A method for identifying nodes in a wireless sensor network

Abstract

The application discloses a kind of identification methods of node in wireless sensor network, comprising: from wireless sensor network, any node is selected and recorded as main node;Through the broadcast RTS signal of main node, traverse all other nodes in wireless sensor network;The node that the back CTS signal received by main node is recorded as participant node is obtained;The interaction parameter between main node and each participant node is obtained;Based on cross-layer fuzzy algorithm, the initial evaluation value of each participant node is generated by processing interaction parameter, and initial evaluation total value is calculated based on initial evaluation value;Based on the Bi-LSTM+BBMO model pre-constructed, the transfer trust value of each participant node is generated by processing interaction parameter;The trust value of each participant node is calculated based on the trust value of main node, and trust total value is calculated based on trust value;According to initial evaluation total value and trust total value, the comprehensive evaluation of each participant node is calculated, and each participant node is judged and identified according to comprehensive evaluation;The application can identify node with high accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to a method for identifying nodes in a wireless sensor network, and belongs to the technical field of sensor networks. BACKGROUND

[0002] Wireless sensor networks (WSNs) have a large number of sensor nodes, which have computing and communication resources for data transmission, and have great technical application value. Future wars are gradually informatized and systematized, and each battlefield combat unit must share intelligence and real-time coordination, which must rely on networking and data sharing of multiple sensor nodes and information terminals. In order to transmit data using WSN, several important factors must be considered, including processing power, storage capacity, and power usage. Previous research focused on a cross-layer intrusion detection system architecture for network security, which can significantly improve network performance, reduce power consumption, and prolong network life. However, the related research algorithm cannot simultaneously monitor multiple-site malicious attacks.

[0003] Currently, researchers usually use two mainstream methods to identify malicious nodes in cluster-based sensor networks. The first method uses two modules of a primary key (PMK) to identify malicious nodes. The authentication module protects the network from interferers, and the monitoring module detects interfered sensor nodes. The "detection rate, network lifetime, data packet delivery rate accuracy, and throughput" are used at the data link layer to correctly distinguish attackers. The second method uses fuzzy logic to optimize interference indicators to correctly detect interference. Unstructured node sets constitute wireless sensor networks, but each sensor node has power and resource limitations. WSNs are resource-constrained and vulnerable to denial-of-service (DOS) attacks. In related optimization strategy research, the energy consumption and security aspects are difficult to balance, and accurately detecting intrusions in wireless networks is the most difficult problem to solve. SUMMARY

[0004] The purpose of the present application is to overcome the deficiencies in the prior art and provide a method for identifying nodes in a wireless sensor network, which can identify nodes in a wireless sensor network, thereby solving the technical problem of being unable to simultaneously monitor multiple-site malicious attacks.

[0005] To achieve the above purpose, the present application adopts the following technical scheme:

[0006] The present application provides a method for identifying nodes in a wireless sensor network, comprising:

[0007] Select any node from the wireless sensor network and mark it as the master node;

[0008] Broadcast the RTS signal through the master node to traverse all other nodes in the wireless sensor network;

[0009] The node receiving the backhaul CTS signal from the master node is recorded as a participating node;

[0010] Interaction parameters between the master node and each participating node are obtained;

[0011] Initial evaluation values of the master node for each participating node are generated based on the interaction parameters processed by a cross-layer fuzzy algorithm, and an initial evaluation total value is calculated based on the initial evaluation values;

[0012] A transfer trust value of the master node for each participating node is generated based on the interaction parameters processed by a pre-constructed Bi-LSTM+BBMO model;

[0013] A trust value of the master node for each participating node is calculated based on the transfer trust value, and a trust total value is calculated based on the trust value;

[0014] A comprehensive evaluation of the master node for each participating node is calculated according to the initial evaluation total value and the trust total value, and each participating node is judged and identified according to the comprehensive evaluation.

[0015] Optionally, the interaction parameters between the master node and each participating node are:

[0016]

[0017]

[0018]

[0019]

[0020] τ=[τ1 τ2 τ3…τ n ]

[0021]

[0022] fr=[fr1 fr2 fr3…fr n ]

[0023] In the formula, ξ RTS , λ relay , β op , ∈ rem , τ, fr are respectively a set of signal-to-noise ratio, relay difference frequency, buffer node capacity, residual energy, signal transmission time, effective data packet delivery rate, and fairness rate in the interaction parameters, τ n , fr n are respectively the interaction parameters between the master node and the participating node n, and n is the number of participating nodes.

[0024] Optionally, the processing of the interaction parameters based on the cross-layer fuzzy algorithm to generate the initial evaluation value of the master node to each participant node includes:

[0025] According to the interaction parameters between the master node and the participant node j, the initial evaluation class of the master node to the participant node j is determined:

[0026]

[0027] In the formula, ξ RTS_j , λ relay_j , β op_j , ∈ rem_j , τ j are the interaction parameters between the master node and the participant node j, are the upper and lower limits of the signal-to-noise ratio, are the upper limits of the relay difference frequency, are the upper limits of the buffer node capacity, are the upper limits of the remaining energy, τ Tl , τ Th are the upper limits of the signal transmission time;

[0028] According to the interaction parameters between the master node and each participant node, the initial evaluation operator θ is determined:

[0029]

[0030] In the formula, ξ RTS_i , τ i , β op_i , λ relay_i are the interaction parameters between the master node and the participant node i, and σ is the Relu function.

[0031] According to the initial evaluation operator and the initial evaluation class of the master node to the participant node j, the initial evaluation value A j of the master node to the participant node j is determined:

[0032]

[0033] Optionally, the initial evaluation total value A′ j is:

[0034] A′ j = [ξ RTS_j λ relay_j β op_j τ j ]*A j .

[0035] Optionally, the processing of the interaction parameters based on the pre-constructed Bi-LSTM+BBMO model to generate the transmission trust value of the master node to each participant node includes:

[0036] M groups of historical interaction parameters between the master node and each participant node are obtained m=1, 2, …, M, and a corresponding label is made A training sample set corresponding to the master node and each participant node is generated;

[0037] A BBMO algorithm model is constructed, and each training sample set is input into each BBMO algorithm model for training, so as to optimize the weight parameters of each BBMO algorithm model;

[0038] λ relay_j , β op_j , τ j , fr j in the interaction parameters between the master node and each participant node are obtained

[0039] A Bi-LSTM network model is constructed, and and the weight parameters of the corresponding trained BBMO algorithm model are input into the Bi-LSTM network model to obtain the transfer trust value T j a of the master node to each participant node.

[0040] Optionally, the calculation of the trust value of the master node to each participant node based on the transfer trust value comprises:

[0041] According to the interaction parameters between the master node and the participant node j, the trust category of the master node to the participant node j is determined:

[0042]

[0043] In the formula, fr j is the interaction parameter between the master node and the participant node j, is the upper and lower limits of the effective data packet delivery rate, fr Tl , fr Th is the upper limit of the fairness rate;

[0044] According to the trust category and the transfer trust value of the master node to the participant node j, the trust value of the master node to the participant node j is determined:

[0045]

[0046] In the formula, σ is a Relu function.

[0047] Optionally, the total trust value T′ j is:

[0048]

[0049] Optionally, the comprehensive evaluation of the main node to each participating node is:

[0050]

[0051] In the formula, R j , A' j , T' j respectively comprehensive evaluation of the main node to participating node j, initial evaluation total value, trust total value.

[0052] Optionally, the judging and identifying each participating node according to the comprehensive evaluation comprises:

[0053] If R j ≥ 0.75, participating node j is a trusted node;

[0054] If 0.25 ≤ R j < 0.75, participating node j is an uncertain node;

[0055] If R j < 0.25, participating node j is a malicious node.

[0056] Compared with the prior art, the beneficial effects achieved by the present application are:

[0057] The node identification method provided by the present application in a wireless sensor network determines the main node and the participating node, processes the interaction parameters between the main node and the participating node through a cross-layer fuzzy algorithm to obtain the initial evaluation total value of each participating node, processes the interaction parameters between the main node and the participating node through a Bi-LSTM+BBMO model to obtain the trust total value of each participating node, constructs the comprehensive evaluation of each node through the initial evaluation total value and the trust total value, and finally identifies the participating node through the comprehensive evaluation; the node identification of the present application has high precision and can ensure the safe routing of the wireless sensor network. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 is a flowchart of the node identification method provided by the first embodiment of the present application in a wireless sensor network. DETAILED DESCRIPTION

[0059] The present application will be further described below in conjunction with the drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and cannot be used to limit the protection scope of the present application.

[0060] Embodiment one:

[0061] For example, Figure 1As shown, the present application provides a method for identifying nodes in a wireless sensor network, comprising the following steps:

[0062] 1. Select any node from the wireless sensor network and mark it as the master node.

[0063] 2. Broadcast the RTS signal through the master node to traverse all other nodes in the wireless sensor network.

[0064] 3. Obtain the nodes that return the CTS signal received by the master node and mark them as participating nodes.

[0065] 4. Obtain the interaction parameters between the master node and each participating node.

[0066] The implementation of the present application is as follows:

[0067]

[0068]

[0069]

[0070]

[0071] τ = [τ1 τ2 τ3…τ n ]

[0072]

[0073] fr = [fr1 fr2 fr3…fr n ]

[0074] In the formula, ξ RTS , λ relay , β op , ∈ rem , τ, fr are the sets of signal-to-noise ratio, relay difference frequency, buffer node capacity, residual energy, signal transmission time, effective data packet delivery rate, and fairness rate in the interaction parameters, τ n , fr n are the interaction parameters between the master node and the participating node n, and n is the number of participating nodes.

[0075] The signal-to-noise ratio refers to the ratio of the trusted storage of the participating node to the total information amount.

[0076] The relay difference frequency refers to the difference between the transmission frequency of the master node and the output frequency of the participating node.

[0077] The buffer node capacity refers to the buffer node capacity of the participating node, which is expressed as:

[0078]

[0079] In the formula, T is the total time of the main node transmitting information to the participating node and the participating node feeding back information to the main node, T delay is the preset maximum delay, generally 0.05s.

[0080] The residual energy refers to the residual energy of the participating node, and its expression is:

[0081]

[0082] In the formula, l is the data length, d is the transmission distance, x is the transmission times, E Default is the default circuit loss of the transmitting electrode and the receiver, OP is the transmitting amplification power consumption;

[0083] The signal transmission time refers to the single information transmission time from the main node to the participating node.

[0084] The effective data packet delivery rate refers to the ratio of the effective data transmitted from the participating node to the main node to the data transmitted from the main node to the participating node.

[0085] The fairness rate refers to the signal strength / buffer node capacity*information transmission time.

[0086] 5. The initial evaluation value of the main node to each participating node is generated by processing the interaction parameters based on the cross-layer fuzzy algorithm, and the initial evaluation total value is calculated based on the initial evaluation value; specifically including:

[0087] (1) The initial evaluation category of the main node to the participating node j is determined according to the interaction parameters between the main node and the participating node j:

[0088]

[0089] In the formula, ξ RTS_j , λ relay_j , β op_j , ∈ rem_j , τ j are the interaction parameters between the main node and the participating node j, are the upper and lower limits of the signal-to-noise ratio, is the upper limit of the relay difference frequency, is the upper limit of the buffer node capacity, is the upper limit of the residual energy, τ Tl , τ Th are the upper limits of the signal transmission time.

[0090] (2) The initial evaluation operator θ is determined according to the interaction parameters between the main node and each participating node:

[0091]

[0092] wherein, ξ RTS_i , τ i , β op_i , λ relay_i are interaction parameters between the main node and the participant node i, and σ is a Relu function.

[0093] (3) The initial evaluation value A j of the main node to the participant node j is determined according to the initial evaluation operator and the initial evaluation class of the main node to the participant node j.

[0094]

[0095] (4) The initial evaluation total value A' j is:

[0096] A' j = [ξ RTS_j λ relay_j β op_j τ j ]*A j .

[0097] 6. The interaction parameters are processed based on the pre-constructed Bi-LSTM+BBMO model to generate the transfer trust value of the main node to each participant node; specifically including:

[0098] (1) M groups of historical interaction parameters between the main node and each participant node are obtained m = 1, 2, …, M, and the corresponding labels are made to generate the training sample set corresponding to the main node and each participant node;

[0099] (2) A BBMO (Bumble Bees mating Optimization) algorithm model is constructed, and each training sample set is input into each BBMO algorithm model for training to optimize the weight parameters of each BBMO algorithm model;

[0100] (3) The interaction parameters λ relay_j , β op_j , τ j , fr j between the main node and each participant node are obtained and input into the corresponding trained BBMO algorithm model to obtain

[0101]

[0102] (4) A Bi-LSTM network model is constructed, and and the weight parameters of the corresponding trained BBMO algorithm model are input into the Bi-LSTM network model, to obtain the transfer trust value T of the master node to each participating node j a ;

[0103] wherein,

[0104] M t =O t tanh(D t )

[0105] O t =σ(U IM X t +U MM M t-1 )

[0106] D t =F t D t-1 +I t G t

[0107] F t =σ(U IF X t +B2+U MF M t-1 )

[0108] I t =σ(U II X t +B1+U MI M t-1 )

[0109] G t =tanh(U IG X t +U MG M t-1 )

[0110] The Bi-LSTM network model includes a plurality of cascaded LSTM units, each LSTM unit including a forgetting layer, an input layer and an output layer; in the formula, B1 and B2 are bias parameters of the input layer and the forgetting layer, and are generally set to 0.22; t is the tth LSTM unit; U IF , U MF are weight parameters of the forgetting layer, U II , U MI are weight parameters of the input layer, U IM , U MM are weight parameters of the output layer, and U IG , U MG are weight parameter update values;

[0111]

[0112]

[0113] U IG =W new1 ·B1

[0114] U MG =W new2 ·B2

[0115]

[0116]

[0117]

[0118]

[0119] In the formula, W new1 , W new2 is the weight parameter of the BBMO algorithm model, N is the total number of nodes in the wireless sensor network, s is the bit number of the master node in the wireless sensor network, t s is the correlation value of the master node s, logit is the Logic Regression of regression analysis.

[0120] 7. Calculate the trust value of the master node to each participating node based on the transmission trust value, and calculate the total trust value based on the trust value; specifically including:

[0121] (1) Determine the trust category of the master node to participating node j according to the interaction parameter between the master node and participating node j:

[0122]

[0123] In the formula, fr j is the interaction parameter between the master node and participating node j, is the upper and lower limit of the effective data packet delivery rate, fr Tl , fr Th is the upper limit of the fairness rate;

[0124] (2) Determine the trust value of the master node to participating node j according to the trust category of the master node to participating node j and the transmission trust value:

[0125]

[0126] In the formula, σ is the Relu function.

[0127] (3) Total trust value T'j is:

[0128]

[0129] 8. Calculating the comprehensive evaluation of the main node to each participant node according to the initial evaluation total value and the trust total value, and judging and identifying each participant node according to the comprehensive evaluation.

[0130] (1) The comprehensive evaluation of the main node to each participant node is:

[0131]

[0132] wherein, R j , A' j , T' j are the comprehensive evaluation of the main node to participant node j, the initial evaluation total value, and the trust total value, respectively.

[0133] (2) The judging and identifying of each participant node according to the comprehensive evaluation includes:

[0134] If R j ≥ 0.75, then participant node j is a trust node;

[0135] If 0.25 ≤ R j < 0.75, then participant node j is an uncertain node;

[0136] If R j < 0.25, then participant node j is a malicious node.

[0137] Those skilled in the art will understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0138] The present application is described with reference to the flowcharts and / or block diagrams according to the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce the functions described in the flowcharts and / or block diagrams. Figure 1 one flow or multiple flows and / or blocksFigure 1 means for performing the function specified in the block or blocks.

[0139] The above merely preferred embodiments of the present application, it should be noted that for those of ordinary skill in the art, without departing from the technical principles of the present application, can make several improvements and variations, these improvements and variations should also be considered as the protection scope of the present application.

Claims

1. A method for identifying nodes in a wireless sensor network, characterized in that, include: Select any node from the wireless sensor network and designate it as the master node; The master node broadcasts the RTS signal to traverse all other nodes in the wireless sensor network. The nodes that receive the CTS signal from the master node are recorded as participating nodes; Obtain the interaction parameters between the master node and each participating node; The interaction parameters are processed using a cross-layer fuzzy algorithm to generate the initial evaluation values ​​of the master node for each participating node, and the initial total evaluation value is calculated based on the initial evaluation values. The interaction parameters are processed based on the pre-built Bi-LSTM+BBMO model to generate the trust value transmitted by the master node to each participating node. The master node calculates the trust value of each participating node based on the transmitted trust value, and the total trust value is calculated based on the trust value. The master node calculates the overall evaluation of each participating node based on the initial total assessment value and the total trust value, and then judges and identifies each participating node based on the overall evaluation.

2. The method for identifying nodes in a wireless sensor network according to claim 1, characterized in that, The interaction parameters between the master node and each participating node are as follows: τ=[τ1 τ2 τ3…τ n ] fr=[fr1 fr2 fr3…fr n ] In the formula, ξ RTS , λ relay β op ,∈ rem ,τ, fr represents a set of interaction parameters including signal-to-noise ratio, relay frequency difference, buffer node capacity, remaining energy, signal transmission time, effective data packet delivery rate, and fairness rate. τ n , fr n These are the interaction parameters between the master node and the participating node n, where n is the number of participating nodes.

3. The method for identifying nodes in a wireless sensor network according to claim 2, characterized in that, The process of generating initial evaluation values ​​for each participating node by the master node based on the cross-layer fuzzy algorithm for interaction parameters includes: The initial evaluation category of the master node for the participating node j is determined based on the interaction parameters between the master node and the participating node j: In the formula, ξ RTS_j , λ relay_j β op_j ,∈ rem_j τ j These are the interaction parameters between the master node and the participating node j, respectively. These represent the upper and lower limits of the signal-to-noise ratio. These are the upper and lower limits of the repeater frequency difference. These are the upper and lower limits of the buffer node capacity. τ represents the upper and lower limits of the remaining energy. Tl τ Th These are the upper and lower limits of signal transmission time; The initial evaluation operator is determined based on the interaction parameters between the master node and each participating node. : In the formula, ξ RTS_i τ i β op_i , λ relay_i are the interaction parameters between the master node and the participating node i, respectively, and σ is the ReLU function; The initial evaluation value A of the master node for participating node j is determined based on the initial evaluation operator and the initial evaluation category of the master node for participating node j. j :

4. The method for identifying nodes in a wireless sensor network according to claim 3, characterized in that, The initial total assessment value A′ j for: A′ j =[ξ RTS_j l relay_j b op_j t j ]*A j 。 5. The method for identifying nodes in a wireless sensor network according to claim 1, characterized in that, The process of generating trust values ​​from the master node to each participating node by processing interaction parameters based on the pre-built Bi-LSTM+BBMO model includes: Retrieve the M group of historical interaction parameters between the master node and each participating node. And create corresponding labels. Generate training sample sets corresponding to the master node and each participating node; Construct BBMO algorithm models and input each training sample set into each BBMO algorithm model for training, and optimize the weight parameters of each BBMO algorithm model; Obtain the interaction parameters λ between the master node and each participating node. relay_j β op_j τ j , fr j And by inputting the corresponding trained BBMO algorithm model, we obtain Construct a Bi-LSTM network model, and The weight parameters of the pre-trained BBMO algorithm model are input into the Bi-LSTM network model to obtain the trust values ​​transmitted from the master node to each participating node.

6. The method for identifying nodes in a wireless sensor network according to claim 5, characterized in that, The calculation of the master node's trust value for each participating node based on the transmitted trust value includes: The trust level of the master node towards the participating node j is determined based on the interaction parameters between the master node and the participating node j: In the formula, fr j The interaction parameters between the master node and participating node j. For the upper and lower limits of the effective packet delivery rate, fr Tl ,fr Th These represent the upper and lower limits of the fairness rate; The master node's trust value for participant node j is determined based on the master node's trust category and the transmitted trust value: In the formula, σ is the ReLU function.

7. The method for identifying nodes in a wireless sensor network according to claim 6, characterized in that, The total trust value T′ j for:

8. The method for identifying nodes in a wireless sensor network according to claim 1, characterized in that, The master node's comprehensive evaluation of each participating node is as follows: In the formula, R j A′ j 、T′ j These are the master node's overall evaluation of participating node j, the initial total evaluation value, and the total trust value, respectively.

9. The method for identifying nodes in a wireless sensor network according to claim 8, characterized in that, The process of judging and identifying each participating node based on a comprehensive evaluation includes: If R j If the value is ≥0.75, then participating node j is a trusted node; If 0.25≤R j If the value is less than 0.75, then participating node j is an uncertain node; If R j If the value is less than 0.25, then participating node j is a malicious node.

Images