CN112202868B - Method for realizing block chain consensus protocol based on wireless air calculation - Google Patents

Method for realizing block chain consensus protocol based on wireless air calculation Download PDF

Info

Publication number
CN112202868B
CN112202868B CN202011030241.2A CN202011030241A CN112202868B CN 112202868 B CN112202868 B CN 112202868B CN 202011030241 A CN202011030241 A CN 202011030241A CN 112202868 B CN112202868 B CN 112202868B
Authority
CN
China
Prior art keywords
base station
consensus
linear superposition
nodes
terminal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202011030241.2A
Other languages
Chinese (zh)
Other versions
CN112202868A (en
Inventor
化存卿
顾鹏文龙
吴越
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Jiaotong University
Original Assignee
Shanghai Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Jiaotong University filed Critical Shanghai Jiaotong University
Priority to CN202011030241.2A priority Critical patent/CN112202868B/en
Publication of CN112202868A publication Critical patent/CN112202868A/en
Application granted granted Critical
Publication of CN112202868B publication Critical patent/CN112202868B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/06—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols the encryption apparatus using shift registers or memories for block-wise or stream coding, e.g. DES systems or RC4; Hash functions; Pseudorandom sequence generators
    • H04L9/0643—Hash functions, e.g. MD5, SHA, HMAC or f9 MAC
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
    • H04L9/3236—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials using cryptographic hash functions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/50—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols using hash chains, e.g. blockchains or hash trees

Abstract

A block chain consensus protocol implementation method based on wireless air calculation includes a base station and N wireless terminal nodes in wireless block chain network, and the base station and the terminal nodes communicate through wireless channel. The method comprises the following steps: step 1, at the beginning of each round of consensus, a base station broadcasts update request information to all terminal nodes; step 2, each terminal node packs a block, calculates and generates a block hash value and sends the block hash value to a base station; step 3, the base station receives the linear superposition signals of all the terminal nodes, processes the linear superposition signals through a calculation forwarding technology and broadcasts the linear superposition signals to all the participating consensus nodes; step 4, the terminal node adopts a two-stage Hash verification method to determine whether consensus is achieved; and 5, each terminal node sends confirmation information to the base station to complete the consensus in the round. The invention reduces the noise influence of transmission in a wireless channel by coding and wireless air computing technology, and realizes an efficient and reliable block chain consensus protocol.

Description

Method for realizing block chain consensus protocol based on wireless air calculation
Technical Field
The invention relates to the technical field of wireless network block verification and consensus, in particular to a method for realizing a block chain consensus protocol based on wireless aerial computing (AirComp).
Background
The block chain technology receives more and more attention in the field of data and network space security in recent years due to the characteristics of distributed type, high encryption safety and traceability and non-repudiation, and provides a brand new idea for network security control and intrusion detection. With the research and discussion going deeper and deeper, how to realize efficient and reliable consensus at the physical layer of the wireless network, the advantages of the blockchain technology are brought into play and the short board is avoided to become a problem which must be faced and solved at the present stage in the network space security field.
Jiangsu Heng is information technology Co., Ltd, 2016 filed a block chain data comparison and consensus method and applied for a patent (application number: 201611133352. X). The hash value of each transaction content is calculated and used as a number, and the hash value can be put into a Memekel-like tree according to a certain rule. Therefore, the difference between different blocks can be found out very quickly, the data transmission amount is reduced, and the time required by consensus is greatly reduced. In most of these consensus protocols, consensus on newly generated blocks relies on cross-validation of hash values between different users. However, in this process, the encoding, decoding and transmission of the hash value will involve a high computational and transmission overhead. Particularly, in a wireless network, hash values generated by different nodes need to be transmitted to a base station end through orthogonal wireless channels and then broadcasted to all nodes, the wireless resource utilization rate in the two stages is linearly increased along with the number of participating nodes, and high transmission delay is caused and the efficiency of a consensus protocol is influenced in a large-scale network scene.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a block chain consensus protocol and a block chain consensus device based on wireless over-the-air calculation aiming at a wireless block chain network consisting of a base station and N wireless terminal nodes, wherein the method comprises the following steps:
step 1, at the beginning of each round of consensus, a base station broadcasts update request information to all terminal nodes; the request message comprises the number sequence of the common identification rounds, the time stamp and the size information of the time window.
Step 2, each terminal node completes the block packaging according to the request sent by the base station, calculates and generates a block hash value and sends the block hash value to the base station;
step 3, the base station processes the received linear superposition signals of the hash value in a calculation forwarding mode and broadcasts the processed linear superposition signals to all nodes participating in consensus;
and 4, after the node receives all the linear superposition lambada of the Hash codes from the base station, restoring the original linear superposition by the corresponding coarse-grained lattice information
Figure BDA0002703407530000027
And determining whether consensus is achieved by adopting a two-stage Hash verification method;
and 5, if the consensus is achieved, the newly generated blocks are stored on each terminal node in a distributed mode, and each terminal node sends confirmation information to the base station to complete the consensus.
In the step 2, each terminal node completes block packing according to a request sent by the base station, calculates and generates a block hash value and sends the block hash value to the base station, and the specific steps are as follows:
(2-1) each terminal node extracts corresponding transactions in a transaction pool of each terminal node according to the time window information in the base station request, packages the transactions into a new block, adds a turn, a timestamp and a last block hash value to the head of the block, and calculates the hash value of the newly generated block;
(2-2) each terminal node completes the hash value coding of the newly generated block by adopting the following lattice coding technology:
first, a generation method of a k-dimensional lattice code in euclidean space is defined as follows:
C＝{uG mod p：u∈Zk}，
where G is a complete generator matrix and p is a prime number;
then, define ΛFIs fine-grained lattice, ΛCFor coarse-grained lattice, respectively passing through ΛF＝C+pZkAnd ΛC＝pZkIs given by
Figure BDA0002703407530000021
Finally, define the set of nested lattices as
Figure BDA0002703407530000022
Wherein
Figure BDA0002703407530000023
Is coarse grain size lattice ΛCAnd the base Voronoi region of the nested lattice set Λ is represented as:
Figure BDA0002703407530000024
mapping hash values to the generated k-dimensional lattice code set ΛCan realize the encoding of the hash value, and the encoding process can be expressed as information belonging to a limited field l
Figure BDA0002703407530000025
Mapping to a lattice code x in the codebook rangei。
And (2-3) controlling the transmitting power of the coded signals by all the terminal nodes according to the wireless channel state information between the terminal nodes and the base station so as to compensate the channel fading of the terminal nodes, and then simultaneously transmitting the terminal nodes to the base station by adopting the same channel.
In step 3, the base station receives the linear superposition signals of all terminal nodes, processes the linear superposition signals in a calculation forwarding manner, and broadcasts the linear superposition signals to all nodes participating in consensus, which is specifically as follows:
(3-1) the linear superposition signals y of the N terminal nodes received by the base station are as follows:
Figure BDA0002703407530000026
in the formula, xiAnd z is white gaussian noise for the coded signal sent by the terminal node i.
(3-2) the base station adopts the scaling factor vector alpha to scale the linear superposition signal y to obtain
Figure BDA00027034075300000313
(3-3) base station pair
Figure BDA00027034075300000314
Performing quantization processing, namely mapping the quantization processing to the nearest point in the fine granularity lattice to obtain an estimated value of the linear superposition signal:
Figure BDA0002703407530000031
from this quantization result, an estimate of the linear superposition signal can be obtained:
Figure BDA0002703407530000032
wherein a isiIs the ith element in the vector alpha.
(3-4) superimposing noiseless linearity by modulo operation
Figure BDA00027034075300000312
Mapping back to coarse-grained lattice ΛCIn the shaping region:
Figure BDA0002703407530000033
where λ is the resulting linear superposition.
(3-5) the base station compares the obtained lambda, and
Figure BDA00027034075300000315
and broadcasting the corresponding coarse-grained lattice information to all nodes.
In step 4, after receiving the linear superposition λ of all hash vectors from the base station, the node restores the original linear superposition by the coarse-grained lattice information corresponding to the linear superposition λ
Figure BDA0002703407530000034
And a two-stage Hash verification method without decoding is adopted to determine whether consensus is achieved, and the specific steps are as follows:
(4-1) each terminal node linearly superposes the restores
Figure BDA0002703407530000035
Encoding x projected onto self block hashiCan obtain
Figure BDA0002703407530000036
At xiModulo | t of the up-projection vectoriI and
Figure BDA0002703407530000037
and xiCosine value of angle cos thetaiThe values, namely:
Figure BDA0002703407530000038
Figure BDA0002703407530000039
(4-2) training a binary classifier pair to obtain a binary group v by each terminal node ii＝(|ti|，cosθi) And (6) classifying. If v isiIf the condition of the consensus establishment is met, the node i obtains the label ui1. Otherwise, it will get tag ui＝-1。
(4-3) any tag u obtained according to the classification result of (4-2)iThe node i which is 1 encodes the lower four bits of the hash value of the node i according to the steps (2-2) to (2-3) to obtain
Figure BDA00027034075300000310
And sends it to the base station for a second round of consensus and in a first phase a tag u is obtainediThe node-1 does not participate;
(4-4) base station has received all codes x'iAfter superimposing the signals, a new linear superposition λ' is calculated according to the algorithm given in (3-1) to (3-5) and broadcast back to all nodes.
(4-5) all the tags u classified in the step (4-3)iThe node 1 repeats the steps from (4-1) to (4-2) according to the received new linear superposition λ ', and obtains the label 1 in the round if more than N/2 nodes exist, and the newly generated linear superposition λ' is used for the second round of classification
Figure BDA00027034075300000311
The scalar projections on these nodes are all greater than a given threshold, and a successful consensus is determined.
Compared with the prior art, the invention has the following technical effects:
and the lattice coding and wireless air computing technology are adopted, so that compared with a consensus algorithm of a network layer, the communication overhead, the computing complexity and the consensus reliability are greatly improved.
Firstly, lattice coding and air computing technology are adopted, and the consensus protocol provided by the inventor breakthroughs the realization of physical layer consensus. Compared with the traditional network layer consensus, the consensus algorithm provided by the invention has great progress in the transmission and operation complexity aspects: for the conventional consensus protocol, the transmission complexity is o (N) because all nodes should communicate with the base station using orthogonal wireless channels, and for the hash verification, the calculation complexity is o (N)2). In our proposed consensus protocol, the transmission complexity is reduced to o (1), and the training of the classifier can be done off-line, thus not incurring much overhead for consensus.
For the consensus reliability, due to the high-dimensional geometric characteristic of lattice coding, the noise influence caused by the transmission of the coding in a wireless channel can be reduced through quantization, and the influence of error codes on the consensus reliability in the transmission process is greatly reduced.
Drawings
Fig. 1 is a diagram of a wireless block chain system consensus.
Fig. 2 is a diagram of a consensus protocol process.
Fig. 3 is a schematic diagram of a process of operation forwarding of a base station.
Fig. 4 is a flow chart of the protocol ensemble algorithm.
Fig. 5 is a graph of quantization accuracy for two codebooks, p-3 and p-5, under different channel conditions.
FIG. 6 is a graph of the detection rates of three SVM classifiers with different numbers of consistent hashes.
FIG. 7 is a consensus error rate graph with different numbers of consistent hashes.
Detailed Description
The invention is further explained below with reference to the figures and examples, without limiting the scope of protection of the invention.
A wireless block-chain network as shown in fig. 1, in which N wireless terminal nodes communicate with one base station through wireless channels. The nodes maintain a same block chain together, and blocks are generated independently by each terminal node periodically based on a consensus protocol and are stored on all the terminal nodes in a distributed mode.
As shown in fig. 2, the overall consensus process is divided into 5 steps:
step 1: at the beginning of each round of consensus, the base station broadcasts an update request to all nodes. The request message includes the number of the common identification rounds, the time stamp and the size information of the time window.
Step 2: and the node finishes packaging the blocks according to the request sent by the base station, calculates and generates a block hash value and sends the block hash value to the base station. The specific process is as follows:
(2-1) each terminal node independently extracts corresponding transactions from the transaction pool according to the time window information in the base station request, packs the transactions into blocks, adds turns, time stamps and the hash value of the previous block into the head of each block, and calculates the 128-bit hash value of the newly generated block by adopting an MD5 algorithm.
(2-2) each terminal node encodes the hash value of the newly generated block by using a lattice encoding technique as follows:
first, a generation method of a k-dimensional lattice code in euclidean space is defined as follows:
C＝{uGmodp：u∈Zk}，
where G is a complete generator matrix and p is a prime number;
then, define ΛFIs fine-grained lattice, ΛCFor coarse-grained lattice, respectively passing through ΛF＝C+pZkAnd ΛC＝pZkIs given by
Figure BDA0002703407530000051
Finally, define the set of nested lattices as
Figure BDA0002703407530000052
Wherein
Figure BDA0002703407530000056
Is coarse grain size lattice ΛCOf a basic VoronoiThe region, and the base Voronoi region of nested lattice set Λ is represented as:
Figure BDA0002703407530000054
by mapping the hash value with the code word in the generated k-dimensional lattice code set lambda, the encoding of the hash value can be realized, and the encoding process can be expressed as information belonging to l finite fields
Figure BDA0002703407530000055
Mapping to a lattice code x in the codebook rangei. In order to reduce the number of collinear lattice codes and improve the detection accuracy, lattice codes with a relatively close modulus of a coding vector are adopted. Thus, the number of available codes in its codebook space can be expressed as:
Mp＝pk-(p-2)k
each terminal node divides the 128-bit hash value into 32 groups, each group has 4 bits, represents a 16-system number, and can pass through 2-dimension G-I2The integer lattice code of the generator matrix modulo-5 is represented, with the number of available codes in the codebook space: mp＝p2-(p-2)216, encoding of a set of 4-bit data can thus be achieved.
And (2-3) controlling the transmitting power of the coded signals by all the terminal nodes according to the wireless channel state information between the terminal nodes and the base station so as to compensate the channel fading of the terminal nodes, and then simultaneously transmitting the terminal nodes to the base station by adopting the same channel.
And step 3: and (3-1) to (3-5) are executed by the base station, the received linear superposition signals of all the terminal nodes are quantized in a calculation and forwarding mode as shown in fig. 3, and then are broadcast to all the nodes participating in the consensus.
And 4, step 4: after the node receives the linear superposition lambada of all the Hash vectors from the base station, the original linear superposition is restored through the corresponding coarse granularity lattice information
Figure BDA0002703407530000057
And a two-stage Hash verification method without decoding is adopted to determine whether consensus is achieved, and the specific steps are as follows.
(4-1) each terminal node i linearly superposes the restored
Figure BDA00027034075300000612
Encoding x projected onto self block hashiCan obtain
Figure BDA00027034075300000611
At xiModulo | t of the up-projection vectoriI and
Figure BDA00027034075300000610
and xiCosine value of angle cos thetaiThe values, namely:
Figure BDA0002703407530000061
Figure BDA0002703407530000062
(4-2) each terminal node i adopts a Support Vector Machine (SVM) classifier based on a kernel function:
fw，b(v)＝k(w，v)+b
for the obtained two-tuple vi＝(|ti|，cosθi) Classifying to obtain binary group viLinear divisible. Where k (,) is a kernel function and w and b are variables.
Wherein, we adopt the following three different SVM kernel functions to ensure the classification precision:
Figure BDA0002703407530000063
Polynomial：k(vi，vj)＝(<vi，vj>+h)d
Figure BDA0002703407530000064
RBF：
Figure BDA0002703407530000065
Figure BDA0002703407530000066
MLP：
Figure BDA0002703407530000067
hyperplane f from trainingw，b(v) And the selected support vector in the training set, judging the binary group viAnd the hyperplane fw，b(v) If v is a positional relationship ofiIf the condition of the consensus establishment is met, the node i obtains the label ui1. Otherwise, it will get tag ui＝-1；
(4-3) any tag u obtained according to the classification result of (4-2)iThe node i which is 1 encodes the lower four bits of the hash value of the node i according to the steps (2-2) to (2-3) to obtain
Figure BDA0002703407530000068
And sends it to the base station for a second round of consensus and in a first phase a tag u is obtainediThe node-1 does not participate;
(4-4) base station has received all codes x'iAfter superimposing the signals, a new linear superposition λ' is calculated according to the algorithm given in (3-1) to (3-5) and broadcast back to all nodes.
(4-5) all the tags u classified in the step (4-3)iThe node 1 repeats the steps from (4-1) to (4-2) according to the received new linear superposition λ ', and obtains the label 1 in the round if more than N/2 nodes exist, and the newly generated linear superposition λ' is used for the second round of classification
Figure BDA0002703407530000069
The scalar projections on these nodes are all greater than a given threshold, and a successful consensus is determined.
And 5: if the consensus is achieved, the newly generated blocks are stored in a distributed manner on each terminal node, and each node sends confirmation information to the base station to complete the current round of consensus.
The algorithm flow of the above 5 steps is shown in fig. 4.
Fig. 5 shows the quantization accuracy using two different codebooks, p-3 and p-5. It can be seen that when the channel quality is low (SNR is 0), the codebook with p 3 has higher quantization accuracy due to the larger angle between two adjacent codevectors. With the improvement of the SNR, the quantization rate gap between the two coding schemes gradually decreases, and when the SNR is equal to 20, the quantization accuracy of the two coding schemes can reach 100%, which greatly reduces the risk of transmission error in a wireless environment.
The simulation results of the detection rate are shown in fig. 6, in which each value is calculated from 1000 simulation results. It can be seen that as the total number of users remains the same, the detection rate of the RBF classifier and the Polynomial classifier increases as the number of consistent hashes increases. With 15 of the 21 hashes being identical, the two classifiers can achieve almost 100% accuracy. Meanwhile, it can be seen that the fault tolerance of the hash verification in the consensus protocol can be greatly improved by using the geometric features of lattice codes, and especially under the condition of poor channel conditions (SNR 0 and SNR 10), the detection rates of the RBF and the multinomial classifier are far greater than the quantization rate of the code words. This indicates that quantization errors caused by transmitted noise can be corrected if they do not cause large deviations in the vector space.
Finally, we evaluated the impact of our proposed two-stage hash validation on consensus error rate. As shown in fig. 7, consensus error rates with different numbers of consistent hashes are given, where each point is obtained based on 500 simulations. It can be observed that the consensus protocol reaches its highest error rate when the number of consistent hashes approaches the threshold for achieving consensus achievement (n 11 in our case). At this time, the detection error of the SVM classifier has the greatest influence on the consensus protocol. In cases where consensus is clearly achieved (n > 12) or clearly not (n < 10), the success rate of our proposed protocol reaches 100%. Furthermore, the consensus error rate can be found to be much lower compared to the detection rate of the SVM classifier presented in FIG. 6. This is because some misclassifications made by the SVM classifier in the first stage can be corrected by the second stage of hash verification, so that the final consensus achievement is not affected.

Claims (5)

1. A method for implementing a block chain consensus protocol based on wireless over-the-air computation, comprising N wireless terminal nodes and a base station in a wireless block chain network, communicating via a wireless channel, the method comprising the steps of:
step 1, at the beginning of each round of consensus, a base station broadcasts update request information to all terminal nodes;
step 2, each terminal node completes the block packaging according to the request sent by the base station, calculates and generates a block hash value and sends the block hash value to the base station, and the specific steps are as follows:
(2-1) each terminal node extracts corresponding transactions in a transaction pool of each terminal node according to the time window information in the base station request, packages the transactions into a new block, adds a turn, a timestamp and a last block hash value to the head of the block, and calculates the hash value of the newly generated block;
(2-2) each terminal node completes the hash value coding of the newly generated block by adopting the following lattice coding technology:
first, a generation method of a k-dimensional lattice code in euclidean space is defined as follows:
C＝{uG mod p:u∈Zk},
where G is a complete generator matrix and p is a prime number;
then, define ΛFIs fine-grained lattice, ΛCFor coarse-grained lattice, respectively passing through ΛF＝C+pZkAnd ΛC＝pZkIs given by
Figure FDA0003225523400000014
Finally, defineThe set of nested lattices is
Figure FDA0003225523400000015
Wherein
Figure FDA0003225523400000016
Is coarse grain size lattice ΛCAnd the base Voronoi region of the nested lattice set Λ is represented as:
Figure FDA0003225523400000011
the encoding of the hash value is realized by mapping the hash value to one code word in the generated k-dimensional lattice code set Λ, and the encoding process can be expressed as information belonging to l finite fields
Figure FDA0003225523400000012
Mapping to a lattice code x in the codebook rangei；
(2-3) all terminal nodes control the transmitting power of the coded signals according to the wireless channel state information between the terminal nodes and the base station so as to compensate the channel fading of the terminal nodes, and then the terminal nodes adopt the same channel and transmit the same channel to the base station at the same time;
step 3, the base station quantizes the received linear superposition signal of the hash value in a computer-and-forward mode, obtains a linear superposition lambda in a shaping region (shaping region) through modular operation and broadcasts the linear superposition lambda to all nodes participating in consensus;
and 4, after the node receives the linear superposition lambada of all the Hash vectors from the base station, restoring the original linear superposition by the corresponding coarse granularity lattice information
Figure FDA0003225523400000013
And determining whether consensus is achieved by adopting a two-stage Hash verification method;
and 5, if the consensus is achieved, the newly generated blocks are stored on each terminal node in a distributed mode, and each terminal node sends confirmation information to the base station to complete the consensus.
2. The method according to claim 1, wherein in step 3, the base station receives the linear superposition signals of all terminal nodes, processes the linear superposition signals in a calculation forwarding manner, and broadcasts the linear superposition signals to all nodes participating in the consensus, specifically as follows:
(3-1) the linear superposition signals y of the N terminal nodes received by the base station are as follows:
Figure FDA0003225523400000021
in the formula, xiA coded signal sent by a terminal node i, wherein z is Gaussian white noise;
(3-2) the base station adopts the scaling factor vector alpha to scale the linear superposition signal y to obtain
Figure FDA0003225523400000022
(3-3) base station pair
Figure FDA0003225523400000023
Performing quantization processing, namely mapping the quantization processing to the nearest point in the fine granularity lattice to obtain an estimated value of the linear superposition signal:
Figure FDA0003225523400000024
from this quantization result, an estimate of the linear superposition signal can be obtained:
Figure FDA0003225523400000025
wherein a isiIs the ith element in the vector α;
(3-4) superimposing noiseless linearity by modulo operation
Figure FDA0003225523400000026
Mapping back to coarse-grained lattice ΛCIn the shaping region of (a):
Figure FDA0003225523400000027
wherein λ is the resulting linear superposition;
(3-5) the base station compares the obtained lambda, and
Figure FDA0003225523400000028
and broadcasting the corresponding coarse-grained lattice information to all nodes.
3. The method of claim 1, wherein in step 4, after the terminal node receives the linear superposition λ of all hash vectors from the base station, the terminal node restores the original linear superposition λ through its corresponding coarse-grained lattice information
Figure FDA0003225523400000029
And a two-stage Hash verification method without decoding is adopted to determine whether consensus is achieved, and the specific steps are as follows:
(4-1) each terminal node linearly superposes the restores
Figure FDA00032255234000000210
Encoding x projected onto self block hashiCan obtain
Figure FDA00032255234000000211
At xiModulo | t of the up-projection vectoriI and
Figure FDA00032255234000000212
and xiCosine value of angle cos thetaiThe values, namely:
Figure FDA00032255234000000213
Figure FDA00032255234000000214
(4-2) training a binary classifier pair to obtain a binary group v by each terminal node ii＝(|ti|,cosθi) Classifying; if v isiIf the condition of the consensus establishment is met, the node i obtains the label ui1 is ═ 1; otherwise, it will get tag ui＝-1；
(4-3) any tag u obtained according to the classification result of (4-2)iThe node i which is 1 encodes the lower four bits of the hash value of the node i according to the steps (2-2) to (2-3) to obtain
Figure FDA00032255234000000215
And sends it to the base station for a second round of consensus and in a first phase a tag u is obtainediThe node-1 does not participate;
(4-4) base station has received all codes x'iAfter superimposing the signals, a new linear superposition λ' is calculated according to the algorithms given in (3-1) to (3-5) and broadcast back to all nodes;
(4-5) all the tags u classified in the step (4-3)iThe node 1 repeats the steps from (4-1) to (4-2) according to the received new linear superposition λ ', and obtains the label 1 in the round if more than N/2 nodes exist, and the newly generated linear superposition λ' is used for the second round of classification
Figure FDA0003225523400000031
The scalar projections on these nodes are all greater than a given threshold, and a successful consensus is determined.
4. The method according to claim 3, wherein in the step (4-2), each terminal node i employs a Support Vector Machine (SVM) classifier based on kernel function:
fw,b(v)＝k(w,v)+b
for the obtained two-tuple vi＝(|ti|,cosθi) Classifying to obtain binary group viLinear separable, where k (,) is a kernel function, w and b are variables;
hyperplane f from trainingw,b(v) And the selected support vector in the training set, judging the binary group viAnd the hyperplane fw,b(v) If v is a positional relationship ofiIf the condition of the consensus establishment is met, the node i obtains the label ui1 is ═ 1; otherwise, it will get tag ui＝-1。
5. The method of claim 1, wherein the request message comprises a number of common identification rounds, a timestamp, and a time window size information.
CN202011030241.2A 2020-09-27 2020-09-27 Method for realizing block chain consensus protocol based on wireless air calculation Active CN112202868B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011030241.2A CN112202868B (en) 2020-09-27 2020-09-27 Method for realizing block chain consensus protocol based on wireless air calculation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011030241.2A CN112202868B (en) 2020-09-27 2020-09-27 Method for realizing block chain consensus protocol based on wireless air calculation

Publications (2)

Publication Number Publication Date
CN112202868A CN112202868A (en) 2021-01-08
CN112202868B true CN112202868B (en) 2021-11-30

Family

ID=74008375

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011030241.2A Active CN112202868B (en) 2020-09-27 2020-09-27 Method for realizing block chain consensus protocol based on wireless air calculation

Country Status (1)

Country Link
CN (1) CN112202868B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114979167A (en) * 2022-01-10 2022-08-30 昆明理工大学 Consensus system, method and device considering storage optimization

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107276765A (en) * 2017-07-04 2017-10-20 中国联合网络通信集团有限公司 The processing method and processing device known together in block chain
CN108965247A (en) * 2018-06-04 2018-12-07 上海交通大学 A kind of threat information exchange shared system and method based on block chain
CN110543515A (en) * 2019-07-26 2019-12-06 天津理工大学 Block chain system realized based on LORA communication technology and working method
CN110913501A (en) * 2019-11-08 2020-03-24 山东大学 Consensus method suitable for wireless block chain network

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107046542B (en) * 2017-04-24 2020-04-14 杭州云象网络技术有限公司 Method for realizing consensus verification by adopting hardware at network level
US20170285720A1 (en) * 2017-05-07 2017-10-05 Jigar Jayesh Shah Method and system for mitigating transmission congestion via distributed computing and blockchain technology
CN107450981B (en) * 2017-05-31 2020-04-24 创新先进技术有限公司 Block chain consensus method and equipment
CN109792451B (en) * 2018-08-22 2022-11-18 袁振南 Communication channel encryption, decryption and establishment method and device, memory and terminal
CN110022216B (en) * 2019-02-18 2022-02-01 西安链融科技有限公司 Efficient asynchronous Byzantine consensus method with low communication complexity and network communication platform

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107276765A (en) * 2017-07-04 2017-10-20 中国联合网络通信集团有限公司 The processing method and processing device known together in block chain
CN108965247A (en) * 2018-06-04 2018-12-07 上海交通大学 A kind of threat information exchange shared system and method based on block chain
CN110543515A (en) * 2019-07-26 2019-12-06 天津理工大学 Block chain system realized based on LORA communication technology and working method
CN110913501A (en) * 2019-11-08 2020-03-24 山东大学 Consensus method suitable for wireless block chain network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
"Consensus in Wireless Blockchain System";Qiang Xu等;《Springer Nat》;20200910;全文 *
"物联网完全检测与防护机制综述";化存卿;《上海交通大学学报》;20181031;第52卷(第10期);全文 *

Also Published As

Publication number Publication date
CN112202868A (en) 2021-01-08

Similar Documents

Publication Publication Date Title
US20230155637A1 (en) Mitigation of transmission errors of quantized channel state information feedback in multi antenna systems
CN107517095B (en) Polarization code coding and decoding method for non-uniform sectional check
US20210319286A1 (en) Joint source channel coding for noisy channels using neural networks
CN104780022B (en) Safe transmission method of physical layer and system based on channel coding matrix dynamic change
CN101019326B (en) File download and streaming system
CN1795679B (en) Method, apparatus and system for encoding and decoding side information for multimedia transmission
US20210351863A1 (en) Joint source channel coding based on channel capacity using neural networks
US20060059403A1 (en) Quantum key distribution method and communication device
CN101459430B (en) Encoding method and apparatus for low density generation matrix code
CN109347604B (en) Multi-hop network communication method and system based on batched sparse codes
WO2020035684A1 (en) Joint source channel coding of information sources using neural networks
CN112202868B (en) Method for realizing block chain consensus protocol based on wireless air calculation
US10367600B2 (en) Forward error correction with contrast coding
CN107332570B (en) Polarization code coding method of segmented cascade Hash sequence
CN108712235B (en) One-way feedback-free transmission method
CN104065463A (en) Method And Apparatus For Processing Piggybacked Acknowledgemet/non-acknowledgement Field
CN110582955B (en) Coding device for polarization code
CN107276596B (en) Polarization code decoding method based on segmented Hash sequence
Karat et al. Optimal error correcting delivery scheme for an optimal coded caching scheme with small buffers
CN111246460B (en) Low-complexity and low-time-delay secure transmission method
GB2508435A (en) Decoding of data received from one source by several receivers
CN103081376B (en) For keeping the system and method for neighborhood in code
Shahani et al. Better Performance of BER in the System Using Block Coding Technique for Optical Communication
Longqing et al. Recovering the parameters of an LDPC code from noisy intercepted sequences
Ramanathan et al. Generalised and channel independent SVM based robust decoders for wireless applications

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant