A physical layer cooperative key generation method based on double key agreement

Abstract

The application relates to a physical layer cooperative key generation method based on double key negotiation, and belongs to the field of Internet of Things security. The method comprises two stages of channel estimation and key negotiation. In the channel estimation stage, legitimate communication parties obtain part of relay channel information through channel estimation, and then simultaneously send training sequences to enable the relay and the eavesdropping node to obtain superposition state information about the relay channel; in the key negotiation stage, double key negotiation is used to generate keys and private keys with different security permission levels. The application can effectively improve the security and efficiency of Internet of Things communication, and is more effective when the eavesdropping channel has correlation.

Description

Technical Field

[0001] This invention belongs to the field of Internet of Things (IoT) security and relates to a physical layer collaborative key generation method based on dual key negotiation. Background Technology

[0002] In recent years, with the rapid development of 5G technology, the Internet of Things (IoT), envisioned as the "Internet of Everything," has gradually entered the lives of ordinary people. At the same time, the security issues of the IoT have also attracted much attention. The IoT generally uses wireless signals as the transmission medium, but due to the openness of wireless signals and the continuous development of attack methods, it faces increasingly severe security threats, which has become a significant constraint on its development. Current research on secure communication in the IoT mainly adopts high-level encryption systems of traditional wireless networks, but this security approach has significant shortcomings. First, it requires complex key distribution and management mechanisms, but key distribution is difficult to implement in the IoT, and there is no trusted key management center. Second, due to the rapid development of quantum computing, key protocols that rely on hardware conditions and computing power will face serious threats.

[0003] Wireless physical layer key generation is one of the effective technologies for securing wireless communication. Unlike traditional security methods, physical layer key generation technology is based on information theory security, enabling theoretical security independent of the eavesdropper's computing power. Legitimate parties can directly extract keys from the shared channel by utilizing the short-term reciprocity and randomness of the wireless channel, avoiding complex key distribution and management processes. This approach has lower implementation complexity and higher security, making it well-suited for the Internet of Things (IoT).

[0004] Current research on typical 4-node models (legal communicating parties Alice and Bob, a relay node Relay, and an eavesdropping node Eve) exists, but most studies do not consider the potential correlation between the eavesdropping and legitimate channels due to factors such as the location of the eavesdropping node and the spatial scattering environment in actual communication. Furthermore, in real-world communication systems, two legitimate users may need to generate multiple keys with different security permission levels in the presence of an eavesdropping node. For example, in tactical networks or wireless networks in the financial industry, Alice and Bob may need to simultaneously exchange two types of data with different security constraints; the data with lower security constraints can be disclosed to an authenticated user (e.g., Relay), but the data with higher security constraints must not be disclosed.

[0005] Therefore, in wireless cooperative networks that consider the correlation between legitimate channels and eavesdropping channels, how to improve the key rate while ensuring the generation of multiple keys with different security clearance levels is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] In view of this, the purpose of this invention is to provide a physical layer collaborative key generation method based on dual key negotiation, which improves the security and efficiency of IoT communication, especially when there is correlation in the eavesdropping channel.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A physical layer cooperative key generation method based on dual key negotiation, targeting situations where eavesdropping channels and legitimate channels are correlated, includes two stages: channel estimation and key negotiation, specifically comprising the following steps:

[0009] S1: Channel estimation: In time slot T1, the relay node Relay sends training sequence S1 so that legitimate users Alice and Bob can obtain partial relay channel information; in time slot T2, legitimate users Alice and Bob simultaneously send training sequence S2 so that the relay node Relay and the eavesdropping node Eve can obtain superposition state information about the relay channel.

[0010] S2: Key Negotiation: In the first negotiation, legitimate users Alice, Bob, and relay node Relay generate a key K1 that is kept secret from the eavesdropping node Eve using their respective channel estimates. In the second negotiation, the estimates of all relay channels and the eavesdropping channel are extracted to generate a private key K2 that is kept secret from both relay node Relay and eavesdropping node Eve. By combining K1 and K2 generated in the two negotiations, legitimate users Alice and Bob finally generate the key K. s With private key K p .

[0011] Furthermore, in step S1, the channel estimate is calculated using the following formula:

[0012]

[0013]

[0014] in, This represents Alice's estimate of the channel gain h. This represents the estimated value of channel gain g by the legitimate user Bob. Let represent the estimated values ​​of the relay channel by the relay node Relay and the eavesdropping node Eve, respectively; h represents the channel gain between the legitimate user Alice and the relay node Relay, and g represents the channel gain between the legitimate user Bob and the relay node Relay. AE g BE Y represents the channel gain from legitimate users Alice and Bob to the eavesdropping node Eve, respectively; A YB Y R Y E These are the signals received by legitimate user Alice, legitimate user Bob, relay node Relay, and eavesdropping node, respectively; N A N B N R N E S1 and S2 are the additive white Gaussian noise at the locations of legitimate user Alice, legitimate user Bob, relay node Relay, and eavesdropping node, respectively; S1 and S2 are the training sequences sent by relay node Relay and legitimate users Alice and Bob in different time slots, and S = S1 = S2, (·) T Represents the transpose of a vector or matrix.

[0015] Furthermore, in step S2, during the first negotiation, the relay node uses Slepian-Wolf coding to generate auxiliary information using its own channel estimation and transmits it through the common channel. Specifically, this includes the following steps:

[0016] S201: Legitimate users Alice and Bob can fully recover the channel estimate of the relay node Relay. The legitimate user Alice obtains information by combining her own channel estimates with information from the public channel. An estimate of the channel gain g is obtained. Bob combines his own channel estimates with information obtained from public channels. Obtain an estimate of the channel gain h.

[0017] S202: Through channel estimation Legitimate users Alice and Bob generate a secret key K1 that is kept confidential from the eavesdropping node Eve.

[0018] Furthermore, in step S2, during the second negotiation, since Alice has already obtained... Bob has received The shared information of both is enhanced. The legitimate user Alice uses Slepian-Wolf coding, generating auxiliary information using observations of the channel gain h and transmitting it through the common channel. Based on the information-theoretic security principle of "one-time pad," Y... R This information was considered public during the negotiation, allowing legitimate users Alice and Bob to generate a private key K2 that was kept secret from both the relay node Relay and the eavesdropping node Eve.

[0019] Furthermore, in step S2, by combining the key K1 generated in the first negotiation with the private key K2 generated in the second negotiation, the legitimate users Alice and Bob finally generate the key K. s For K s =K1+K2, generating the private key K p For K p =K2.

[0020] The beneficial effects of this invention are as follows: The method comprises two stages: channel estimation and key negotiation. In the channel estimation stage, the legitimate communicating parties obtain partial relay channel information through channel estimation, and then simultaneously send a training sequence so that the relay and eavesdropping nodes obtain superposition state information about the relay channel. In the key negotiation stage, dual key negotiation is used to generate keys and private keys with different security permission levels. This invention can improve the key generation rate, thereby effectively improving the security and efficiency of IoT communication, especially when there is correlation in the eavesdropping channel.

[0021] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0023] Figure 1 This is a flowchart of the physical layer key generation process of the present invention;

[0024] Figure 2 This is the system model of the present invention;

[0025] Figure 3 A time slot allocation diagram for sending training sequences to each node;

[0026] Figure 4 The curve shows the key rate as a function of the signal-to-noise ratio.

[0027] Figure 5 The curve shows the key rate as a function of the correlation coefficient.

[0028] Figure 6 This represents the relationship between the key rate and the number of relay nodes. Detailed Implementation

[0029] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0030] First, some of the nouns or terms appearing in the embodiments of this invention are explained as follows:

[0031] Key: In this embodiment, the key is kept secret only from the eavesdropping node Eve.

[0032] Private key: In this embodiment, it is a key that is kept secret from both the eavesdropping node Eve and the relay node Relay.

[0033] The model considered in this embodiment is as follows: Figure 2 As shown, Alice and Bob are legitimate users; Relay is a relay node that can communicate directly with Alice and Bob; an eavesdropping node, Eve, exists around Relay, attempting to steal confidential information; all nodes are equipped with a single antenna. Due to path loss and shadowing effects, there is no direct link between Alice and Bob, and they can only communicate via Relay. It is assumed that there is a noiseless common channel between the nodes; Eve can also access the common channel through which legitimate nodes exchange information. Eve obtains noisy transmission information through the wireless channel and noiseless transmission information through the common channel. Furthermore, Eve is a passive node and will not send signals to interfere with legitimate transmission processes.

[0034] Let h AR h represents the channel gain from Alice to Relay. RA Given the channel gain from Relay to Alice, h can be obtained based on channel reciprocity. AR =h RA =h; for the channel gain between Bob and Relay, g BR =g RB =g. h AE g BE Let h and h' represent the channel gains from Alice and Bob to Eve, respectively. Considering a real-world communication scenario where there is a correlation between the relay channel and the eavesdropping channel, the channel gains h and h' ... AE The correlation coefficient between them can be expressed as:

[0035]

[0036] in,(·) H This is the conjugate transpose of the matrix. Similarly, g and g BE The correlation coefficient between them can be expressed as:

[0037]

[0038] When using the Jakes model, the correlation coefficient depends only on the distance between Relay and Eve, which is sufficient to obtain...

[0039] The channel between nodes is modeled as a time-varying Rayleigh block fading channel, meaning the channel gain remains constant within the coherence time T and changes to other independent random values ​​in the next coherence time. It is assumed that the noise received by Alice, Bob, and Relay is independent and identically distributed zero-mean circularly-symmetric complex Gaussian (ZMCSCG) white noise with variance... Channel gain h and h AE The variance is and The ZMCSCG random variable, i.e. Similarly, we can obtain

[0040] To reflect the path loss and small-scale fading of the channel model, we define in And f h and The correlation coefficient between them is ρ, d h and Let represent the distances from Alice to Relay and Eve, respectively, and l be the path loss coefficient. Therefore, the channel gains h and h AE The variances can be calculated as follows: Similarly, the channel gain g and g BE It can also be defined as in And f g with f gBE The correlation coefficient between them is ρ, d g and Let these represent the distances from Bob to Relay and Eve, respectively; then the corresponding channel gain variance can be obtained.

[0041] The key generation process of this invention consists of two steps: the first step is channel estimation, and the second step is key negotiation. The key negotiation process includes two negotiations: the first negotiation generates a key K1 that is kept secret from Eve, and the second negotiation generates a private key K2 that is kept secret from both Relay and Eve. Finally, by combining K1 and K2, the goal of simultaneously generating multiple keys with different security permission levels is achieved.

[0042] The physical layer key generation process is as follows: Figure 1 As shown, the detailed process is as follows:

[0043] 1. Channel estimation

[0044] This step consists of two time slots, where the coherence time T is evenly distributed into two time slots T1 and T2, i.e., T1 = T2 = T / 2, as shown in the time slot allocation. Figure 3 As shown. Assume S1 = S2 = S, and the transmission power of Alice, Bob, and Relay is all P, therefore ||S|| 2 =PT / 2.

[0045] 1) In time slot T1, Relay sends the same training sequence S1 to Alice and Bob, and Alice and Bob receive signals Y respectively. A =hS1+N A Y B =gS1+N B At this point, Alice can obtain an estimate of h. Bob can obtain an estimate of g.

[0046]

[0047]

[0048] 2) In time slot T2, Alice and Bob simultaneously send the same training sequence S2 to Relay, and Relay and Eve receive signals Y respectively. R =hS² + ​​gS² + N R Y E =h AE S2+g BE S2+N E At this point, Relay and Eve each obtain an estimated value. and

[0049]

[0050]

[0051] Therefore, we can obtain All are ZMCSCG variables, and

[0052] 2. Key Negotiation

[0053] The key negotiation process consists of two stages: In the first stage, Alice and Bob use the auxiliary information transmitted by the Relay on the public channel to negotiate and generate a key K1. In the second stage, after Alice and Bob each obtain an estimate of the relay channel, they use relevant observation information about the channel gain h to negotiate and generate a private key K2. Alice and Bob combine K1 and K2 to obtain the final key K. s and private key K p .

[0054] 1) First negotiation

[0055] To increase the shared information between Alice and Bob, Relay can use Slepian-Wolf encoding to send auxiliary information to Alice and Bob, enabling them to negotiate and generate a key. Relay will then estimate... Quantized into binary sequence Where Δ represents the quantization interval, Alice and Bob also respectively provide their estimation information. and Quantized into binary sequence and Then Relay randomly... The typical set of sequences is divided into non-overlapping subsets, each subset containing multiple sequences. Typical sequences are defined, with each typical sequence containing two indices: the subset index and the sequence index within the subset. The Relay estimates... get Following the sequence, the subset index number of the subset containing the sequence is sent to Alice and Bob as negotiation information. The Relay needs to be sent via a public channel. Bit information, where H(X|Y) represents the conditional entropy of random variable X given random variable Y. By combining their own channel estimation with information transmitted on the public channel, Alice and Bob can accurately recover the... Then, the index of this sequence within the subset is used as the key K1, and the key rate is R. 11 ,Right now

[0056]

[0057] Since the subset index and the sequence index within the subset are independent, even if Eve obtains the subset index, she still doesn't know the sequence index within the subset, meaning she cannot obtain any information about key K1. Let the quantization interval Δ approach 0, then the rate R1 of key K1 is...

[0058]

[0059] Where I(X;Y) represents the mutual information between random variables X and Y.

[0060] 2. Second negotiation

[0061] Since Alice has already obtained an estimate of the channel gain h Bob can combine his own channel estimation values. And information Y obtained through public channels R This allows us to obtain an estimate of the channel gain h. Because Relay sent Y during the first negotiation R The auxiliary information is used to generate key K1. According to the "one-time pad" principle of information theory security, Y R In the second negotiation, this should be considered public information to ensure the security of K2. Similar to the first negotiation, Alice uses the channel estimate. Auxiliary information is generated using Slepian-Wolf encoding and sent to Bob via a public channel. Alice and Bob then negotiate and generate key K2 with a key rate of:

[0062]

[0063] Note that K2 is kept secret from both Relay and Eve simultaneously, therefore K2 is also known as the private key, which has higher security constraints.

[0064] Therefore, Alice and Bob combine K1, generated from the first negotiation, and K2, generated from the second negotiation, to obtain the final key K. s and private key K p Their rates are R s and R p :

[0065]

[0066]

[0067] Experimental Verification: The method of this invention was verified using the Matlab simulation platform. Assuming coherence time T = 20, path loss coefficient l = 3, and noise variance... Alice and Bob are 2m and 3m away from Relay, respectively. During the simulation process, Figure 4 and Figure 6 The correlation coefficient ρ = 0.2. Figure 5 and Figure 6 The signal-to-noise ratios (SNR) are SNR=20dB and SNR=30dB, respectively.

[0068] Figure 4 The simulation shown is a curve depicting the key rate as a function of the signal-to-noise ratio. From... Figure 4 As can be seen from the curve, the total key rate of the key generation method proposed in this invention is equal to the sum of the rates of the first two negotiations, and the total private key rate is equal to the private key rate of the second negotiation; and the rate will increase with the increase of the signal-to-noise ratio.

[0069] Figure 5 The simulation shown is a curve depicting the key rate as a function of the correlation coefficient. From... Figure 5 As can be seen from the above, the total key rate and private key rate of the method proposed in this invention both decrease or remain unchanged as the correlation coefficient increases. When the correlation coefficient is greater than 0.5, the key rate and private key rate are the same.

[0070] The method proposed in this invention can also be extended to multi-relay scenarios. Figure 6 The simulation shown illustrates the relationship between the key rate and the number of relay nodes. From... Figure 6 As can be seen from this, the key rate increases linearly with the increase in the number of relay nodes.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A physical layer collaborative key generation method based on dual key negotiation, characterized in that, The method specifically includes the following steps: S1: Channel estimation: In In the time slot, the relay node (Relay) sends the training sequence. Enable legitimate users Alice and Bob to obtain partial relay channel information; During the time slot, both legitimate users Alice and Bob simultaneously send the training sequence. This enables the relay node Relay and the eavesdropping node Eve to obtain superposition state information about the relay channel; S2: Key Negotiation: In the first negotiation, legitimate users Alice, Bob, and relay node Relay generate a key to keep secret from the eavesdropping node Eve using their respective channel estimates. The second negotiation extracts estimates for all relay channels and eavesdropping channels to generate a private key that is kept secret from both the relay node Relay and the eavesdropping node Eve. ; generated through two joint negotiations and The legitimate users Alice and Bob eventually generate the key. With private key ; During the initial negotiation, the relay node Relay used Slepian-Wolf coding and sent auxiliary information to Alice and Bob. By combining their own channel estimates with information transmitted on the common channel, the legitimate users Alice and Bob were able to completely recover the channel estimates from the relay node Relay. The legitimate user Alice obtains information by combining her own channel estimation values ​​with information from the public channel. Obtain the channel gain The estimated value Bob combines his own channel estimates with information obtained from the common channel. Obtain the channel gain The estimated value ; through channel estimation Legitimate users Alice and Bob generate a key that is kept secret from the eavesdropping node Eve. ; In the second negotiation, the legitimate user Alice used Slepian-Wolf coding, utilizing the channel gain. The observation information generates auxiliary information and is transmitted through a public channel. Legitimate users Alice and Bob generate private keys that are kept secret from both the relay node Relay and the eavesdropping node Eve. .

2. The physical layer cooperative key generation method according to claim 1, characterized in that, In step S1, the channel estimate is calculated using the following formula: ， ， in, This indicates that the legitimate user Alice has a positive effect on the channel gain. The estimated value, This indicates that legitimate user Bob has a positive effect on channel gain. The estimated value, , These represent the estimates of the relay channel by the relay node Relay and the eavesdropping node Eve, respectively. This represents the channel gain between the legitimate user Alice and the relay node Relay. This represents the channel gain between legitimate user Bob and relay node Relay. , These represent the channel gains from legitimate users Alice and Bob to the eavesdropping node Eve, respectively. , , , These are the signals received by the legitimate user Alice, the legitimate user Bob, the relay node Relay, and the eavesdropping node Eve, respectively. , , , The additive white Gaussian noise at the locations of legitimate user Alice, legitimate user Bob, relay node Relay, and eavesdropping node Eve are respectively. and The training sequences sent by the relay node Relay and legitimate users Alice and Bob in different time slots, and , Represents the transpose of a vector or matrix.

3. The physical layer cooperative key generation method according to claim 1, characterized in that, In step S2, the key generated through the first negotiation is combined. The private key generated in the second negotiation The key ultimately generated by legitimate users Alice and Bob for The generated private key for .

Images