A physical layer identity authentication method and device of a network time synchronization system

Abstract

The application provides a physical layer identity authentication method and device of a network time synchronization system, and belongs to the technical field of network security, and specifically comprises the following steps: encoding and modulating a sending end original time synchronization message, and inserting an identity authentication pilot sequence; performing pre-distortion of the message signal related to the sending end identity; performing step-by-step equalization of the message signal by each device based on a pre-measured and stored frequency domain channel matrix of a physical channel; a security demodulator extracts a current sub-block identity authentication pilot sequence, estimates, interpolates and stores the pre-distorted sequence, and then performs frequency domain equalization, demodulation and decoding on the current sub-block data sub-carrier by using the pre-distorted sequence to recover the original time synchronization message and a message security indication vector; identity authentication is judged based on the error code rate of the message security indication vector, and the original time synchronization message is sent to a receiving end or discarded according to the judgment. The scheme of the application effectively overcomes external and internal attackers, and realizes time source identity authentication.

Description

Technical Field

[0001] This application relates to the field of network security technology, and in particular to a physical layer authentication method and apparatus for a network time synchronization system. Background Technology

[0002] Network time reference is one of the main forms of modern time reference, widely used in distributed, networked systems that rely on a unified time reference. Network time synchronization is a crucial link in establishing and maintaining a network time reference. Time synchronization refers to the state in which the clock faces of all clocks in a system reach a consistent state. Network time synchronization technology uses private or public networks as a medium, and through message exchange between the clock to be synchronized and the reference clock, it synchronizes the clocks of all computers and devices connected to the synchronization network to a common reference time. Because the network environment is insecure and unreliable, and not all time-frequency terminals that exchange time synchronization messages are trustworthy, network time synchronization has become the most vulnerable link in the establishment and maintenance of a network time reference. Attacks on the time synchronization system can cause the target application system to synchronize to an incorrect time source, reduce time synchronization accuracy, or lose time synchronization, which can interfere with and disrupt the normal operation and function of the entire application system, and even lead to system paralysis, thus causing significant losses to national defense, military, social politics, and economic and social fields.

[0003] Typical network time synchronization protocols include Network Time Protocol (NTP) and Precision Time Protocol (PTP, IEEE 1588). NTP is designed for large, dynamic, variable-delay packet-switched networks, with message exchange based on the connectionless UDP protocol. PTP is designed for infrastructure networks, requiring dedicated hardware for network construction and the implementation of well-controlled packet-switched networks. PTP overcomes the time synchronization accuracy limitations of NTP, providing high-precision time synchronization services for precision instruments, industrial automation, and military applications. Both NTP and PTP protocols are continuously evolving and improving, with security issues becoming a key focus of the working groups in recent years. The latest NTP security protocol is Network Time Security (NTS) (RFC 8915), released in 2020, while the latest PTP security standardization achievement is Appendix P of the upgraded version IEEE 1588v2.1, released in 2019.

[0004] Time source spoofing attacks occur when an attacker impersonates a server / master clock in the network and manipulates the target client / slave clock's local time to a false time set by the attacker through time synchronization message exchanges. This type of attack is one of the important methods for implementing "time spoofing." An effective way to suppress time source spoofing attacks is to perform data source authentication on the received time synchronization data packets at the receiving end. While the latest security mechanism standardization achievements of NTP and PTP show that current NTP and PTP security mechanisms have authentication functions, they still have vulnerabilities and limitations and cannot completely suppress time source spoofing attacks. These limitations are mainly reflected in the following aspects:

[0005] (1) For NTP, according to RFC 8915, in peer mode and control mode, mutual authentication is provided through the DTLS protocol; in client / server mode, the first phase completes the initial authentication between the client and server through the TLS protocol, and the second phase provides mutual authentication during message exchange through the AEAD_AES_SIV_CMAC_256 message authentication code. The first phase authentication is only performed once when the NTP connection is established, and the second phase authentication can be performed repeatedly while the connection is not closed; in broadcast mode, no encryption authentication algorithm has been defined. The TESLA-like algorithm, as the most important alternative, is vulnerable to latency attacks and has security vulnerabilities. In summary, the current NTP authentication mechanism still has the following problems: 1) It does not cover all working modes. The message exchange parties in peer mode, control mode, and client / server mode have one-to-one secure authentication, while the broadcast mode has not defined an effective authentication mechanism; 2) It adopts the external security mechanism TLS, which has integration problems with the NTP protocol. On the one hand, the latency and latency changes of the TLS security stack significantly reduce the time synchronization accuracy. On the other hand, due to limitations in efficiency and the number of keys, the TLS mechanism is only applicable to unicast. If applied to broadcast mode, it can only provide group-level authentication functionality.

[0006] (2) For PTP, IEEE 1588v2.1 defines two types of authentication mechanisms: internal and external. The internal mechanism is based on the Security Tag Length Value (SecurityTLV) and the Message Integrity Verification Value (ICV), and proposes two key management strategies. The instantaneous key sharing strategy using GDOI can only provide group-level authentication, while the delayed key sharing strategy using TESLA has data source authentication functionality. However, the security of this strategy depends on time synchronization between the sender and receiver, making it vulnerable to targeted delay attacks. The external mechanism includes all available security mechanisms outside of the PTP protocol, with MACSec and IPSec being typical examples. However, similar to TLS, IPSec also has integration issues with the PTP protocol, such as significantly reducing time synchronization accuracy, only providing group-level authentication, and incompatibility with PTP transparent clock components. The MACSec mechanism is applied at layer 2, and therefore can only provide security protection for the local network. In addition, this mechanism is based on the hop-by-hop encryption method, thus requiring every node in the network to be trusted.

[0007] Based on the current standardization status of NTP and PTP security mechanisms, from the perspective of suppressing time-source spoofing attacks, NTP unicast mode (peer mode, control mode, client / server mode) using DTLS / TLS combined with message authentication codes has the ability to suppress time-source spoofing attacks, but there is a problem of integration between external security mechanisms and the NTP protocol; as for NTP broadcast mode and PTP (whose master-slave clock working mode has inherent broadcast attributes), currently only the TESLA algorithm has the ability to authenticate data sources, but this algorithm still has security vulnerabilities in the current security mechanism protocol framework.

[0008] As the above analysis shows, existing protocols mainly rely on external security mechanisms and Message Authentication Codes (NTP unicast) and the TESLA algorithm (PTP) to suppress time source spoofing attacks, but all of these have certain problems. Essentially, the main contradiction that network time synchronization protocols need to consider when defining data source authentication strategies is security performance versus complexity, i.e., the impact on time synchronization accuracy. Message Authentication Codes and the TESLA algorithm, with their lower complexity, have limitations in terms of coverage of working modes and security. While asymmetric key-based digital signature algorithms offer higher security, their high computational complexity (2-3 orders of magnitude higher than Message Authentication Codes), significant processing delays, and delay uncertainties prevent their application in network time synchronization systems. In other words, these algorithms cannot currently solve the problem of data source authentication in network time synchronization. In conclusion, using cryptographic techniques to address the problem of suppressing time source spoofing attacks will always face the aforementioned contradictions. Summary of the Invention

[0009] In view of this, embodiments of this application provide a physical layer authentication method and apparatus for a network time synchronization system. It employs physical layer technology to solve the data source authentication problem, covering all NTP / PTP operating modes, achieving a reasonable trade-off between security performance and computational complexity, and providing a new approach to suppressing time source spoofing attacks in network time synchronization. The network time synchronization physical layer authentication method provided by this invention differs from cryptographic authentication technologies such as digital signature technology and message authentication code technology. It achieves time source authentication of legitimate messages by utilizing the different signal characteristics of message signals sent by legitimate time sources and those sent by illegitimate time sources (attackers), as well as the different channel transmission characteristics they undergo before reaching the legitimate receiver.

[0010] In a first aspect, embodiments of this application provide a physical layer authentication method for a network time synchronization system. The network time synchronization system includes a sending end and multiple receiving ends communicating with the sending end. Each receiving end is connected to the sending end via a physical link, which is used for transmitting network time synchronization messages. The sending end is connected to a secure coding modulator, and the receiving ends are connected to a secure demodulation decoder. A multi-level secure network infrastructure is provided between the secure coding modulator and the secure demodulation decoder. The secure network infrastructure includes a channel simulation system and secure switching equipment. The method comprises:

[0011] The secure coding modulator encodes and modulates the original time synchronization message sent by the transmitting end to generate a first message signal. Messages sent to different receiving ends are carried by different sub-blocks of the first message signal.

[0012] An authentication pilot sequence is inserted into the first message signal to obtain a second message signal. Different authentication pilot sequences are used for the sub-blocks corresponding to different sending ends and receiving ends.

[0013] The second message signal is pre-distorted according to the identity of the sending end using a distortion sequence to obtain the third message signal. Different sending ends use different distortion sequences.

[0014] The third message signal is leveled up by means of the multi-level security network infrastructure and the security demodulation decoder, based on the frequency domain channel matrix of the physical channels measured and stored at each level in advance.

[0015] The secure demodulation decoder extracts the authentication pilot sequence of the target sub-block, estimates, interpolates and stores the pre-distorted sequence based on the extracted authentication pilot sequence, and then uses the pre-distorted sequence to perform frequency domain equalization on the data subcarriers of the target sub-block.

[0016] The frequency-domain equalized message signal is demodulated and decoded to recover the original time synchronization message and message security indication vector;

[0017] An authentication decision is made based on the bit error rate of the message security indication vector, and the original time synchronization message is sent to the receiving end or discarded according to the decision result.

[0018] According to a specific implementation of an embodiment of this application, the formula for determining the identity authentication pilot sequence is as follows:

[0019]

[0020] 0≤n≤N ZC -1,0 <N CS ≤N ZC ,0≤v≤N SPR -1,

[0021] Where, x u,v [n] represents the identity authentication pilot sequence, N ZC Let u be the sequence length, u be the root, and vN be the number of substrings. CS N is the number of signs that are cyclically shifted from the basic sequence (v=0). CS N is the interval for the cyclic shift. SPR The number of sequences in the sequence group obtained by cyclic shifting refers to the fact that the u of the authentication pilot sequence used by different receivers corresponding to the same transmitter is the same, but the v is different.

[0022] According to a specific implementation of an embodiment of this application, the step of inserting an authentication pilot sequence into the first message signal to obtain the second message signal includes:

[0023] The identity authentication pilot sequence is inserted into the subcarrier of the first message signal at fixed intervals to obtain the second message signal;

[0024] If N ZC If the number of authentication pilots required for the current sub-block is greater than the number of authentication pilots needed, then the actual generated authentication pilot sequence x... u,v [n] is truncated;

[0025] If N ZC If the number of authentication pilots is less than the number required for the current sub-block, then the actual generated authentication pilot sequence x... u,v [n] is expanded in a loop until the requirement is met.

[0026] According to a specific implementation of an embodiment of this application, the pre-distortion formula is as follows:

[0027] S k =Xk ·J k ,1≤k≤N dp ,

[0028] Where k is the index of the data subcarrier and the authentication pilot subcarrier of the second message signal, and N dp J represents the number of data subcarriers and authentication pilot subcarriers in the second message signal. k For the distorted sequence, X k The second message signal carries the original data and authentication pilot signal, S k This is a pilot for distorted data and identity authentication.

[0029] According to a specific implementation of an embodiment of this application, the method further includes:

[0030] The distorted sequence is determined by a specific frequency selective channel model. Each transmitter pre-establishes its own distorted sequence library and periodically changes the distorted sequence used.

[0031] The insertion of the authentication pilot sequence is set to be periodic and corresponds to the transformation period of the distorted sequence. The authentication pilot sequence is inserted only in the message signal where the distorted sequence is transformed.

[0032] The secure demodulation decoder at each legitimate receiver estimates the current distortion sequence in the corresponding sub-block of the transmitter based on the authentication pilot sequence in the corresponding sub-block of the message signal. (k∈Subblock i), and the current distorted sequence (k∈Subblock i) are stored in the locally pre-established distorted sequence library;

[0033] According to the current malformation sequence (k∈Subblock i) Equalize the data subcarriers of this subblock. During the maintenance of this distortion sequence, the message signal is not inserted with the authentication pilot sequence. The authentication pilot subcarriers are modulated with a random symbol sequence, or modulated with a second type of pilot sequence for channel state information estimation of the physical channel. Each secure demodulator / decoder determines the channel state information based on the currently stored data. (k∈Subblock i) Equalize the received signal.

[0034] According to a specific implementation of this application, the third message signal is subjected to step-by-step equalization based on the frequency domain channel matrix of the physical channels pre-measured and stored at each level, wherein the frequency domain channel matrix is ​​N. i,all ×N i,all order, N i,allLet i be the total number of subcarriers contained in sub-block i. The diagonal elements of each frequency domain channel matrix are the channel frequency responses of each subcarrier, and the other elements of each frequency domain channel matrix are 0.

[0035] The equilibrium can be, but is not limited to, zero-forcing equilibrium, and the zero-forcing equilibrium process is as follows:

[0036]

[0037] Among them, S q,j This is the j-th level balanced sequence; H represents the frequency domain channel matrix of the j-th level obtained after channel estimation; j Let be the frequency domain channel matrix of the j-th level physical channel; N is the frequency domain symbol vector carried by the third message signal sub-block i. i,all N represents the total number of subcarriers contained in subblock i; i The number of message signals that need to be leveled sequentially before reaching the receiving end; (·) H This is the conjugate transpose operation.

[0038] According to a specific implementation of an embodiment of this application, the method further includes:

[0039] Channel estimation and message signal transmission of each device are performed in a time-division manner, requiring channel estimation to precede message transmission. The result of each channel estimation can only be applied to the physical channel equalization of subsequent message signals. Each device includes the channel simulation system and the secure demodulation decoder.

[0040] Each device performs channel estimation based on dedicated pilot signals during the interval of message signal transmission, and / or performs channel estimation based on the second type of pilot sequence modulated in the message signal.

[0041] Each device establishes a channel state information database, storing the most recently measured channel state information between itself and all possible other devices with which it interacts. Before the next channel estimation, the channel state information database will maintain the current channel state information, and the channel estimation period will be dynamically adjusted by each device according to the actual physical channel conditions.

[0042] According to a specific implementation of an embodiment of this application, the method further includes:

[0043] Two bits are added to the end of the transmitter identification code as a pilot identifier. The transmitter identification code is carried by the first cluster of each sub-block. The pilot identifier is used to indicate whether the current message signal is inserted with the authentication pilot sequence, the second type of pilot sequence, or the random symbol sequence.

[0044] According to a specific implementation of an embodiment of this application, the step of performing an authentication decision based on the bit error rate of the message security indication vector, and sending the original time synchronization message to the receiving end or discarding it according to the decision result, includes:

[0045] If the bit error rate of the message security indication vector is lower than a preset threshold, the time synchronization message is determined to be secure. After source identity authentication, the secure demodulator sends the recovered original time synchronization message to the receiving end.

[0046] If the bit error rate of the security indication vector of the message reaches or exceeds a preset threshold, the time synchronization message is determined to be illegal. The security demodulation decoder will discard the time synchronization message and inform the receiver that the time synchronization message has been discarded and the security network infrastructure it has passed through. The physical channel state information updated based on the time synchronization message will be deleted, and the previous channel estimation result will be maintained.

[0047] Secondly, embodiments of this application also provide a physical layer authentication device for a network time synchronization system. The network time synchronization system includes a sending end and multiple receiving ends communicating with the sending end. Each receiving end is connected to the sending end via a physical link, which is used for transmitting network time synchronization messages. The sending end is connected to a secure coding modulator, and the receiving end is connected to a secure demodulation decoder. A multi-level secure network infrastructure is provided between the secure coding modulator and the secure demodulation decoder. The secure network infrastructure includes a channel simulation system and secure switching equipment. The device comprises:

[0048] The encoding and modulation module is used to encode and modulate the original time synchronization message sent by the transmitting end to generate a first message signal. Messages sent to different receiving ends are carried by different sub-blocks of the first message signal.

[0049] An authentication pilot sequence insertion module is used to insert an authentication pilot sequence into the first message signal to obtain a second message signal. Different authentication pilot sequences are used for the sub-blocks corresponding to different sending ends and receiving ends.

[0050] The distortion module is used to perform pre-distortion of the second message signal related to the identity of the sending end using a distortion sequence to obtain the third message signal, wherein different sending ends use different distortion sequences;

[0051] The secure network infrastructure progressive equalization module is used to progressively equalize the third message signal based on the frequency domain channel matrix of the physical channels pre-measured and stored at each level through the multi-level secure network infrastructure and the secure demodulation decoder.

[0052] The distortion equalization module is used to extract the identity authentication pilot sequence of the target sub-block, estimate, interpolate and store the pre-distortion sequence based on the extracted identity authentication pilot sequence, and then use the pre-distortion sequence to perform frequency domain equalization on the data subcarriers of the target sub-block.

[0053] The demodulation and decoding module is used to demodulate and decode the frequency-domain equalized message signal to recover the original time synchronization message and message security indication vector.

[0054] The identity authentication decision module is used to make an identity authentication decision based on the bit error rate of the message security indication vector, and send the original time synchronization message to the receiving end or discard it according to the decision result.

[0055] Beneficial effects

[0056] The physical layer authentication method and apparatus for the network time synchronization system in this application overcomes the limitations of current authentication methods in NTP / PTP security mechanisms and solves the problem of suppressing time source spoofing attacks. Based on dual authentication of the physical characteristics of the source signal and the signal transmission link, this method can effectively overcome external and internal attackers and achieve time source authentication. This method is applicable to unicast and broadcast, single-link and multi-sublink scenarios, and is not limited by operating mode, operating scenario, or channel environment. Furthermore, this method solves the source authentication problem at the physical layer, employing techniques with low complexity and low processing latency, having minimal impact on time synchronization accuracy; it does not rely on system time synchronization; and it is compatible with current NTP / PTP security mechanisms. Attached Figure Description

[0057] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a network time synchronization system architecture diagram according to an embodiment of the present invention;

[0059] Figure 2(a) is a schematic diagram of the spectrum of the pre-insertion message signal for identity authentication pilot according to an embodiment of the present invention;

[0060] Figure 2(b) is a schematic diagram of the message signal spectrum after the identity authentication pilot is inserted according to an embodiment of the present invention. Detailed Implementation

[0061] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0062] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. This application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0063] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this application, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.

[0064] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. The drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0065] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0066] This application provides a physical layer authentication method for a network time synchronization system, which is described below. Figure 1 See Figure 2 for a detailed description.

[0067] In one embodiment, refer to Figure 1The network time synchronization system based on this embodiment includes a sending end and multiple receiving ends communicating with the sending end. Each receiving end is connected to the sending end via a physical link for transmitting network time synchronization messages. The sending end is connected to a secure coding modulator, and the receiving ends are connected to a secure demodulation decoder. A multi-level secure network infrastructure is provided between the secure coding modulator and the secure demodulation decoder. The secure network infrastructure includes a channel simulation system and secure switching equipment. The physical layer authentication method of the network time synchronization system includes:

[0068] Step S101: The secure coding modulator encodes and modulates the original time synchronization message sent by the transmitting end to generate a first message signal. Messages sent to different receiving ends are carried by different sub-blocks of the first message signal.

[0069] Step S102: Insert an authentication pilot sequence into the first message signal to obtain a second message signal. Different authentication pilot sequences are used for the sub-blocks corresponding to different sending ends and receiving ends.

[0070] Step S103: The second message signal is pre-distorted according to the identity of the sending end using a distortion sequence to obtain the third message signal. Different sending ends use different distortion sequences.

[0071] Step S104: Through the multi-level security network infrastructure and the security demodulation decoder, the third message signal is equalized step by step based on the frequency domain channel matrix of the physical channels pre-measured and stored at each level;

[0072] Step S105: The secure demodulation decoder extracts the authentication pilot sequence of the target sub-block, estimates, interpolates and stores the pre-distorted sequence based on the extracted authentication pilot sequence, and then uses the pre-distorted sequence to perform frequency domain equalization on the data subcarriers of the target sub-block.

[0073] Step S106: Demodulate and decode the frequency-domain equalized message signal to recover the original time synchronization message and message security indication vector;

[0074] Step S107: Make an identity authentication decision based on the bit error rate of the message security indication vector, and send the original time synchronization message to the receiving end or discard it according to the decision result.

[0075] In practice, the method of this embodiment is based on a network time synchronization system and uses existing signal processing technology for sending and receiving ends to expand the identity authentication function.

[0076] This method for secure transmission of network time synchronization messages in arbitrary channel environments addresses the problem of suppressing typical forwarding-based time attacks (delay attacks, timestamp tampering attacks). It fully considers the more general and network-transmission-compatible broadcast and multi-sublink transmission modes in network time synchronization message transmission, as well as the arbitrary channel environment with maximized applicability. The method proposes Polar coding and OFDM-IM modulation of the original time synchronization message at the sending end, utilizing the polarization characteristics of Polar codes to construct a secure bit channel, enabling secure transmission of protected information bits such as timestamps. OFDM-IM modulation is used to implement orthogonal frequency division multiple access and physical layer routing control for time synchronization messages sent to various legitimate receivers. Combined with the system infrastructure and its operating mechanism, message transmission quality control is implemented, ensuring that, under any channel quality comparison between legitimate and illegitimate channels (including cases where the illegitimate channel's quality is superior to the legitimate channel's), the transmission quality of legitimate messages is significantly better than that of illegitimate messages (attacked messages), satisfying the aforementioned conditions for constructing a secure bit channel.

[0077] This invention inherits the aforementioned secure transmission method, focusing on time source authentication. For simplicity, the working scenario is simplified to unicast (one-to-one), single link, with no sub-block division in the frequency domain of the corresponding OFDM-IM message signal, containing 4 clusters numbered 1, 2, 3, and 4. Cluster 1 carries a sender identifier (pre-assigned, unique in the system) in its IM data; cluster 2 carries a receiver identifier (pre-assigned, unique in the system); cluster 3 carries routing information, i.e., sub-link identifiers (pre-assigned, uniquely identifying each sub-link in the network); and cluster 4 carries sub-link channel simulation system indication data. This data indicates the identifiers (pre-assigned, different channel simulation system identifiers correspond to different transfer functions) and traversal order (according to the order of the identifiers) that the time synchronization message must pass through on that sub-link. Each cluster contains n data subcarriers, and the active subcarrier index of each cluster is determined. The active subcarriers of the first three clusters are dedicated to carrying link security information (see invention patent, application number 202211665095.X). The active subcarriers of cluster 4 are used to carry network time synchronization message data symbols and other symbols used to enhance security functions (such as pseudo-data symbols), which are also referred to as data subcarriers in this invention. It should be noted that the physical layer authentication method of the network time synchronization system proposed in this invention can be directly extended to broadcast (one-to-many) and multi-sublink (actual network) scenarios, without limitations in terms of working mode, working scenario, or channel environment. The relevant content will be described in a generalized manner below.

[0078] In one embodiment, the formula for determining the identity authentication pilot sequence is:

[0079]

[0080] 0≤n≤N ZC -1,0 <N CS ≤N ZC ,0≤v≤N SPR -1,

[0081] Where, x u,v [n] represents the identity authentication pilot sequence, N ZC Let u be the sequence length, u be the root, and vN be the number of substrings. CS N is the number of signs that are cyclically shifted from the basic sequence (v=0). CS N is the interval for the cyclic shift. SPR The number of sequences in the sequence group obtained by cyclic shifting refers to the fact that the u of the authentication pilot sequence used by different receivers corresponding to the same transmitter is the same, but the v is different.

[0082] In one embodiment, inserting an authentication pilot sequence into the first message signal to obtain the second message signal includes:

[0083] The identity authentication pilot sequence is inserted into the subcarrier of the first message signal at fixed intervals to obtain the second message signal;

[0084] If N ZC If the number of authentication pilots required for the current sub-block is greater than the number of authentication pilots needed, then the actual generated authentication pilot sequence x... u,v [n] is truncated;

[0085] If N ZC If the number of authentication pilots is less than the number required for the current sub-block, then the actual generated authentication pilot sequence x... u,v [n] is expanded in a loop until the requirement is met.

[0086] In one embodiment, the pre-distortion formula is:

[0087] S k =X k ·J k ,1≤k≤N dp ,

[0088] Where k is the index of the data subcarrier and the authentication pilot subcarrier of the second message signal, and N dp J represents the number of data subcarriers and authentication pilot subcarriers in the second message signal. k For the distorted sequence, Xk The second message signal carries the original data and authentication pilot signal, S k This is a pilot for distorted data and identity authentication.

[0089] In one embodiment, the method further includes:

[0090] The distorted sequence is determined by a specific frequency selective channel model. Each transmitter pre-establishes its own distorted sequence library and periodically changes the distorted sequence used.

[0091] The insertion of the authentication pilot sequence is set to be periodic and corresponds to the transformation period of the distorted sequence. The authentication pilot sequence is inserted only in the message signal where the distorted sequence is transformed.

[0092] The secure demodulation decoder at each legitimate receiver estimates the current distortion sequence in the corresponding sub-block of the transmitter based on the authentication pilot sequence in the corresponding sub-block of the message signal. (k∈Subblock i), and the current distorted sequence (k∈Subblock i) are stored in the locally pre-established distorted sequence library;

[0093] According to the current malformation sequence (k∈Subblock i) Equalize the data subcarriers of this subblock. During the maintenance of this distortion sequence, the message signal is not inserted with the authentication pilot sequence. The authentication pilot subcarriers are modulated with a random symbol sequence, or modulated with a second type of pilot sequence for channel state information estimation of the physical channel. Each secure demodulator / decoder determines the channel state information based on the currently stored data. (k∈Subblock i) Equalize the received signal.

[0094] In one embodiment, the third message signal is subjected to step-by-step equalization based on the frequency domain channel matrix of the physical channels pre-measured and stored at each level, wherein the frequency domain channel matrix is ​​N. i,all ×N i,all order, N i,all Let i be the total number of subcarriers contained in sub-block i. The diagonal elements of each frequency domain channel matrix are the channel frequency responses of each subcarrier, and the other elements of each frequency domain channel matrix are 0.

[0095] The equilibrium can be, but is not limited to, zero-forcing equilibrium, and the zero-forcing equilibrium process is as follows:

[0096]

[0097] Among them, S q,j This is the j-th level balanced sequence; H represents the frequency domain channel matrix of the j-th level obtained after channel estimation; j Let be the frequency domain channel matrix of the j-th level physical channel; N is the frequency domain symbol vector carried by the third message signal sub-block i. i,all N represents the total number of subcarriers contained in subblock i; i The number of message signals that need to be leveled sequentially before reaching the receiving end; (·) H This is the conjugate transpose operation.

[0098] In one embodiment, the method further includes:

[0099] Channel estimation and message signal transmission of each device are performed in a time-division manner, requiring channel estimation to precede message transmission. The result of each channel estimation can only be applied to the physical channel equalization of subsequent message signals. Each device includes the channel simulation system and the secure demodulation decoder.

[0100] Each device performs channel estimation based on dedicated pilot signals during the interval of message signal transmission, and / or performs channel estimation based on the second type of pilot sequence modulated in the message signal.

[0101] Each device establishes a channel state information database, storing the most recently measured channel state information between itself and all possible other devices with which it interacts. Before the next channel estimation, the channel state information database will maintain the current channel state information, and the channel estimation period will be dynamically adjusted by each device according to the actual physical channel conditions.

[0102] In one embodiment, the first message signal of each sub-block includes four clusters: the first cluster carries the sender identification code, the second cluster carries the receiver identification code, the third cluster carries the sub-link identification code, and the fourth cluster carries the sub-link channel simulation system indication data. The sub-link channel simulation system indication data is the data subcarrier, and the sub-link channel simulation system indication data is used to indicate the identification codes of the channel simulation systems that the time synchronization message needs to pass through on the current sub-link and the order in which they pass through.

[0103] In one embodiment, the method further includes:

[0104] Two bits are added to the end of the transmitter identification code as a pilot identifier. The transmitter identification code is carried by the first cluster of each sub-block. The pilot identifier is used to indicate whether the current message signal is inserted with the authentication pilot sequence, the second type of pilot sequence, or the random symbol sequence. The second type of pilot sequence is used for channel estimation, carried by the authentication pilot subcarrier, and transmitted in a time-division multiplexing manner with the authentication pilot sequence. When the authentication pilot subcarrier does not modulate any pilot sequence, it carries the random symbol sequence.

[0105] In one embodiment, the step of making an authentication decision based on the bit error rate of the message security indication vector, and sending the original time synchronization message to the receiving end or discarding it according to the decision result, includes:

[0106] If the bit error rate of the message security indication vector is lower than a preset threshold, the time synchronization message is determined to be secure. After source identity authentication, the secure demodulator sends the recovered original time synchronization message to the receiving end.

[0107] If the bit error rate of the security indication vector of the message reaches or exceeds a preset threshold, the time synchronization message is determined to be illegal. The security demodulation decoder will discard the time synchronization message and inform the receiver that the time synchronization message has been discarded and the security network infrastructure it has passed through. The physical channel state information updated based on the time synchronization message will be deleted, and the previous channel estimation result will be maintained.

[0108] The following is a detailed description using a specific embodiment.

[0109] The system architecture diagram of the network time synchronization physical layer authentication method designed in this embodiment is as follows: Figure 1 As shown. The method consists of three parts: a signal part, a transmission part, and a decision part, each composed of... Figure 1 The secure coding modulator (secure terminal infrastructure), secure network infrastructure, and secure demodulator / decoder (secure terminal infrastructure) are completed.

[0110] (I) Signal Section

[0111] The sending end generates a raw time synchronization message. The message bits are Polar encoded and OFDM-IM modulated in a secure coding modulator to generate a message signal. For detailed encoding and modulation processes, please refer to the invention patent application number 202211656673.3. To achieve physical layer authentication, this embodiment adds the following design to the above message signal:

[0112] (1) Insert an identity authentication pilot on the basis of the existing message signal spectrum.

[0113] The purpose of inserting the authentication pilot is to enable the secure demodulation decoder at the legitimate receiver to estimate and frequency-domain equalize the transmitter identity-related pre-distortion implemented by the secure coding modulator at the transmitter. The authentication pilot uses a Zadoff-Chu (ZC) sequence (also known as a generalized chirp-like sequence). The ZC sequence is a non-binary unit amplitude sequence that satisfies the constant amplitude zero autocorrelation (CAZAC) property. Since the amplitude is constant, the ZC sequence only needs to calculate and store the phase, effectively simplifying the implementation process. In this invention, the authentication pilot is determined by equation (1):

[0114]

[0115] In the formula, N ZC Let N be the sequence length, and assume the number of transmitters in the system is N. T The maximum number of receivers connected to each transmitter is N. R In this embodiment, N is taken as ZC For is greater than or equal to max(N) T, N R The smallest prime number of N+1, where max(a,b) refers to the larger of a and b. u is the root, and according to the properties of the ZC sequence, when N... ZC When u is a prime number, the values ​​of u can be 1, ..., N. ZC -1, total N ZC -1, corresponding to N ZC -1 sets of sequences, from which N are selected. T Groups, respectively assigned to N T There are multiple transmitters, therefore the authentication pilot sequences used by different transmitters are different. CS It is the number of signs obtained by cyclically shifting the basic sequence (v=0). According to the properties of ZC sequences, the sequence group obtained by cyclically shifting the same basic ZC sequence (with the same u, 0) is... <N CS ≤N ZC ,0≤v≤N SPR -1, ), and they have zero autocorrelation with each other. CS N is the interval for the cyclic shift. SPR Let N be the number of sequences in the sequence group obtained through cyclic shifting. In this embodiment, let N be... SPR The number of receivers connected to the corresponding transmitter, and then according to Calculate N CS The N obtained by cyclically shifting the basic ZC sequence assigned to this transmitter is... SPREach sequence, known as an authentication pilot sequence, is inserted into the message signal sent to each legitimate receiver (for the same sender, the authentication pilot sequence u is the same for different receivers, but v is different). A specific authentication pilot sequence is known only to the corresponding message exchange terminals. To increase the number of authentication pilots and improve selectivity and confidentiality, the sequence length N can also be appropriately increased. ZC .

[0116] Figure 2 shows a schematic diagram of the message signal spectrum before and after the insertion of the authentication pilot. In the figure, the subcarrier of sub-block i is used to carry message data, routing information, link security information, etc., destined for receiver i, as shown in Figure 2(a). The authentication pilot is inserted into the atomic carrier at fixed intervals based on the spectral structure of atomic block i, as shown in Figure 2(b). The insertion position of the authentication pilot (at the start point and interval of each sub-block) is predetermined by the system, with the principle of saving spectrum resources as much as possible while satisfying equalization quality.

[0117] Figure 2 shows the message signal spectrum before and after the identity authentication pilot insertion: (a) before insertion; (b) after insertion, such as N. ZC If the number of authentication pilots required for the sub-block is greater than the number of authentication pilots needed, then the actual generated authentication pilot sequence x... u,v [n] is used for truncation; e.g., N ZC If the number of authentication pilots required for the sub-block is less than the actual number of authentication pilot sequences x, then the generated authentication pilot sequence x... u,v [n] is expanded using a loop to meet actual needs.

[0118] (2) Perform pre-distortion of the message signal spectrum related to the sender's identity.

[0119] After the frequency domain signal is generated, the message signal spectrum shown in Figure 2(b) is pre-distorted based on the sender's identity. The distortion range includes the data subcarriers and the identity authentication pilot subcarriers of each sub-block. The distortion process is shown in Equation (2):

[0120] S k =X k ·J k ,1≤k≤N dp (2)

[0121] In the formula, k represents the index of the data subcarrier and the authentication pilot subcarrier of the message signal, arranged sequentially starting from the first sub-block. N dp This refers to the number of data subcarriers and authentication pilot subcarriers in the message signal. k (1≤k≤N dpX represents a distorted sequence, which is sender-specific; different senders use different distorted sequences. The distorted sequence can be determined through a specific frequency-selective channel model. Each sender can pre-build its own distorted sequence library and periodically change the distorted sequence it uses. k S k (1≤k≤N dp These are the original data and authentication pilot signals carried by the message signal, and the distorted data and authentication pilot signals, respectively.

[0122] Corresponding to the transformation period of the distorted sequence, the insertion of the authentication pilot is also designed to be periodic. That is, the authentication pilot is inserted only in the message signal where the distorted sequence has transformed. The secure demodulation decoder of each legitimate receiver estimates the current distorted sequence in the corresponding sub-block of the transmitter based on the authentication pilot in the corresponding sub-block of the message signal. (k∈Subblock i), and store it in the local pre-distorted sequence library. (k∈Subblock i) Equalize the data subcarriers of this subblock. During the maintenance of this distortion sequence, the message signal is not inserted with authentication pilots. The authentication pilot subcarriers are modulated with random symbol sequences, and if necessary, second-type pilot sequences (assigned by the system to each transmitter) can also be modulated for channel state information estimation of the physical channel (the purpose of which is described in the transmission section below). Each secure demodulator / decoder uses the currently stored... (k∈Subblock i) Equalize the received signal. Whether an authentication pilot has been inserted into a message signal can be identified by adding two bits to the end of the sender's identifier code as an identifier, which indicates whether the current message signal has an authentication pilot sequence, a second type pilot sequence, or a random symbol sequence inserted. The above two bits are carried by the IM data of each subblock cluster 1.

[0123] The pre-distorted message signal is transmitted to each legitimate receiving end via the network.

[0124] (II) Transmission Section

[0125] To mitigate the impact of the actual physical channel on the message signal, the various levels of secure network infrastructure and secure demodulation decoders that the message signal passes through will perform step-by-step equalization of the message signal, such as... Figure 1 As shown in the figure. H 1, …,H Ni Here is the frequency domain channel matrix for each level of the physical channel, and the matrix is ​​N. i,all ×N i,all order (N) i,allLet H1 be the total number of subcarriers contained in subblock i, and let the diagonal elements be the channel frequency response of each subcarrier, with other elements being 0. The influence of the initial channel simulation system on H1 is balanced, and the influence of the subsequent channel simulation system on H2 is balanced, until the secure demodulation decoder... The effects of the equalization process are balanced. Step-by-step equalization requires knowledge of the channel state information between itself and the previous level device, which can be obtained through periodic channel estimation. The equalization process is shown in equation (3) (taking zero-forcing equalization as an example):

[0126]

[0127] In the formula, S q,j This is the j-th level balanced sequence; This is the frequency domain channel matrix of the j-th level obtained after channel estimation; N is the frequency domain symbol vector carried by sub-block i in the pre-distorted message signal; i Let be the number of step-by-step equalization steps required to transmit the message signal to receiver i. H This refers to the conjugate transpose operation.

[0128] Channel estimation by each device (end-channel simulation system, channel simulation system, secure demodulation decoder) is performed in a time-division multiplexing manner with message signal transmission. Channel estimation must precede message transmission; that is, the result of each channel estimation can only be applied to the physical channel equalization of subsequent message signals. Each device can perform channel estimation based on dedicated pilot signals during message signal transmission intervals, or based on the modulated Type II pilot sequence in the message signal, or a combination of both methods. Each device establishes a channel state information database, storing the most recently measured channel state information between itself and all possible other devices with which it interacts. The channel state information database maintains the current channel state information until the next channel estimation. The channel estimation period is dynamically adjusted by each device according to the actual physical channel conditions.

[0129] (III) Judgment Section

[0130] The message signal arriving at the secure demodulator, except Besides the impact of physical channels, the effects of physical channels have been balanced. The secure demodulation decoder first performs physical channel equalization on the received signal, then extracts the authentication pilot sequence of the sub-block, and applies the pre-distorted sequence... (k∈Subblock i) are estimated, interpolated, and stored. For message signals without an inserted authentication pilot sequence, the corresponding sequence is retrieved from the local pre-distorted sequence library based on the sender's identifier. (k∈Subblocki). Then utilize... (k∈Subblock i) The data subcarriers of this subblock are frequency domain equalized (as shown in Equation (3) zero-forcing equalization) to eliminate the effects of pre-distortion. Then the message signal is demodulated by IM, OFDM and Polar decoding to recover the original network time synchronization message and message security indication vector (known in advance by both the sender and receiver). If the bit error rate of the message security indication vector is lower than the preset threshold, the received message is judged to be secure. After source identity authentication, the secure demodulator sends the network time synchronization message to the receiver. If the bit error rate of the message security indication vector reaches or exceeds the preset threshold, the received message is judged to be illegal. The secure demodulator will discard the network time synchronization message and inform the receiver and the security network infrastructure it has passed through, delete the physical channel state information updated based on the message, and continue to maintain the previous channel estimation results.

[0131] According to the aforementioned authentication decision mechanism, the transmission quality of the message security indication vector can only be guaranteed when the effects of pre-distortion and the physical channel are effectively balanced. The effects of the physical channel are eliminated through step-by-step equalization. Since the channel state information at each level is predetermined, the message signal must be transmitted to the secure demodulation decoder strictly according to the established route. Otherwise, there will be a deviation, even a significant deviation, between the actual traversed channel at a certain level and the pre-estimated channel state information used for equalization. This will introduce additional channel effects during the equalization process, causing signal distortion. The effects of pre-distortion are eliminated through pre-distortion sequence estimation, interpolation, and storage by the secure demodulation decoder, as well as subsequent frequency domain equalization. This requires that the estimated distortion sequence be consistent with or substantially consistent with the distortion sequence actually used by the transmitter. The premise for sequence consistency is that the influence of the physical channel on the authentication pilot is effectively balanced, and the distortion of the received signal is consistent with that at the transmitter. External pseudo-time sources (attackers) cannot know the distortion sequence used by the sender, cannot apply the same distortion to the pseudo-message signal as the real message signal, and cannot control the routing of the pseudo-message signal to be completely consistent with the real signal. Therefore, they are easily identified during authentication decisions. For internal pseudo-time sources (attackers), in the worst case, they can somehow obtain the distortion sequence used by the sender. They also know the correct transmission route, message signal structure, and generation method, and can generate pseudo-time messages with the same structure and distortion as the real message. However, due to the attacker's location, they still cannot control the actual routing of the pseudo-message signal to be completely consistent with the real signal. This causes the pseudo-message signal to be affected by additional channel interference, resulting in inconsistencies between the estimated distortion sequence and the actual distortion sequence used by the sender, or inconsistencies between the locally queried distortion sequence (previously estimated) and the actual distortion sequence (due to the superimposed channel interference). Furthermore, physical channel interference will also cause IM data to fail to demodulate correctly, which can also be effectively identified during authentication decisions.

[0132] Secondly, embodiments of this application also provide a physical layer authentication device for a network time synchronization system. The network time synchronization system includes a sending end and multiple receiving ends communicating with the sending end. Each receiving end is connected to the sending end via a physical link, which is used for transmitting network time synchronization messages. The sending end is connected to a secure coding modulator, and the receiving end is connected to a secure demodulation decoder. A multi-level secure network infrastructure is provided between the secure coding modulator and the secure demodulation decoder. The secure network infrastructure includes a channel simulation system and secure switching equipment. The device comprises:

[0133] The encoding and modulation module is used to encode and modulate the original time synchronization message sent by the transmitting end to generate a first message signal. Messages sent to different receiving ends are carried by different sub-blocks of the first message signal.

[0134] An authentication pilot sequence insertion module is used to insert an authentication pilot sequence into the first message signal to obtain a second message signal. Different authentication pilot sequences are used for the sub-blocks corresponding to different sending ends and receiving ends.

[0135] The distortion module is used to perform pre-distortion of the second message signal related to the identity of the sending end using a distortion sequence to obtain the third message signal, wherein different sending ends use different distortion sequences;

[0136] The secure network infrastructure progressive equalization module is used to progressively equalize the third message signal based on the frequency domain channel matrix of the physical channels pre-measured and stored at each level through the multi-level secure network infrastructure and the secure demodulation decoder.

[0137] The distortion equalization module is used to extract the identity authentication pilot sequence of the target sub-block, estimate, interpolate and store the pre-distortion sequence based on the extracted identity authentication pilot sequence, and then use the pre-distortion sequence to perform frequency domain equalization on the data subcarriers of the target sub-block.

[0138] The demodulation and decoding module is used to demodulate and decode the frequency-domain equalized message signal to recover the original time synchronization message and message security indication vector.

[0139] The identity authentication decision module is used to make an identity authentication decision based on the bit error rate of the message security indication vector, and send the original time synchronization message to the receiving end or discard it according to the decision result.

[0140] The embodiments provided by this invention overcome the limitations of current authentication methods in NTP / PTP security mechanisms and solve the problem of suppressing time source spoofing attacks. This method, based on dual authentication of the physical characteristics of the source signal and the signal transmission link, can effectively overcome external and internal attackers and achieve time source authentication. This method is applicable to unicast and broadcast, single-link and multi-sublink scenarios, and is not limited by operating mode, operating scenario, or channel environment. Furthermore, this method solves the source authentication problem at the physical layer, employing techniques with low complexity and low processing latency, minimizing the impact on time synchronization accuracy; it does not rely on system time synchronization; and it is compatible with current NTP / PTP security mechanisms.

[0141] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A physical layer authentication method for a network time synchronization system, the network time synchronization system comprising a sending end and a plurality of receiving ends in communication with the sending end, each of the receiving ends being provided with a physical link with the sending end, the physical link being used for network time synchronization message transmission, the sending end being connected with a secure coding modulator, the receiving ends being connected with secure demodulation decoders, a plurality of secure network infrastructures being provided between the secure coding modulator and the secure demodulation decoders, the secure network infrastructures comprising channel simulation systems and secure switching devices, characterized in that, The method comprises: The security coding modulator codes and modulates the original time synchronization message sent by the sending end to generate a first message signal, and the message sent to different receiving ends is carried by different sub-blocks of the first message signal; An identity authentication pilot sequence is inserted into the first message signal to obtain a second message signal, and different sub-blocks corresponding to different sending ends and receiving ends use different identity authentication pilot sequences; A distortion sequence is used to pre-distort the second message signal based on the identity of the sending end to obtain a third message signal, and different sending ends use different distortion sequences; The third message signal is equalized level by level based on the frequency domain channel matrix of the physical channel pre-measured and stored by the multi-level secure network infrastructure and the security demodulator; The security demodulator extracts the identity authentication pilot sequence of the target sub-block, estimates, interpolates and stores the pre-distortion sequence based on the extracted identity authentication pilot sequence, and then uses the pre-distortion sequence to perform frequency domain equalization on the data sub-carrier of the target sub-block; The message signal after frequency domain equalization is demodulated and decoded to recover the original time synchronization message and the message security indication vector; Identity authentication judgment is performed based on the error rate of the message security indication vector, and the original time synchronization message is sent to the receiving end or discarded according to the judgment result. 2.The method of claim 1, wherein, The determination formula of the identity authentication pilot sequence is: wherein x u,v [n] is the identity authentication pilot sequence, N ZC is the sequence length, u is the root, and vN CS is the number of symbols for cyclic shift of the basic sequence (v=0), N CS is the interval of cyclic shift, N SPR is the number of sequences in the sequence group obtained by cyclic shift, u of the identity authentication pilot sequence used by different receiving ends corresponding to the same sending end is the same, and v is different. 3.The method of claim 2, wherein, The identity authentication pilot sequence is inserted into the sub-carrier of the first message signal at a fixed interval to obtain a second message signal; The pre-distortion formula is: If N ZC is greater than the required number of identity authentication pilots for the current sub-block, the actual generated identity authentication pilot sequence x u,v [n] is truncated. If N ZC is less than the required number of identity authentication pilots for the current sub-block, the actual generated identity authentication pilot sequence x u,v [n] is cyclically extended until the requirement is met.

4. The method of claim 1, wherein the method further comprises: The method further comprises: S k = X k · J k , 1≤k≤N dp , Wherein, k is the data sub-carrier and authentication pilot sub-carrier index of the second message signal, N dp N is the number of data sub-carrier and authentication pilot sub-carrier in the second message signal, J k X is the distortion sequence, k S is the original data and authentication pilot carried by the second message signal, k is the distorted data and authentication pilot.

5. The method of claim 4, wherein the authentication comprises, The distortion sequence is determined by a specific frequency selective channel model, each sending end pre-establishes its own distortion sequence library, and periodically changes the distortion sequence used; The insertion of the identity authentication pilot sequence is set to be periodic and corresponds to the change period of the distortion sequence, and the identity authentication pilot sequence is inserted only in the message signal in which the distortion sequence changes; The equalization can use, but is not limited to, zero-forcing equalization, and the equalization process of the zero-forcing equalization is: The security demodulation decoder of each legal receiving end estimates the current scrambling sequence in the corresponding sub-block of the sending end according to the identity authentication pilot sequence in the corresponding sub-block of the message signal and stores the current scrambling sequence in the locally pre-established scrambling sequence library. According to the current scrambling sequence The data subcarriers of the sub-block are equalized, during the maintenance of the scrambling sequence, no authentication pilot sequence is inserted into the message signal, the authentication pilot subcarriers modulate a random symbol sequence, or modulate the second type of pilot sequence for channel state information estimation of the physical channel, and each security demodulator decodes according to the current stored The received signal is equalized.

6. The method of claim 5, wherein the authentication comprises: The third message signal is equalized step by step based on the pre-measured and stored frequency domain channel matrices of physical channels at different levels, wherein the frequency domain channel matrix is N i,all ×N i,all order, N i,all is the total number of subcarriers included in the subcarrier i, the diagonal element of each frequency domain channel matrix is the channel frequency response of each subcarrier, and other elements of each frequency domain channel matrix are 0. The method further comprises: wherein S q,j is the jth equalized sequence; is the jth frequency domain channel matrix obtained by channel estimation; H j is the jth frequency domain channel matrix of the physical channel; is the frequency domain symbol vector carried by the third subcarrier i of the subcarrier block, N i,all is the total number of subcarriers included in the subcarrier i; N i is the number of equalization levels required for the transmission of the message signal to the receiving end;(·) H is the conjugate transpose operation.

7. The method of claim 6, wherein the method further comprises: Channel estimation and message signal transmission of each device are performed at different times, and channel estimation is required to precede message transmission, and the result of each channel estimation can only be applied to physical channel equalization of subsequent message signals, and the devices include the channel simulation system and the security demodulator; Each device performs channel estimation based on special pilot signals during the interval of message signal transmission, and / or performs channel estimation based on the second type of pilot sequence modulated in the message signal; Each device respectively establishes a channel state information library to store the last measured channel state information between all possible other devices that have an interaction relationship with itself, and the channel state information library will maintain the current channel state information before the next channel estimation, and the channel estimation period is dynamically adjusted by each device according to the actual physical channel condition. ​ 8. The method of claim 5, wherein the physical layer authentication of the network time synchronization system is characterized by, The method further comprises: adding two bits as pilot identification at the tail of the sending-end identification code, the sending-end identification code being carried by the first cluster of each sub-block, and the pilot identification being used to indicate that the current message signal inserts the identity authentication pilot sequence, the second type of pilot sequence or the random symbol sequence.

9. The method of claim 1, wherein the method further comprises: The identity authentication decision based on the error rate of the message security indication vector includes: if the error rate of the message security indication vector is lower than a preset threshold, judging that the time synchronization message is secure, passing the source identity authentication, and sending the recovered original time synchronization message to the receiving end by the secure demodulation decoder; if the error rate of the message security indication vector reaches or is higher than the preset threshold, judging that the time synchronization message is illegal, discarding the time synchronization message by the secure demodulation decoder, and informing the receiving end that the time synchronization message has been discarded and the security network infrastructure, deleting the physical channel state information updated based on the time synchronization message, and continuing to maintain the last channel estimation result.

10. A physical layer authentication device of a network time synchronization system, the network time synchronization system comprising a sending end and a plurality of receiving ends in communication with the sending end, each of the receiving ends being provided with a physical link with the sending end, the physical link being used for network time synchronization message transmission, the sending end being connected with a secure coding modulator, the receiving end being connected with a secure demodulation decoder, a plurality of secure network infrastructures being provided between the secure coding modulator and the secure demodulation decoder, the secure network infrastructure comprising a channel simulation system and a secure switching device, characterized in that, The device comprises: an encoding and modulation module, configured to encode and modulate the original time synchronization message sent by the sending end to generate a first message signal, and to carry the messages sent to different receiving ends by different sub-blocks of the first message signal; an identity authentication pilot sequence insertion module, configured to insert an identity authentication pilot sequence into the first message signal to obtain a second message signal, and to use different identity authentication pilot sequences for the sub-blocks corresponding to different sending ends and receiving ends; a distortion module, configured to use a distortion sequence to pre-distort the second message signal in relation to the identity of the sending end to obtain a third message signal, and to use different distortion sequences for different sending ends; a security network infrastructure step-by-step equalization module, configured to perform step-by-step equalization on the third message signal based on the frequency domain channel matrix of the physical channel pre-measured and stored at each level through the multi-level security network infrastructure and the secure demodulation decoder; a distortion equalization module, configured to extract the identity authentication pilot sequence of a target sub-block, estimate, interpolate and store a pre-distortion sequence based on the extracted identity authentication pilot sequence, and use the pre-distortion sequence to perform frequency domain equalization on the data sub-carrier of the target sub-block; a demodulation and decoding module, configured to demodulate and decode the message signal after frequency domain equalization to recover the original time synchronization message and a message security indication vector; and an identity authentication decision module, configured to make an identity authentication decision based on the error rate of the message security indication vector, and to send the original time synchronization message to the receiving end or discard it according to the decision result.

Images