CN111934863A - Secret key sharing method based on artificial noise and safety coding in edge calculation - Google Patents

Abstract

The invention discloses a secret key sharing method based on artificial noise and safety coding in edge calculation, which comprises the following steps of S1: s2, channel estimation: bob sends a channel estimation sequence to Alice, and the Alice estimates to obtain a channel matrix H_A(ii) a S3, safety coding: the edge side device Alice sets the key message b as (b) through security coding₁,b₂,...b_m) Coded as binary bits s ═(s)₁,s₂,...s_n) (ii) a S4, forming a transmitting signal by combining artificial noise: alice combines artificial noise through MIMO beamformingForming a transmission signal x: x is fs + Gz; s5, signal receiving and decoding: legal terminal Bob receives and demodulates the signal and decodes m key bits from the received signal

S6, repeatedly executing the step S3 to the step S5 until Bob obtains the keys with the length of L symbols.

Description

Secret key sharing method based on artificial noise and safety coding in edge calculation

Technical Field

The invention relates to key sharing in edge calculation, in particular to a key sharing method based on artificial noise and safety coding in edge calculation.

Background

The edge computing moves part or all of the computing tasks of the original cloud computing model to the network edge equipment, so that the computing load of a cloud computing center is reduced, and network congestion is relieved. The edge computing has rich application scenes, such as cloud computing task migration, video monitoring, intelligent transportation, an intelligent power grid and the like. With the rapid development of the 5G Internet of things and the edge computing network, a large number of novel services and applications are continuously emerging. Various confidential and sensitive data and the like in a wireless network are increased in a massive manner, the problem of information security is more and more prominent, and the security is gradually becoming a precondition for various different service applications. In a conventional wireless communication network, cryptography-based encryption technology is generally used at the network layer and the upper layer to secure system communication. Performing various cryptographic authentications requires establishing a secure shared key between the communicating parties. In a large number of novel application scenarios of edge computing, such as a large-scale IoT network and a smart grid network, a large number of resource-limited sensing node terminals are accessed, so that the complexity of key distribution and management based on cryptography is extremely high and even difficult to implement. The key generation and distribution technology based on physical channel uses the randomness and reciprocity of fading channel to generate and distribute the key between legal users.

At present, there are some preliminary research results on physical layer key generation, but the current physical layer key generation rate is slow, and the rate is highly related to the channel change speed. Experiments have verified that a typical physical layer key distribution system with 3 transmit and receive antennas takes 10 seconds or more to build a 128 bit length AES symmetric encryption key in an indoor channel with a key error rate of less than an order of magnitude. Because the two parties need to perform interaction at least for more than 3 times on the public channel in the process of establishing the key (including the processes of pilot frequency transmission, key agreement, privacy amplification, final consistency confirmation and the like of the two parties of the transceiver), the complexity of the communication protocol is higher, and the hidden danger of information leakage is promoted.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide a key sharing method based on artificial noise and security coding in edge computing, which realizes the key distribution of a physical layer under an edge computing environment by using the artificial noise and the security coding, only needs two times of interaction to realize the key distribution of the physical layer and reduces the time delay and the complexity caused by key sharing.

The purpose of the invention is realized by the following technical scheme: a secret key sharing method based on artificial noise and safety coding in edge calculation comprises the following steps

S1, setting a protection area:

setting that an edge side device Alice and a legal terminal Bob need to share a secret key, wherein Eve is an eavesdropper;

the edge device Alice and the legal terminal Bob respectively surround the edge device Alice and the legal terminal Bob by using a protection area with the radius of R, so that an eavesdropper Eve cannot enter the protection area for eavesdropping, namely, the distance between Eve and Alice is ensured to be larger than R, and the distance between Eve and Bob is ensured to be larger than R;

N_Aand N_BRespectively representing the number of antennas, N, of Alice and Bob_A＞N_B≥1；

S2, channel estimation: bob sends a channel estimation sequence to Alice, and the Alice estimates to obtain a channel matrix H_A；

S3, safety coding: the edge side device Alice sets the key message b as (b) through security coding₁,b₂,...b_m) Coded as binary bits s ═(s)₁,s₂,...s_n)；

Assuming binary security coding is used, the code rate is described as R_SFor code rate R ═ m/n_SWhere m is the packet length of the legitimate user information bits and n is the code length, by adjusting the transmit SNR signal-to-noise ratio such that Bob's decoded BER ρ_cSatisfies the following conditions

S4, forming a transmitting signal by combining artificial noise:

forming a transmitting signal x by combining MIMO beam forming and artificial noise by Alice:

x＝fs+Gz；

where s is the modulated unit scalar transmit key symbol and z is N_T-a 1 x 1-dimensional randomly generated complex gaussian artificial noise AN vector; the transmit beamformer is represented as:

f＝V(:,1)；

wherein H_A＝UΣV^HRepresenting singular value decomposition, f ═ V (: 1) is the first column vector of V;

meanwhile, the interference signal needs to be in the null space of Bob to avoid interference to legal users, and G is equal to V (: 2: N)_T-1), thus:

H_AG·z⊥H_Af

in terms of power allocation, the transmit signal covariance matrix E { xx^H}＝Q_x，T_r(Q_x)≤P_maxWherein T is_r(-) represents the trace-finding operator; the power allocated to the legitimate signal is P_STransmitting the interference signal with the residual power, then P_AN＝P_max-P_SAnd E: (a)_ZZ ^H)＝[P_AN/(N_T-1)I]；

S5, signal receiving and decoding: legal terminal Bob receivesDemodulating the signal and decoding the m key bits from the received signal

In the step S5, N is set_A＞N_BNot less than 1 and N is guaranteed_A-1>N_E(ii) a Wherein N is_ENumber of antennas for eavesdroppers; at the receiving end, w_AAnd w_ERespectively represent N_R×1，N_EAnd receiving the combining vector in the dimension of x 1, wherein after receiving and combining, the signals received by Bob and Eve are represented as follows:

wherein n is_AAnd n_EWhite noise vectors corresponding to Bob and Eve receiving signals respectively, and the power covariance of the white noise vectors is satisfied by a matrix:

H_Aand H_EIndependent of each other, channel matrices for the main channel and the eavesdropping channel, Alice and Bob unknown H_E(ii) a Wherein Alice feeds back w_A＝H_Af. To Bob, there are:

due to H_AG·z⊥H_Af, then there are:

that is, Bob is not interfered by artificial noise, and the received signal-to-interference-and-noise ratio is expressed as:

eve received artifact H_EThe reception quality of the interference of G · z decreases with increasing noise power, and the received signal-to-interference-and-noise ratio is expressed as:

the legitimate receiver Bob gets the sequence

Is a noisy version of the sequence s; meanwhile, an eavesdropper Eve can observe a noisy symbol sequence

Selecting a transmission power P_SSum noise power P_ANThe receiving signal-to-noise ratio of Bob is better than that of Eve, so that when Bob normally receives signals, Eve is interfered, and Eve is interfered

It is difficult for an eavesdropper to obtain the key message b (b) by demodulating and securely decoding the received signal₁,b₂,...b_m) A consistent key bit; and the legal terminal Bob can obtain the message b (b) of the key by demodulating and safely decoding the received signal₁,b₂,...b_m) Consistent m-bit key bits

S6, repeatedly executing the step S3 to the step S5 until Bob obtains the keys with the length of L symbols.

In step S6, after Bob has obtained L symbol-length keys, the method further includes a consistency confirmation step:

and B, confirming the consistency of the key between Bob and Alice, finishing the key sharing if the key obtained by Bob is consistent with the key shared by Alice, returning to the step S2 if the key obtained by Bob is consistent with the key shared by Alice, and re-executing the key sharing process according to the steps S2-S6.

Preferably, when Bob and Alice confirm the key consistency, the consistency confirmation method adopted includes:

alice uses the shared secret key to generate a signature through a hash function, encrypts the digital signature by using the shared secret key to obtain a ciphertext signature to be sent, and sends the ciphertext signature to Bob;

bob decrypts the ciphertext digital signature by using the key obtained by the Bob, generates the digital signature by using the key obtained by the Bob through a Hash function, compares the decoded digital signature with the digital signature of the Bob, and if the digital signature is consistent with the digital signature of the Bob, the key consistency is passed.

The invention has the beneficial effects that: (1) the invention only needs 2 times of interaction to realize the key distribution of the physical layer, realizes the low error rate of the terminal, and reduces the time delay and the complexity caused by key sharing; meanwhile, only one channel estimation is needed when the secret key is shared, which is beneficial to saving the expenditure. (2) According to the key distribution process, the key quantization, negotiation and privacy amplification do not need to be carried out by two parties, so that the two parties do not need to carry out multiple interactive negotiations related to the process on a public channel, and the complexity is reduced. (3) In the current technology, at least more than 3 times of interaction (including the processes of pilot frequency transmission, key agreement, privacy amplification, final consistency confirmation and the like of a transmitting party and a receiving party) needs to be performed on a public channel in the process of establishing a key, so that the complexity of a communication protocol is higher, and the hidden danger of information leakage is promoted. The technology does not need the processes of key quantization, negotiation and privacy amplification in public channel negotiation, avoids the hidden danger of information leakage, and improves the safety.

Drawings

FIG. 1 is a flow chart of a method of the present invention;

FIG. 2 is a schematic diagram of a key distribution architecture of a physical layer under an edge computing network;

FIG. 3 is a schematic diagram of the key bit error probability of Bob and Eve;

FIG. 4 is a diagram illustrating the probability of key bit errors of Bob and Eve under different transmission signal-to-noise ratios.

Detailed Description

The technical solutions of the present invention are further described in detail below with reference to the accompanying drawings, but the scope of the present invention is not limited to the following.

As shown in FIG. 1, a key sharing method based on artificial noise and security coding in edge calculation comprises the following steps

S1, setting a protection area:

setting that an edge side device Alice and a legal terminal Bob need to share a secret key, wherein Eve is an eavesdropper;

the edge device Alice and the legal terminal Bob respectively surround the edge device Alice and the legal terminal Bob by a protective area with the radius of R, and the protective area is an area where an eavesdropper Eve is forbidden to enter, so that the eavesdropper Eve cannot enter the protective area for eavesdropping, and the eavesdropper can be realized by a fence and a fence on a physical layer, and can also be a forbidden area for manual duty; namely, ensuring that the distance between Eve and Alice is greater than R and the distance between Eve and Bob is greater than R; the radius length of the protection region is generally required to be greater than the uncorrelated distance of the channel, and in the embodiment of the present application, the length is determined by the propagation environment of the channel and the frequency of the carrier frequency, and is generally 10cm to 100cm in the environment of sufficient scattering.

N_AAnd N_BRespectively representing the number of antennas, N, of Alice and Bob_A＞N_B≥1；

S2, channel estimation: bob sends a channel estimation sequence to Alice, and the Alice estimates to obtain a channel matrix H_A；

S3, safety coding: the edge side device Alice sets the key message b as (b) through security coding₁,b₂,...b_m) Coded as binary bits s ═(s)₁,s₂,...s_n)；

Assuming binary security coding is used, the code rate is described as R_SFor code rate R ═ m/n_SWhere m is the packet length of the legitimate user information bits and n is the code length, by adjusting the transmit SNR signal-to-noise ratio such that Bob's decoded BER ρ_cSatisfies the following conditions

S4, forming a transmitting signal by combining artificial noise:

forming a transmitting signal x by combining MIMO beam forming and artificial noise by Alice:

x＝fs+Gz；

where s is the modulated unit scalar transmit key symbol and z is N_T-a 1 x 1-dimensional randomly generated complex gaussian artificial noise AN vector; the transmit beamformer is represented as:

f＝V(:,1)；

wherein H_A＝UΣV^HRepresenting singular value decomposition, f ═ V (: 1) is the first column vector of V;

meanwhile, the interference signal needs to be in the null space of Bob to avoid interference to legal users, and G is equal to V (: 2: N)_T-1), thus:

H_AG·z⊥H_Af

in terms of power allocation, the transmit signal covariance matrix E { xx^H}＝Q_x，T_r(Q_x)≤P_maxWherein T is_r(-) represents the trace-finding operator; the power allocated to the legitimate signal is P_STransmitting the interference signal with the residual power, then P_AN＝P_max-P_SAnd E: (a)_ZZ ^H)＝[P_AN/(N_T-1)I]；

S5, signal receiving and decoding: legal terminal Bob receives and demodulates the signal and decodes m key bits from the received signal

In the step S5, N is set_A＞N_BNot less than 1 and N is guaranteed_A-1>N_E(ii) a Wherein N is_ENumber of antennas for eavesdroppers; at the receiving end, w_AAnd w_ERespectively represent N_R×1，N_EAnd receiving the combining vector in the dimension of x 1, wherein after receiving and combining, the signals received by Bob and Eve are represented as follows:

wherein n is_AAnd n_EWhite noise vectors corresponding to Bob and Eve receiving signals respectively, and the power covariance of the white noise vectors is satisfied by a matrix:

H_Aand H_EIndependent of each other, channel matrices for the main channel and the eavesdropping channel, Alice and Bob unknown H_E(ii) a Wherein Alice feeds back w_A＝H_Af. To Bob, there are:

due to H_AG·z⊥H_Af, then there are:

that is, Bob is not interfered by artificial noise, and the received signal-to-interference-and-noise ratio is expressed as:

eve received artifact H_EThe reception quality of the interference of G · z decreases with increasing noise power, and the received signal-to-interference-and-noise ratio is expressed as:

the legitimate receiver Bob gets the sequence

Is a noisy version of the sequence s; meanwhile, an eavesdropper Eve can observe a noisy symbol sequence

Selecting a transmission power P_SSum noise power P_ANThe receiving signal-to-noise ratio of Bob is better than that of Eve, so that when Bob normally receives signals, Eve is interfered, and Eve is interfered

It is difficult for an eavesdropper to obtain the key message b (b) by demodulating and securely decoding the received signal₁,b₂,...b_m) A consistent key bit; and the legal terminal Bob can obtain the message b (b) of the key by demodulating and safely decoding the received signal₁,b₂,...b_m) Consistent m-bit key bits

S6, repeatedly executing the steps S3 to S5 until Bob obtains L keys with symbol lengths, wherein L is greater than m, and L is generally an integral multiple of m, generally, for the key to be shared, L is a determined known value, and codes m with different lengths can be selected according to actual conditions and L requirements.

In step S6, after Bob has obtained L symbol-length keys, the method further includes a consistency confirmation step:

and B, confirming the consistency of the key between Bob and Alice, finishing the key sharing if the key obtained by Bob is consistent with the key shared by Alice, returning to the step S2 if the key obtained by Bob is consistent with the key shared by Alice, and re-executing the key sharing process according to the steps S2-S6.

In the embodiment of the present application, when Bob and Alice perform key consistency confirmation, the adopted consistency confirmation method includes:

alice uses the shared secret key to generate a signature through a hash function, encrypts the digital signature by using the shared secret key to obtain a ciphertext signature to be sent, and sends the ciphertext signature to Bob;

bob decrypts the ciphertext digital signature by using the key obtained by the Bob, generates the digital signature by using the key obtained by the Bob through a Hash function, compares the decoded digital signature with the digital signature of the Bob, and if the digital signature is consistent with the digital signature of the Bob, the key consistency is passed.

In the embodiment of the present application, an edge computing network lower physical layer key distribution architecture is as shown in fig. 2, and fig. 3 shows key bit error probabilities of Bob and Eve (error probabilities when PAN is 10dB, 20dB, and with or without security coding), and fig. 4 shows key bit error probabilities of Bob and Eve under different transmission signal-to-noise ratios, it can be seen that, as the signal-to-noise ratio of the transmission signal and the artificial interference power increase, the Eve decoding key bit error rate remains 0.5, which ensures the security thereof, while the Bob key bit decoding error probability decreases exponentially and tends to 0, which ensures the high consistency thereof.

The foregoing is a preferred embodiment of the present invention, it is to be understood that the invention is not limited to the form disclosed herein, but is not to be construed as excluding other embodiments, and is capable of other combinations, modifications, and environments and is capable of changes within the scope of the inventive concept as expressed herein, commensurate with the above teachings, or the skill or knowledge of the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (3)

1. A secret key sharing method based on artificial noise and safety coding in edge calculation is characterized in that: comprises the following steps

S1, setting a protection area:

setting that an edge side device Alice and a legal terminal Bob need to share a secret key, wherein Eve is an eavesdropper;

the edge device Alice and the legal terminal Bob respectively surround the edge device Alice and the legal terminal Bob by using a protection area with the radius of R, so that an eavesdropper Eve cannot enter the protection area for eavesdropping, namely, the distance between Eve and Alice is ensured to be larger than R, and the distance between Eve and Bob is ensured to be larger than R;

N_Aand N_BRespectively representing the number of antennas, N, of Alice and Bob_A＞N_B≥1；

S2, channel estimation: bob sends a channel estimation sequence to Alice, and the Alice estimates to obtain a channel matrix H_A；

S3, safety coding: the edge side device Alice sets the key message b as (b) through security coding₁,b₂,...b_m) Coded as binary bits s ═(s)₁,s₂,...s_n)；

Assuming binary security coding is used, the code rate is described as R_SFor code rate R ═ m/n_SWhere m is the packet length of the legitimate user information bits and n is the code length, by adjusting the transmit SNR signal-to-noise ratio such that Bob's decoded BER ρ_cSatisfies the following conditions

S4, forming a transmitting signal by combining artificial noise:

forming a transmitting signal x by combining MIMO beam forming and artificial noise by Alice:

x＝fs+Gz；

where s is the modulated unit scalar transmit key symbol and z is N_T-a 1 x 1-dimensional randomly generated complex gaussian artificial noise AN vector; the transmit beamformer is represented as:

f＝V(:,1)；

wherein H_A＝UΣV^HRepresenting singular value decomposition, f ═ V (: 1) is the first column vector of V;

meanwhile, the interference signal needs to be in the null space of Bob to avoid interference to legal users, and G is equal to V (: 2: N)_T-1), thus:

H_AG·z⊥H_Af

in terms of power allocation, the transmit signal covariance matrix E { xx^H}＝Q_x，T_r(Q_x)≤P_maxWherein T is_r(-) represents the trace-finding operator; the power allocated to the legitimate signal is P_STransmitting the interference signal with the residual power, then P_AN＝P_max-P_SAnd E: (a)_ZZ ^H)＝[P_AN/(N_T-1)]I；

S5, signal receiving and decoding: legal terminal Bob receives and demodulates the signal and decodes m key bits from the received signal

S6, repeatedly executing the step S3 to the step S5 until Bob obtains the keys with the length of L symbols.

2. The method of claim 1, wherein the method comprises: in the step S5, N is set_A＞N_BNot less than 1 and N is guaranteed_A-1>N_E(ii) a Wherein N is_ENumber of antennas for eavesdroppers;

at the receiving end, w_AAnd w_ERespectively represent N_R×1，N_EThe x 1-dimensional received combined vector,after the reception and combination process, the signals received by Bob and Eve are represented as follows:

wherein n is_AAnd n_EWhite noise vectors corresponding to Bob and Eve receiving signals respectively, and the power covariance of the white noise vectors is satisfied by a matrix:

H_Aand H_EIndependent of each other, channel matrices for the main channel and the eavesdropping channel, Alice and Bob unknown H_E(ii) a Wherein Alice feeds back w_A＝H_Af. To Bob, there are:

due to H_AG·z⊥H_Af, then there are:

that is, Bob is not interfered by artificial noise, and the received signal-to-interference-and-noise ratio is expressed as:

eve received artifact H_EThe reception quality of the interference of G · z decreases with increasing noise power, and the received signal-to-interference-and-noise ratio is expressed as:

the legitimate receiver Bob gets the sequence

Is a noisy version of the sequence s; meanwhile, an eavesdropper Eve can observe a noisy symbol sequence

Selecting a transmission power P_SSum noise power P_ANThe receiving signal-to-noise ratio of Bob is better than that of Eve, so that when Bob normally receives signals, Eve is interfered, and Eve is interfered

It is difficult for an eavesdropper to obtain the key message b (b) by demodulating and securely decoding the received signal₁,b₂,...b_m) A consistent key bit; and the legal terminal Bob can obtain the message b (b) of the key by demodulating and safely decoding the received signal₁,b₂,...b_m) Consistent m-bit key bits

3. The method of claim 1, wherein the method comprises: in step S6, after Bob has obtained L symbol-length keys, the method further includes a consistency confirmation step:

and B, confirming the consistency of the key between Bob and Alice, finishing the key sharing if the key obtained by Bob is consistent with the key shared by Alice, returning to the step S2 if the key obtained by Bob is consistent with the key shared by Alice, and re-executing the key sharing process according to the steps S2-S6.

Images