A wireless sensor network multi-period data fusion method and system

Abstract

The invention discloses a multi-time period data fusion method for a wireless sensor network, comprising: clustering the sensor network, each cluster including a cluster head node and a common node, obtaining data respectively received by the common node in multiple time periods, encrypting and hiding the received data by adopting an improved Paillier homomorphic encryption system, signing the data received in different time periods by adopting an identity-based signature method to obtain signature data, sending the hidden data and the signature data to the cluster head node, each cluster head node respectively aggregates the hidden data and signature data of the common nodes in the cluster, and sends them to a base station, the base station recovers the aggregated data hidden in multiple time periods of all sensors, and verifies whether the recovered data is tampered with according to the aggregated signature data. The method reduces the computational complexity and reduces the redundancy of the data.

Description

Technical Field

[0001] The present invention relates to the technical field of sensor network data fusion, and more specifically to a method and system for multi-period data fusion of a wireless sensor network. Background Art

[0002] Wireless sensor networks (WSNs) are constructed by a large number of sensor nodes in a wireless, multi-hop manner. With the rapid development of the Internet of Things, wireless sensor networks are more widely used in agricultural monitoring, environmental monitoring, intelligent transportation, smart homes, medical monitoring, logistics management, military and other fields. Since sensor nodes are limited by computing, storage and communication, it is difficult to transmit a large amount of sensor data. Therefore, using data fusion technology for data transmission can greatly reduce the amount of data transmission in the network, reduce node energy consumption, and extend the life of the entire network. In some fields with high security requirements (such as military and public health fields), it is not only required to achieve aggregation of collected data during the data fusion process, but also to verify the integrity of the collected data and protect the privacy of the data.

[0003] Existing privacy-protected wireless sensor network data fusion methods are mainly based on hop-by-hop encryption, privacy homomorphism, secure multi-party computing, data slicing and data perturbation technologies. The hop-by-hop encryption method uses hop-by-hop encryption and decryption to protect privacy, which requires a large computational overhead. The privacy protection method based on privacy homomorphism achieves the hiding of sensor data through end-to-end encryption, which can effectively resist most attacks on data during transmission, avoid encryption and decryption operations during aggregation, and provide efficient network data aggregation. The privacy protection method based on privacy homomorphism has obvious advantages over the hop-by-hop encryption method, and its research has become a trend. However, since the encryption strength of the homomorphic addition method mainly depends on the complexity of the key, in order to pursue data security, it is inevitable to sacrifice some network energy consumption. The traditional privacy protection scheme based on secure multi-party computing has a large amount of computation, so it is not used in resource-constrained wireless sensor networks. However, in recent years, some improved secure multi-party computing schemes have been proposed to reduce the complexity and energy consumption of traditional secure multi-party computing to a certain extent. In order to support data privacy and data integrity verification, the data slicing method needs to generate a large amount of redundant data, but the redundant data brings additional communication overhead. The data perturbation method is a non-encrypted privacy protection method. It hides data by modifying the true value of the data, thereby achieving the purpose of preventing attackers from obtaining the real data without encrypting the data. The data perturbation method can greatly reduce the amount of data transmission in the network without affecting the confidentiality of the data, thereby saving network energy consumption.

[0004] In most existing wireless sensor network data fusion schemes, after one round of data fusion, the base station can only obtain the fused data within a certain time period in a certain area. In order to obtain the fused data of multiple time periods in a certain area, multiple rounds of data fusion are required, which will greatly increase the consumption of network resources. Therefore, it is very necessary to study a data fusion method that can obtain different fused data of multiple time periods in a certain area after only one round of data fusion and meet the requirements of integrity verification and data privacy protection. Summary of the invention

[0005] The embodiment of the present invention provides a method and system for multi-period data fusion in a wireless sensor network, comprising: clustering the sensor network, each cluster including a cluster head node and a common node;

[0006] Obtain the data received by ordinary nodes in multiple time periods, use the improved Paillier homomorphic encryption system to encrypt and hide the received data, use the identity-based signature method to sign the data received in different time periods to obtain signature data, and send the hidden data and signature data to the cluster head node;

[0007] Each cluster head node aggregates the hidden data and signature data of ordinary nodes in the cluster and sends them to the base station. The base station recovers the aggregated data hidden in multiple time periods of all sensors and verifies whether the recovered data has been tampered with based on the aggregated signature data.

[0008] The improved Paillier homomorphic encryption system is used to encrypt the received data to obtain the private data, which specifically includes:

[0009] Using formula The received data is encrypted, wherein: represents a super-increasing sequence, represents the data received by the sensor node, S i,j represents the identity-based private key of each network node, g represents the generator, q represents a large prime number, and mod represents the modulus operator;

[0010] The identity-based signature method for signing data received in different time periods specifically includes:

[0011] Calculate P i,j =H1(ID i,j ), calculate P β =H2(β)∈G1, where P i,j Represents the unique ID assigned to each sensor network node i,j The corresponding value of Hash function H1, P β The corresponding value of the Hash function H2 representing the common state parameter β;

[0012] For time periods t1, t1, …, t w The data received by the sensor node is signed to obtain the signature data.

[0013]

[0014] Wherein, w represents the time period.

[0015] Furthermore, before clustering the sensor network, the sensor network parameters should be selected, which include:

[0016] Choose a network security parameter k∈Z q , choose a large prime number q>2 k , generate two additive groups G1 and multiplicative groups G2 of order q, a bilinear pairing map e:G1×G1→G2, and select two secure hash functions H1,H2:{0,1} * →G1, select a generator P from group G1, and select a random number s∈Z q is the system master key, Q = sP is the system public key, and a unique ID is assigned to each sensor network node i,j ∈{0,1} * , ID i,j Represents the unique identity of the jth network node in the i-th cluster, and calculates the identity-based private key S of each network node i,j =sP i,j , choose a common state parameter β∈{0,1} * , calculate P β =H2(β)∈G1, calculate T=P β P, choose a generator Select a super-increasing sequence in, is a large prime number, each Length And meet Where N represents the total number of nodes in the sensor network, and d represents the maximum value of the sensor data.

[0017] Next, after the sensor network parameters are selected and before the ordinary nodes receive data, the network sensor node parameters are stored: ID i,j , S i,j , H1, H2, β, g, q; store the private keys and parameters of all base station nodes: k, q, e, G1, G2, H1, H2, P, s, Q, T, g.

[0018] Next, we aggregate the hidden data and signature data of ordinary nodes in the cluster, including using the formula Aggregate the hidden data of common nodes, where C i,j represents the encrypted data of multiple times of ordinary node j in cluster i, m represents the number of ordinary nodes in cluster i, C i Represents aggregated privacy data, using the formula Aggregate signature data.

[0019] Furthermore, the base station recovers the aggregated data hidden by all sensors over multiple time periods, including:

[0020] Take out the private key S of all nodes in turn i,j ,calculate:

[0021]

[0022] Where n represents the number of cluster head nodes;

[0023] Aggregate the private data sent to it by n cluster head nodes to get:

[0024]

[0025] calculate

[0026]

[0027] calculate

[0028]

[0029] According to the obtained D, perform the following steps to restore the time periods {t1, t2, ..., t w} The corresponding original aggregate data D1, D2, ..., D w ;

[0030] Let Y w =D;

[0031] For x = w to 2, execute

[0032] because so

[0033] Repeat the above three steps to recover the time period {t1, t2, …, t w}The original aggregation number corresponding to

[0034] Next, the recovered data is verified to see if it has been tampered with based on the aggregated signature data, including:

[0035] calculate

[0036]

[0037] calculate

[0038] calculate

[0039] Calculate e(Sig,P) and e(SumD,T)·e(R,Q);

[0040] Where e represents a bilinear pair mapping, comparing e(Sig, P) and e(SumD, T)·e(R, Q) to see if they are equal. If they are equal, then the time period {t1, t2, …, t w} The corresponding original aggregate data D1, D2, ..., D w , otherwise it means that the data has been tampered with and will not be accepted.

[0041] The present invention provides a wireless sensor network multi-period data fusion system, characterized in that:

[0042] A data acquisition module is provided on each node in the wireless sensor network and is used to acquire data of each node in the wireless sensor network;

[0043] The data processing module is located on the common node, and uses the improved Paillier homomorphic encryption system to encrypt and hide the received data, uses the identity-based signature method to sign the data received in different time periods to obtain the signature data, and sends the hidden data and signature data to the cluster head node;

[0044] The data fusion module is located on the cluster head node and is used to aggregate the hidden data and signature data of the common nodes in the cluster and send them to the base station;

[0045] A data recovery module, located on the base station, is used to recover the aggregated data hidden in multiple time periods of all sensors;

[0046] The data verification module is located on the base station and performs integrity verification by comparing the signature of the original data with the aggregate signature. If the verification is passed, the collected perception data is accepted, otherwise the perception data will be deleted.

[0047] The embodiment of the present invention provides a method and system for multi-period data fusion in a wireless sensor network. Compared with the prior art, the advantageous effects thereof are as follows:

[0048] In most existing wireless sensor network data fusion solutions, after one round of data fusion, the base station can only obtain fused data within a certain time period in a certain area. In order to obtain fused data for multiple time periods in a certain area, multiple rounds of data fusion are required, which will greatly increase the consumption of network resources. The advantage of the present invention is that different fused data for multiple time periods in a certain area can be obtained after one round of data fusion. The present invention can efficiently realize the fusion of multi-time period data in wireless sensor networks, greatly improve data fusion efficiency, and reduce the consumption of data processing and transmission resources.

[0049] Many existing wireless sensor network data fusion methods based on hop-by-hop encryption, privacy homomorphism and secure multi-party computation use a large number of security mechanisms based on discrete logarithm problems with high computational complexity for encryption or signing. The advantage of the present invention is that it adopts an improved Paillier homomorphic encryption system, which transforms the original discrete logarithm problem into a simple multiplication operation, greatly reducing the computational complexity.

[0050] In order to support data privacy and data integrity verification, the existing wireless sensor network data fusion method based on data slicing needs to generate a large amount of redundant data, but the redundant data brings additional communication overhead. The advantage of the present invention is that it is based on the idea of ​​clustering. Each sensor node uses the improved Paillier homomorphic encryption system to encrypt and hide the perception data collected in different time periods, and uses the identity-based signature method to sign the perception data collected in different time periods, and then sends the hidden and signed perception data to the cluster head node of the cluster where it is located. After the cluster head node receives the data sent by the cluster node, it aggregates all the received hidden data and signature data respectively, and finally sends the aggregated data to the base station. After receiving all the aggregated data, the base station recovers the hidden aggregated data and verifies the recovered aggregated data. Therefore, the present invention does not require multiple rounds of information interaction, does not generate any redundant data, and greatly reduces the communication overhead of the network. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A wireless sensor network multi-period data fusion method and system overall architecture provided by the embodiment of the present invention;

[0052] Figure 2 A method and system for fusion of multi-period data in a wireless sensor network provided by an embodiment of the present invention provides a privacy data generation process;

[0053] Figure 3 A data aggregation process of a wireless sensor network multi-period data fusion method and system provided by an embodiment of the present invention;

[0054] Figure 4A wireless sensor network multi-period data fusion method and system data recovery process provided by an embodiment of the present invention;

[0055] Figure 5 A wireless sensor network multi-period data fusion method and a system data verification process are provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] See also Figures 1 to 5 The embodiment of the present invention provides a method for multi-period data fusion in a wireless sensor network, the method comprising:

[0058] In a clustered wireless sensor network, each sensor node uses the improved Paillier homomorphic encryption system to encrypt and hide the sensor data collected in different time periods, uses the identity-based signature algorithm to sign the sensor data collected in different time periods, and then sends the hidden and signed sensor data to the cluster head node of its cluster. After receiving the data sent by the cluster head node, it aggregates all the received hidden data and signature data, and finally sends the aggregated data to the base station. After receiving all the aggregated data, the base station first recovers the hidden aggregated data collected in different time periods, and then verifies the recovered aggregated data. If the verification is passed, the collected sensory aggregated data in different time periods will be accepted, otherwise the collected sensory aggregated data in different time periods will be deleted.

[0059] A multi-period data fusion method for wireless sensor networks mainly consists of five modules: system initialization, privacy data generation, data aggregation, data recovery and verification.

[0060] 1. System initialization

[0061] (1) Select a system security parameter k∈Z q , choose a large prime number q>2 k , generate two additive groups G1 and multiplicative groups G2 of order q, and a bilinear pairing map e:G1×G1→G2. Choose two secure hash functions H1, H2: {0,1} * →G1. Select a generator P from group G1 and a random number s∈Z qis the master secret key, Q = sP is the system public key, and a unique ID is assigned to each sensor node i,j ∈{0,1} * , ID i,j Represents the unique identity of the jth node in the i-th cluster, and calculates the identity-based private key S of each node i,j =sP i,j , choose a common state parameter β∈{0,1} * , calculate P β =H2(β)∈G1, calculate T=P β P, choose a generator Select a super-increasing sequence in, is a large prime number, each Length And meet Where N represents the total number of nodes in the network, and d represents the maximum value of the sensor data.

[0062] (2) Before the network is deployed, store (ID i,j , S i,j , H1, H2, β, g, q), the base station stores (k, q, e, G1, G2, H1, H2, P, s, Q, T, g) and the private keys of all nodes, and then deploys the nodes to the target area. The entire network is divided into n clusters, each cluster has m ordinary nodes, and each cluster selects a node called cluster head (Cluster Head, CH).

[0063] 2. Generation of private data

[0064] Assume that each common node j in the i-th cluster is in the time period {t1, t1, …, t w The perceived data are Node j first uses the improved Paillier homomorphic encryption system to protect the data privacy, then signs the data, and finally sends the private data and the signature of the data to the cluster head node CH i , the specific implementation process is as follows.

[0065] (1) Use your own private key S i,j and generator g to calculate the blinding factor

[0066] (2) For the time periods {t1, t1, …, t w}Perceived data Encrypt to get private data:

[0067]

[0068] (3) Calculate P i,j =H1(ID i,j ), calculate P β =H2(β)∈G1.

[0069] (4) For the time period {t1, t1, …, t w}'s sense data signature is signed:

[0070]

[0071] (5) Send data (C i,j , Sig i,j , R i,j ) to the cluster head node CH i .

[0072] 3. Data Aggregation

[0073] When the cluster head node CH i After receiving the data sent by all nodes j in the cluster, it first aggregates the private data of all m member nodes in the cluster to obtain aggregated private data, then aggregates the signature data of the m member nodes to obtain an aggregated signature, and finally sends the aggregated data to the base station. The specific execution process is as follows.

[0074] Compute aggregated private data:

[0075]

[0076] Calculate the aggregate signature:

[0077]

[0078] Send data (C i , Sig i , R i ) to the base station.

[0079] 4. Data Recovery

[0080] When the base station receives the aggregated data sent to it by all n cluster head nodes, it first aggregates the n aggregated private data to obtain new aggregated private data C, and then recovers the time periods {t1, t2, ..., t w} The corresponding original aggregate data D1, D2, ..., D w , the specific implementation process is as follows.

[0081] (1) Take out the private key S of all nodes in turn i,j ,calculate:

[0082]

[0083] (2) Aggregate the private data sent to it by n cluster head nodes to obtain:

[0084]

[0085]

[0086] (3) Calculation

[0087]

[0088] (4) Calculation

[0089]

[0090] (5) According to the obtained D, perform the following steps to restore the time periods {t1, t2, ..., t w} The corresponding original aggregate data D1, D2, ..., D w .

[0091] ①Let Y w =D;

[0092] ②For x=w to 2, execute

[0093] ③Because so

[0094] Through the above three steps, the time period {t1, t2, …, t w}The original aggregation number corresponding to

[0095] 5. Data Validation

[0096] The base station recovers the time period {t1, t2, ..., t w} The corresponding original aggregate data D1, D2, ..., D w Afterwards, it first aggregates the signature data sent to it by n cluster head nodes to obtain the aggregate signatures Sig and R, and then verifies the integrity of the restored original aggregate data. The specific integrity verification process is as follows.

[0097] (1) Calculation

[0098]

[0099] (2) Calculation

[0100]

[0101] (3) Calculation

[0102]

[0103] (4) Calculate e(Sig, P) and e(SumD, T)·e(R, Q).

[0104] (5) Compare e(Sig, P) and e(SumD, T)·e(R, Q) to see if they are equal. If they are equal, accept the time period {t1, t2, …, t w} The corresponding original aggregate data D1, D2, ..., D w , otherwise it means that the data has been tampered with and will not be accepted.

[0105] Disclosed above are only a few specific embodiments of the present invention. Those skilled in the art may make various changes and modifications to the embodiments of the present invention without departing from the spirit and scope of the present invention. However, the embodiments of the present invention are not limited thereto. Any changes that can be conceived by those skilled in the art should fall within the scope of protection of the present invention.

Claims

1. A multi-period data fusion method for wireless sensor networks, characterized in that: include: Divide the sensor network into clusters, each cluster contains cluster head nodes and ordinary nodes; Obtain the data received by ordinary nodes in multiple time periods, use the improved Paillier homomorphic encryption system to encrypt and hide the received data, use the identity-based signature method to sign the data received in different time periods to obtain signature data, and send the hidden data and signature data to the cluster head node; Each cluster head node aggregates the hidden data and signature data of ordinary nodes in the cluster and sends them to the base station. The base station recovers the aggregated data hidden in multiple time periods of all sensors and verifies whether the recovered data has been tampered with based on the aggregated signature data. The improved Paillier homomorphic encryption system is used to encrypt the received data to obtain the private data, which specifically includes: Using formula The received data is encrypted, wherein: represents a super-increasing sequence, is a large prime number, each Length And meet N represents the total number of nodes in the sensor network, d represents the maximum value of the sensor data, represents the data received by the sensor node, S i,j represents the identity-based private key of each network node, g represents the generator, q represents a large prime number, and mod represents the modulus operator; The identity-based signature method for signing data received in different time periods specifically includes: Calculate P i,j =H1(ID i,j ), calculate P β =H2(β)∈G1, where P i,j Represents the unique ID assigned to each sensor network node i,j The corresponding value of Hash function H1, P β The corresponding value of the Hash function H2 representing the common state parameter β; For time periods t1, t1, …, t w The data received by the sensor node is signed to obtain the signature data. Where w represents the subscript value of the time period.

2. A wireless sensor network multi-period data fusion method as claimed in claim 1, characterized in that: Before clustering the sensor network, the sensor network parameters must be selected, including: Choose a network security parameter k∈Z q , choose a large prime number q>2 k , generate two additive groups G1 and multiplicative groups G2 of order q, a bilinear pairing map e:G1×G1→G2, and select two secure hash functions H1,H2:{0,1} * →G1, select a generator P from group G1, and select a random number s∈Z q is the system master key, Q = sP is the system public key, and a unique ID is assigned to each sensor network node i,j ∈{0,1} * , ID i,j Represents the unique identity of the jth network node in the i-th cluster, and calculates the identity-based private key S of each network node i,j =sP i,j , choose a common state parameter β∈{0,1} * , calculate P β =H2(β)∈G1, calculate T=P β P, choose a generator Select a super-increasing sequence 3. A wireless sensor network multi-period data fusion method as claimed in claim 2, characterized in that: After the sensor network parameters are selected and before the common nodes receive data, the network sensor node parameters are stored: ID i,j , S i,j , H1, H2, β, g, q; store the private keys and parameters of all base station nodes: k, q, e, G1, G2, H1, H2, P, s, Q, T, g.

4. A wireless sensor network multi-period data fusion method as claimed in claim 2, characterized in that: The hidden data and signature data of ordinary nodes in the cluster are aggregated separately, including using the formula Aggregate the hidden data of common nodes, where C i,j represents the encrypted data of multiple times of ordinary node j in cluster i, m represents the number of ordinary nodes in cluster i, C i Represents aggregated privacy data, using the formula Aggregate signature data.

5. A wireless sensor network multi-period data fusion method as claimed in claim 2, characterized in that: The base station recovers the aggregated data hidden by all sensors in multiple time periods, including: Take out the private key S of all nodes in turn i,j ,calculate: Where n represents the number of cluster head nodes; Aggregate the private data sent to it by n cluster head nodes to get: q)q(mod q 2 )) calculate calculate According to the obtained D, perform the following steps to restore the time periods {t1, t2, ..., t w } The corresponding original aggregate data D1, D2, ..., D w ; Let Y w =D; For x = w to 2, execute because so Repeat the above three steps to recover the time periods t1, t2, ..., t w The corresponding original aggregation number 6. A wireless sensor network multi-period data fusion method as claimed in claim 1, characterized in that: The verifying whether the restored data has been tampered with according to the aggregated signature data includes: calculate calculate calculate Calculate e(Sig,P) and e(SumD,T)·e(R,Q); Where e represents a bilinear pair mapping; Compare e(Sig,P) and e(SumD,T)·e(R,Q) to see if they are equal. If they are equal, accept the time period {t1, t2, …, t w } The corresponding original aggregate data D1, D2, ..., D w , otherwise it means that the data has been tampered with and will not be accepted.

7. A wireless sensor network multi-period data fusion system, characterized in that: A data acquisition module is provided on each node in the wireless sensor network and is used to acquire data of each node in the wireless sensor network; The data processing module is located on the common node and uses the improved Paillier homomorphic encryption system to encrypt and hide the received data. Specifically, the formula is used: The received data is encrypted, wherein: represents a super-increasing sequence, is a large prime number, each Length And meet N represents the total number of nodes in the sensor network, d represents the maximum value of the sensor data, represents the data received by the sensor node, S i,j represents the identity-based private key of each network node, g represents the generator, q represents a large prime number, and mod represents the modulus operator; the identity-based signature method is used to sign the data received in different time periods to obtain the signature data, specifically: calculate P i,j =H1(ID i,j ), calculate P β =H2(β)∈G1, where P i,j Represents the unique ID assigned to each sensor network node i,j The corresponding value of Hash function H1, P β The corresponding value of the hash function H2 representing the common state parameter β for time periods t1, t1, …, t w The data received by the sensor node is signed to obtain the signature data. Send hidden data and signature data to the cluster head node; The data fusion module is located on the cluster head node and is used to aggregate the hidden data and signature data of the common nodes in the cluster and send them to the base station; A data recovery module, located on the base station, is used to recover the aggregated data hidden in multiple time periods of all sensors; The data verification module is located on the base station and performs integrity verification by comparing the signature of the original data with the aggregate signature. If the verification is passed, the collected perception data is accepted, otherwise the perception data will be deleted.

Images