2.2.1

Data layer

The data layer has three characteristics, which are tamper proof, full backup and full equality (data, authority and code) To achieve this feature, we rely on the chain structure which contains prev hash, nonce and tx, as shown in Figure 1:

Figure 1.

Chain structure

2.2.2

Data structure of data layer

The data layers are structured as blocks, block heads, and block bodies, respectively. As shown in table1.

Table 1.

Data structure of data layer

The container data structure that aggregates the transaction information contained in the public account book (blockchain), Including block header and a body. The block header + block body < = 1m, the block header is 80 bytes, and the block body indicates that each block contains 2000 transactions, with an average of at least 250 bytes per transaction. Thus, the block containing the full transaction is 4000 times the block header size.

The digital fingerprint obtained by hashing the head of the data block twice with the sha256 algorithm produces a fast hash value. It is unique, computable and storable. It is not included in the data structure of the block, but can be stored in an independent database table. The height of the block comes from the bitcoin network. Nodes dynamically identify the location of blocks in the network (also known as block height), which is not unique in a short time.

A complete Bitcoin node saves the local blockchain copy [3] from the created block, and the block local blockchain [4] copy will be constantly updated to expand blockchain. Upon receiving incoming blocks from the network, the node checks these blocks, and then connects them to the existing blockchain network. In the case of new blocks, the node will find the hash value of its parent block in the field of “parent block hash value”.

2.2.3

The timestamp of the data layer

The timestamp of the data layer refers to the local timestamp of each node. The block generation frequency is 10 minutes. When the timestamp of the data layer is greater than the median value of the timestamp of the first 7 blocks and less than 2 hours after the average time of the system, the block will be rejected.

2.2.4

Transaction records of data layer

Figure 2 shows the transaction records of the data layer. The private key, public key and wallet address of the data layer bitcoin system uses the elliptic curve signature algorithm. The private key of the algorithm is composed of 32 bytes random numbers. The public key can be calculated through the private key. The public key gets the bitcoin address through a series of hash algorithms and coding algorithms, and the address can also be understood as the summary (hash) of the public key.

Figure 2.

Transaction records of data layer

2.2.5

Maintaining the Integrity of the Specifications

The template is used to format your paper and style the text. All margins, column widths, line spaces, and text fonts are prescribed; please do not alter them. You may note peculiarities. For example, the head margin in this template measures proportionately more than is customary. This measurement and others are deliberate, using specifications that anticipate your paper as one part of the entire proceedings, and not as an independent document. Please do not revise any of the current designations.

The broadcast block transaction data includes three parts: original data, public key and the transferor’s public key. The original data includes the transfer amount and the other party’s wallet address, signature, and signature of the original data using the transferor’s private key, The public key is a publicly visible key in the whole network. It is used to verify the information and encrypt the data generated according to the private key. It can be verified by the data signed by the private key.

The asymmetric encryption algorithm of the data layer uses the public key and the private key respectively in the process of “encryption” and “decryption”, including

Asymmetric encryption, elliptic curve encryption algorithm and Merkle tree[2] of data layer. A Merkle tree is a binary hash tree used for fast recursion and large scale data integrity checking purposes. This tree is called balanced.

In the blockchain[5],all blocks consist of all transactions generated from the block and are represented by Merkle trees., as shown the Formula 1 and 2:

The proof of the existence of a specific transaction in the block requires only the computation of log2 (n) hashing. 16 transactions can be proved by 4 hashes + Merkle tree root. For 65535 transactions, only 16 transactions are required, which can be seen in Figure 3.

Figure 3.

The broadcast block transaction data

As shown in the above figure, if we want to determine the accuracy of the transaction H (k), the proof method is that the detection server a has only the block header and no block, but it can communicate with all 2 nodes to obtain the hash of the Merkle tree. and then to detect server B. The Merkle path (H (L), H (ij), H (mnop), H (abcdeffgh)) can be determined. It can be verified with the last requested data from the complete access list (H (L), H (ij), H (mnop), H (abcdeffgh)), perform several hash operations, and finally perform hash comparison with the Merkle root. By using the same original information of the hash function, the same summary information can always be obtained by using the same hash. Any minor changes to the original information hash out unidentifiable summary information. The three features of the original information cannot be inferred from the summary information and a short summary information can be obtained.

2.3.1

P2P technology of network layer.

P2P networking technology was early applied in P2P download software such as BT, which means that blockchain[6] has automatic networking function and supports TCP, UDP and other communication protocols.

2.4.1

Data desensitization module

Data desensitization refers to the use of masking, generalization, encryption and other technologies to eliminate the identifiable sensitive information contained in the original data, but at the same time, it needs to retain the data characteristics in a certain use environment [6]. This scheme will first adopt data desensitization strategy to protect sensitive information in medical data.

2.4.2

Digital signature module

SM9 is an identification cipher algorithm constructed by bilinear pairs on elliptic curves. Its main body includes numbers, word signature algorithm, public key encryption algorithm, and etc. This scheme uses SM9 identification algorithm in the digital signature module to perform the digital signature operation. The algorithm generates public-private key pairs based on the user’s ID. In this mode, we no longer rely on the keystore and Ca in the traditional PKI system to issue and verify certificates. Therefore, using SM9 identification algorithm can greatly reduce the resource overhead in computing and storage, which is in line with application requirements of medical scenarios.

2.4.3

Data encryption and decryption module

The scheme designed in this paper adopts SM9 identification national secret algorithm to sign and count the medical data first. According to the signature verification, the data source is safe and reliable. Since the blockchain system[8] is open and transparent,the data stored on the blockchain can be viewed by all the nodes in the network, and it is stored directly. In essence, text messages don’t solve privacy concerns about medical data, so it needs to be stored on the chain. The stored medical data and relevant personal privacy information are encrypted, so that they can be safely stored in the form of ciphertext in blockchain.

2.4.4

Implementation of digital signature module

This scheme adopts SM9 identification algorithm as digital signature algorithm. The significance of using digital signature in this paper is to ensure that medical data has a true and reliable source. Only when the source is reliable, can the sharing of medical data have practical significance, so the use of digital signature technology in this scheme is essential. In this scenario, doctors can generate a pair of public and private keys through their own identity, and use their own private keys to digitally sign the medical data containing patient information and diagnosis and treatment information. The data receiver can verify the signature with the doctor’s public key. The verified medical data can be regarded as a reliable source, and files that can be encrypted and stored in the distributed file system of IPFs.

2.4.5

Implementation of data encryption and decryption module

In this scheme, IPFs from distributed file systems are adopted as a medical record storage aid. Since the data stored on the blockchain [9] is open and transparent, themultiple nodes can participate in the backup. If data is directly stored in the distributed file storage system as plain text, and the hash value returned by data stored in IPFs is stored in blockchain network, when the blockchain data is synchronized with other nodes, data address becomes open and transparent. If the illegal user finds the corresponding information through hash, it will inevitably lead to the privacy disclosure of patients. Therefore, medical data can only be effectively stored in distributed storage system after encryption.

2.4.6

Implementation of distributed storage module for medical data

As a major component of blockchain technology [10], the choice of specific implementation scheme needs careful consideration. In this experiment, the delegated proof of stake (dpos) algorithm is selected. By voting, a certain number of proxy nodes are selected, and these nodes obtain the accounting right of the block. In the process of experimental testing, the system preset 10 nodes to meet the experimental needs, and took port number 8881 to port number 8890 as the node identification. Through the voting mechanism, the nodes that get the top three votes are selected as accounting nodes, which are responsible for uploading medical data and generating blocks. Other nodes are responsible for consensus verification, and all election results will be subject to consensus verification.