A kind of financial big data processing method and its system platform based on block chain
Technical field
The invention belongs to block chain financial technology fields, and in particular to a kind of financial big data processing side based on block chain
Method and its system platform.
Background technique
As the rapid development of information science technology, especially block chain, cloud computing, big data technology are in internet finance
It is widely applied, especially block chain has potential in financial fields such as international exchange, the letter of credit, equity registration and stock exchanges
Huge applications value.By block chain technical application in financial industry, third party's intermediacy can be saved, is realized point-to-point
Direct docking, so that transaction payment is rapidly completed, the core competitiveness of the following financial circles is very while substantially reducing cost
Speed and ability in big degree dependent on extraction information and knowledge from big data, and this speed and ability, depend on number
According to processing and application level.It is flat to finance data with continuing to bring out for the Novel finicals industry situations such as internet finance, mobile payment
Platform requires also to correspondingly increase.
Financial database is very high to the accuracy of data, reliability requirement, and it must have the stronger property checked.
Most financial product needs long-time tracking to handle, and data also have diversity, complexity.But current actual conditions
It is that financial product data handling procedure is complicated, in interaction, the bit error rate is high and safety is again and again by prestige for financial business core data
The side of body, bank finance financing, security, insurance company VI P client carry out business operation and access when and normal client data volume
It tests and speed experience difference is little, meanwhile, there is also financial products newly developed and existing operation system data sharing to deposit
In incompatible problem.
Summary of the invention
The purpose of the invention is to overcome above-mentioned deficiency to provide a kind of financial big data processing method based on block chain
And its system platform.
A kind of financial big data processing method based on block chain, comprising the following steps:
Data input unit receives the data of financial transaction of input, and data of financial transaction is sent to data-optimized list
Member;
Data-optimized unit receives data of financial transaction and optimizes to it, and provides financial transaction to Data Management Unit
Data storage path;
Data Management Unit receives data storage path and optimized data of financial transaction, and to by optimizing
Data of financial transaction afterwards is encoded, and data storage path and data encoding are sent to distributed block cache node;
Distributed block cache node receives store path and data encoding, and data encoding is converted to data storage version
And be sent in database and stored,
When user needs to inquire relevant financial transaction data, corresponding block cache node is found out by query unit, and
Data inquiry request is sent to the block cache node, distributed block cache node is searched corresponding according to data inquiry request
Storage version, and the specified inquiry port address of query unit is pushed to when finding corresponding storage version.
Before data input unit receives the data of financial transaction of input: client carries out typing by data input unit,
Data input unit sends key sequence and common parameter to client by safe lane, and client stores key sequence;
Financial institution's staff's work number and password login, which send typing to data input unit, requests;Data input unit is from public
A random number is selected in parameter at random, wherein only to known to data input unit before random number;Data input unit passes through
Random number is sent to client and stored it in client by safe lane;Financial institution staff logs off again
Work number and password are inputted to its client, client selectes a random number as its private key at random from common parameter, and counts
Calculate its public key;Data input unit verifies client public key by key sequence, is verified laggard row information typing.
The system platform of the financial big data processing method based on block chain includes: that data input unit, data are excellent
Change unit, Data Management Unit, distributed block cache node and query unit, the data input unit, data-optimized list
Member, Data Management Unit, distributed block cache node and query unit are according to being sequentially connected.
The present invention has the effect that the present invention improves the efficiency of authentication verification algorithm and agreement, to its security level
It compares, attack may be born by avoiding in finance data transmission, abandon traditional communication computer system, only need to be according to number
Error code rate can be made to be preferably minimized according to the transmission time that administrative unit is deployed to distributed block cache node, and also improved
The efficiency of secure communication while meeting financial system diversified demand, improves resource by distributed resource allocation method
Utilization rate, enormously simplify the experience of user data access.
Detailed description of the invention
Fig. 1 is present system structural schematic diagram.
Specific embodiment
Below in conjunction with specific embodiment, the present invention is further illustrated:
A kind of financial big data processing method based on block chain, comprising the following steps:
Data input unit receives the data of financial transaction of input, and data of financial transaction is sent to data-optimized list
Member;
Data-optimized unit receives data of financial transaction and optimizes to it, and provides financial transaction to Data Management Unit
Data storage path;
Data Management Unit receives data storage path and optimized data of financial transaction, and to by optimizing
Data of financial transaction afterwards is encoded, and data storage path and data encoding are sent to distributed block cache node;
Distributed block cache node receives store path and data encoding, and data encoding is converted to data storage version
And be sent in database and stored,
When user needs to inquire relevant financial transaction data, corresponding block cache node is found out by query unit, and
Data inquiry request is sent to the block cache node, distributed block cache node is searched corresponding according to data inquiry request
Storage version, and the specified inquiry port address of query unit is pushed to when finding corresponding storage version.
Before data input unit receives the data of financial transaction of input: client carries out typing by data input unit,
Data input unit sends key sequence and common parameter to client by safe lane, and client stores key sequence;
Financial institution's staff's work number and password login, which send typing to data input unit, requests;Data input unit is from public
A random number is selected in parameter at random, wherein only to known to data input unit before random number;Data input unit passes through
Random number is sent to client and stored it in client by safe lane;Financial institution staff logs off again
Work number and password are inputted to its client, client selectes a random number as its private key at random from common parameter, and counts
Calculate its public key;Data input unit verifies client public key by key sequence, is verified laggard row information typing.
It is described when data storage path and data encoding are sent to distributed block cache node, if receive financial number
It is I according to the distributed block cache node sum that store path and finance data encode, for distributed block cache node i ∈
{ 1,2 ..., I }, the size for needing to transmit finance data packet representated by each coding is dx, calculates the number of finance data coding
According to transmission rate vsiAre as follows:
Wherein, zyiIndicate channel gain;PiFor transimission power;σ2It is sampled value variance of the noise after over-sampling;liFor
The maximum data length that link layer can bear;wiFor the bit error rate of transmission finance data coding;stiFor the degree of strength of channel;
E indicates natural constant;Q-1(x) be function Q inverse function, whereinX is independent variable, and t is meaningless;
Then
Wherein, B is channel width, tiThe transmission time of distributed block cache node i is deployed to for Data Management Unit;
When distributed block cache node receives store path and data encoding, Data Management Unit is by adjusting tiTo reach minwiI.e.
It can.
When the distributed block cache node receives store path and data encoding, Data Management Unit passes through safety letter
The parameter that road sending time is stabbed and randomly selected gives distributed block cache node;Distributed block cache node checks data pipe
It manages unit and receives a message and if only if it includes one for distributed block cache node close enough current time
Timestamp;If it is not, distributed block cache node may think that Data Management Unit, there are Replay Attack, distributed blocks
The reception of cache node stopping finance data;If so, distributed block cache node utilizes its private key and Data Management Unit
Public key check both password equation it is whether equal;If unequal, distributed block cache node stops finance data
It receives;If equal, Data Management Unit selectes random parameter, and session key, distributed block cache node at random
The parameter stabbed and randomly selected by safe lane sending time is to Data Management Unit;Data Management Unit checks data management
Unit receives a message and if only if it includes the timestamps at a close enough current time for Data Management Unit;
If it is not, Data Management Unit may think that distributed block cache node there are Replay Attack, Data Management Unit stops gold
Melt the transmission of data;If so, Data Management Unit checks the password of the two using the public key of its private key and Data Management Unit
Whether equation is equal;If unequal, Data Management Unit stops the transmission of finance data;If equal, Data Management Unit is calculated
First arranging key is simultaneously sent to distributed block cache node by safe lane;Distributed block cache node calculates second
Whether arranging key simultaneously judges equal with the first arranging key;If unequal, secure key negotiation unsuccessfully stops finance data
Transmission;If equal, Data Management Unit and distributed block cache node have all reached a trust threshold, are negotiated with second
Key simultaneously judges to continue finance data transmission with the first arranging key.
The system platform of the financial big data processing method based on block chain includes: that data input unit, data are excellent
Change unit, Data Management Unit, distributed block cache node and query unit, the data input unit, data-optimized list
Member, Data Management Unit, distributed block cache node and query unit are according to being sequentially connected.