the two sides.When using Python language for data analysis, data is transmitted through the interface, the analysis results are placed on the front page, and the open source database management system is used to access and store the existing system.The economic data analysis platform of this paper finally determines the main functions of the platform by investigating the traditional stock analysis software and integrating the relevant data of the actual stock market: (1) Data collection: basic stock data, historical price data and stock market news data are obtained by using crawling technology module and multithreading technology.The acquisition module will acquire data, create a data pool, perform data preprocessing, save the database in batches to specific fields, and establish automatic and regular updating methods to achieve a large amount of historical data.
(2) Data analysis: rank and analyze the data according to the historical transaction data and text information of the listed companies' shares, as well as statistical data, corresponding cycles, seasons, correlations, etc.Through the relationship between stock data, targeted analysis is conducted on individual stocks or comparative analysis is conducted on multiple stocks.Based on the analysis of financial data needs, the system is mainly responsible for data collection, data analysis, trend prediction and intelligent evaluation.The system architecture is shown in Figure 2:  The entire financial data analysis system can run on independent servers or Alibaba Cloud and other hosts.The financial analysis system in this project mainly consists of front-end pages, display layers, business layers, data layers, and database access layers.The home page visualizes the important functions of the financial analysis system, which provides users with good interactivity and functionality.

Flow of financial data processing module
FPGA hardware decoding in line with the network communication protocol also adopts modular design, which is divided into three modules: data segmentation, data merging, and data decoding.The data segmentation module realizes the functions of extracting existing bitmaps, marking the start and end positions of data fields, and slicing data fields.The data merging module realizes the judgment of data type and field merging, and the data decoding module recognizes the fast decoding of data fields, as well as the updating of previous value lists and other functions.The overall design process is shown in Figure 3.In the decryption module, the data domain is solved through the previous location map and value table.At the same time, the latest FAST data has been updated to the previous value table to analyze the next FAST protocol.

System real-time data processing performance analysis
The experimental results are shown in Figure 4: Vrf-poa Within a certain limit, it can be seen from Figure 4 that with the increase of the test load, the throughput of the system increases through synchronization and increase.When the test load is 800 tps, the system reaches the maximum throughput, which is close to 600 tps.When the test load exceeds 800 tps, the actual system throughput starts to decline, indicating that the simulation system is overloaded.As shown in Figure 5, when there is no information about the mc, the intelligent gateway is not very successful in tracking device node data, and the data integrity also has errors.When the running time exceeds 800ms, Merkle's traceability success rate based on the reliable tree is higher than the original Merkle tree.In a short time, Merkle's trusted tree will question and update the Mac address data, which will take more time.When the local device information list is established, its traceability greatly improves the success rate of the query.This section statistically analyzes the resource consumption of the FAST protocol analysis module and the market reconfiguration module.See Table 3 for specific data.It can be seen from Table 3 that the resource consumption of LUTS and FIFQ is still large.BRAM is used for the market reconfiguration module, while the market reconfiguration module is used because it uses the search and storage functions of the market reconfiguration module.In addition, FPGA system mainly collects and removes data through field computing, mainly through field computing, so DSP resources are also few.

Analysis of financial measurement results of Yangtze River Delta urban agglomeration
The level of economic development and the scale of economic industries affect the economic benefits of cities to a large extent, so there is a large gap in the economic benefit indicators of cities in the Yangtze River Delta.From the perspective of economies of scale, due to small-scale effects, some cities have low economic benefits, but their pure technical efficiency is not very low.This means that these cities with relatively backward economic development can improve their development efficiency by encouraging them to expand the scope of economic and industrial development.4 that the average value of the economic benefit index of City S is significantly higher than that of the other three provinces, and there is still a large gap between Province Z and Province A, compared with City A and Province J.According to the results of the overall economic benefit measurement, S City is the largest economic center in China, and has ranked first in the national economic benefit ranking for 15 consecutive years.The forward correlation represents the role of the industry in the development of other industries.The previous connection is the contribution of industry to the development of other industries.The final products are mostly concentrated in the financial sector, followed by the industrial sector.The industries in S City and Z Province are more closely related to the financial sector, and the financial sector itself is also closely related.In provinces J and A, the direct and full distribution coefficients are higher, mostly in the service sector, but not in the financial sector.As the financial industry of province A provides most of its products and services to one industry, while other sectors only account for a small part of its weight, province A is expanding its financial business to increase its positive contribution to other industries.

Conclusion
The Yangtze River Delta has an active economic activity, developed economy, leading national reform, perfect financial market system and high level of financial development, laying a foundation for economic and industrial innovation.Based on this practical background, this research is based on blockchain technology and real-time data processing system.It is conducted in the financial field of the long triangle urban agglomeration.Starting from the current development situation, expanding financial sectors and expanding business activities, the overall financial progress of the four provinces has greatly improved financial efficiency.However, the main expected achievements of this stage need to further improve the structure of each component, starting from the Expo, We will expand the scale of the development of the financial industry.To discuss that the regional financial integration project in the Yangtze River Delta has a significant impact on financial development, so as to accelerate long-term integration, deepen the deep integration of regional finance in the Yangtze River Delta, and promote the development content of regional economic integration.Therefore, this discussion is the internal requirement of the "National Strategy for Yangtze River Delta Integration".Through regional financial integration, we can promote development, so as to better safeguard economic and social stability, and more actively participate in global cooperation and competition, Innovate.

Conflict of interest
The authors declare that they have no conflict of interests

Ethical approval
This article does not contain any studies with human participants performed by any of the authors.

( 3 )
Trend prediction: forecast and analyze the change trend of the stock market in the form of discrete k-line according to the multi stock index, predict the price of a single stock, select different methods to predict the recent trend and price of the stock, and use different methods to compare the prediction algorithms to test the effectiveness of the prediction results.(4) In terms of intelligent evaluation: according to the trading data and text information of the stocks obtained, identify individual stocks, analyze positive and negative correlations, and analyze the stock clusters of specific industries to facilitate investors to analyze the stock market.

Figure 2
Figure 2 Basic architecture of the system

Figure 3 FAST
Figure 3 FAST Protocol Data Decoding Process

Figure 4
Figure 4 Throughput Test Results

Figure 5
Figure 5 Comparison of bit error rates in different Merkle tree environments

Table 1 ARIMA
Model Training Data Fitting Table

Table 2
Test results with the log level of ERROR

Table 3
Utilization rate of main hardware resources of financial module The pure technical efficiency of the representative cities in the middle of the Yangtze River Delta is 0.6-0.7,and the economic and industrial scale of these cities has developed moderately.From the perspective of current urban development, the key to improving the technical development level of the financial industry is to improve economic efficiency and promote the healthy development of the financial market, including advanced scientific management technology, advanced technical capabilities, and the organizational structure of foreign-funded enterprises.The average of the comprehensive technical efficiency of urban economic development of communities Yangtze River Delta cities from 2016 to 2021 is shown in Table4: It can be seen from Table