Lightweight Authentication Protocol Based on LFSR-APUF and Private Cover Function

Abstract

The present invention discloses a lightweight identity authentication protocol of LFSR‑APUF and private Cover function, and proposes a lightweight identity authentication protocol based on LFSR‑APUF structure to realize dynamic obfuscation scheme. The protocol model includes multiple devices, a server and a database. The protocol has two parts: registration phase and authentication phase. The switching strategy of LFSR unit changes with the change of external stimulus. Through deep integration with authentication protocol, dynamic time-varying fuzziness is realized, and bidirectional authentication can be realized from device to server and from server to device.

Description

Technical Field

[0001] The present invention relates to the field of information and communication technology, and in particular to a lightweight identity authentication protocol of LFSR-APUF and private Cover function. Background Art

[0002] Based on the existing and evolving information and communication technologies, the Internet of Things (IoT) connects and integrates the physical world and the digital space. It is generally considered to be a global network of interconnected objects that are uniquely addressable based on standard communication protocols. The IoT is widely used in industrial control, environmental monitoring, warehousing and logistics, agricultural production, national defense and military, medical and health care, and other fields, and has gradually become one of the focuses of global scientific and technological development. However, with the continuous expansion of the scale of the IoT, the number and types of access devices have gradually increased, and the information security of IoT devices has become a major concern for users. This problem can be solved by authentication protocols based on various encryption algorithms (such as AES and RSA). The key is crucial to the encryption algorithm and should be kept strictly confidential. It is usually stored in non-volatile memory (NVM). However, some IoT devices are low-cost, small in size, ultra-low in power consumption, and resource-constrained, making it difficult to implement complex encryption algorithms. In addition, the NVM that stores the key is vulnerable to various physical attacks.

[0003] Physically Unclonable Function (PUF) is a very promising hardware security primitive for resource-constrained systems. PUF uses the inevitable random physical variations in the chip manufacturing process to extract a unique bit stream and generate unique stimulus-response pairs (CRPs). Therefore, PUF is an alternative solution for low-cost key generation and device authentication.

[0004] Although strong PUF-based protocols provide an extremely effective and lightweight solution for authentication of resource-constrained IoT devices, they are not completely reliable and are vulnerable to modeling attacks, which can clone the CRPs behavior of a PUF instance, essentially undermining the security of authentication protocols based on the physical unclonability and uniqueness of strong PUFs.

[0005] Each delay chain of a traditional APUF is the linear cumulative sum of its delay elements. CRPs fuzzification can improve the PUF's ability to resist machine learning attacks, and this technology does not require potential structural changes to the PUF. It only introduces a specific mapping circuit at the stimulus input or response output port of the PUF to confuse the true mapping relationship between the stimulus and the response and prevent attackers from obtaining valid CRPs. However, the traditional time-invariant fuzzification scheme converts external stimuli into fuzzy stimuli with static mapping functions. It essentially uses a fixed mapping function and cannot effectively resist advanced machine learning attacks. In addition to being vulnerable to modeling attacks, most existing protocols based on strong PUFs assume that the server is secure during the authentication phase. Since the server stores CRPs about strong PUFs or strong PUF software models, once the server in the authentication network is compromised, the attacker can obtain information related to strong PUFs and other encryption functions, destroying the entire authentication protocol;

[0006] Disadvantages of existing technology:

[0007] (1) The structure of the prior art 1 has low resistance to modeling attacks.

[0008] (2) The block containing NVM in the structure of the prior art 2 increases the device resource overhead and is vulnerable to physical attacks.

[0009] (3) Existing technologies 1 and 2 only support one-way authentication from server to device, which is vulnerable to spoofing attacks by fake servers, thereby destroying the entire authentication network. Summary of the invention

[0010] In view of the insufficient protection provided by the existing PUF against machine learning attacks using lightweight obfuscation logic, the lightweight authentication protocol based on the novel LFSR-APUF and the private Cover function provided by the present invention makes full use of the special structure of the APUF (LFSR-APUF) of the proposed linear feedback shift register (LFSR) that is resistant to modeling attacks, realizes a dynamic obfuscation scheme, further improves the anti-attack capability of protocol-level modeling, and helps to uniquely encrypt each device in the authentication network.

[0011] To achieve the above object, the present invention provides the following technical solutions:

[0012] A lightweight identity authentication protocol of LFSR-APUF and private Cover function, including a registration part and an authentication part;

[0013] The registration part includes the following steps:

[0014] Step 1: After the two weak PUFs (WPUF1 and WPUF2) in the server are powered on, two server keys K are generated. s1 and Ks2 ;

[0015] Step 2: The server assigns a unique ID to each device k in the authentication network. k ), the only primitive polynomial (p k ) and the only ultra-lightweight Cover function (Cover k );

[0016] Step 3: The server uses the encryption key K s1 p for device k k and Cover k The encrypted content is encrypted with Indicates, abbreviated as Enc k , where the symbol || is used to represent the connection of two vectors; then ID k The encrypted information is stored in the database as an index;

[0017] Step 4: The true random number generator (TRNG) in the server generates y sets of random numbers (n c ,Base); for the i-th round of device k authentication, (n c ,Base) is defined as (n ci ,Base i );n ci There are n bits, which is equivalent to the number of APUF circuit stages, and y determines the size of the CRP data set used for authentication with the corresponding device;

[0018] Step 5: For the i-th round of authentication of device k, receive (n ci ,Base i ) input to the LFSR k In the LFSR k The seed value and shift count value of the i-th round of authentication generate the corresponding fuzzy stimulus, using n Lci express;

[0019] Step 6: LFSR in device k k Replace n in step 5 Lci As a seed value for incentive expansion, the incentive is used after expansion <c i > n express;

[0020] Step 7: APUF in device k k Use the extended <c i > n Generate multi-bit response r i , the response is sent to the server;

[0021] Step 8: The server uses the encryption key Ks2 (defined as For the LFSR in device k k Related data and APUF k The response data is encrypted and stored in the reserved space under the corresponding index of the database.

[0022] The authentication part includes the following steps:

[0023] Step 1: Device k sends an identity token to the server for initial authentication communication; after receiving the identity token from device k, the server only needs to check its validity and send it to the database to request the encrypted message related to device k;

[0024] Step 2: The server receives the relevant encrypted information about device k from the database and uses the key K generated by WPUF1 s1 Decrypt it and configure device k;

[0025] Step 3: When the server completes the configuration in step 2, device k uses TRNG to generate (n d ,ind i1 ), where n d Cover k The random l bits of ind i1 Used to request the i-th round of encrypted messages in the database;

[0026] Step 4: Receive (n d ,ind i1 ), the server uses (ID k ,ind i1 ) Request the relevant encryption information Enc of device k from the database ki ;Then, the database sends the i-th round encrypted message about device k to the server;

[0027] Step 5: After the server receives the encrypted message from the database, it uses the key K generated by WPUF2 s2 Decrypt it and generate (n ci ||Base i ||r i ); Use Covers(n) on the server ci ||Base i ||r i ,n d ), generate n dc ; Finally, n dc Send to device k;

[0028] Step 6: Device k uses the Uncovers function to decrypt to obtain relevant information (n ci ||Basei ||r i ); Then, device k uses the received (n ci ||Base i )Configure LFSR k For fuzzy excitation n Lci , then use n Lci Used as a seed to expand the stimulus and send it to the APUF k Generates a multi-bit response r' i ; r' i With r i Compare them. If they are the same, the device successfully authenticates the server.

[0029] Step 7: After authenticating the server, the server uses its own TRNG to generate (n s , ind i2 ), similar to step 3; then, the server will n s Send to device k and send ind i2 Request relevant encrypted information from the database;

[0030] Step 8: The server receives Enc from the database kj Then, use the key K s2 Decrypt it and generate (nc j ||Base j ||r j ); then, the server (nc j ||Base j ) and n s Use the Covers function to generate n sc , and send it to device k;

[0031] Step 9: Device k receives n from the server in sequence s and n sc After that, use n s And Uncovers function for n sc Decrypt to obtain (n cj ||Base j ); then (n cj ||Base j ) Input LFSR-APUF generates multi-bit response r' j , and send it to the server;

[0032] Step 10: After receiving r' sent by device k j After that, the server compares it with the r extracted from the database j Compare; if they are the same, the server will successfully authenticate device k; the server and the device complete mutual identity authentication.

[0033] As a further solution of the invention: the LFSR-APUF comprises four parts: an excitation generator, a conventional APUF unit, a LFSR unit and a control unit;

[0034] The specific workflow is summarized as follows: the stimulus generator generates stimulus C, where C is divided into two parts, C1 and C2, which serve as the input of the LFSR unit and the control unit respectively; the LFSR unit receives C1 generated by the stimulus generator and uses it as the initial value; then, it waits for the control unit to send a shift signal S L The control unit receives C2 generated by the excitation generator and determines the shift signal S according to the value of C2 L The number of times it should be sent; the LFSR unit is based on S L After shifting the corresponding number of times, the output can be used as the C of the fuzzy excitation O ; Finally, when the control unit sends a shift signal S L After the specified number of times, it sends a pulse signal S to the APUF unit A ; APUF receives the fuzzy stimulus C generated by LFSR O and pulse signal S A ;When the pulse signal is transmitted from the beginning to the last D flip-flop through the two delay chains of APUF, APUF generates a response R one bit at a time.

[0035] As a further solution of the invention: in the LFSR-APUF, the external excitation C is (n+m) bits, C1 is n bits, and C2 is m bits.

[0036] As a further solution of the invention: in the LFSR-APUF, both the LFSR and the APUF are of n levels.

[0037] As a further solution of the invention: the LFSR-APUF, wherein an integer Base corresponding to a shift count is predefined, further improves the modeling attack resistance of the LFSR obfuscation scheme, and the final shift count can be expressed as:

[0038] Shift count = ||C2||*Base;

[0039] Where ||C2|| represents the value of m-bit C2, Base is a predefined integer value, and the shift count is the shift signal S L Controls the final number of times the LFSR is shifted.

[0040] Compared with the prior art, the beneficial effects of the invention are: the advantage of the present invention is that a circuit structure based on a new LFSR-APUF is proposed to blur external stimuli, and the switching strategy of the LFSR unit changes with the change of the external stimuli. Through deep integration with the authentication protocol, dynamic time-varying fuzzification is achieved, and at the same time, two-way authentication can be achieved from device to server and from server to device. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a circuit structure diagram of the LFSR-APUF of the present invention;

[0042] Figure 2 This is a specific schematic diagram of the registration phase of the lightweight identity authentication protocol provided in the present invention;

[0043] Figure 3 The figure is a specific schematic diagram of the authentication phase of the lightweight identity authentication protocol provided in the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the invention to clearly and completely describe the technical solutions in the embodiments of the invention. Obviously, the described embodiments are only part of the embodiments of the invention, not all of them. Based on the embodiments in the invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the invention.

[0045] The anti-modeling attack design based on LFSR-APUF proposed in the present invention is mainly divided into four parts: an excitation generator, a traditional APUF unit, a LFSR unit, and a control unit. Among them, the excitation generator mainly generates random excitations from the outside when implemented on an FPGA development board, and the external excitations generated are respectively input into the LFSR unit and the control unit. The control unit mainly controls the LFSR shift to generate fuzzy excitations according to the external partial excitations, and controls the APUF to generate response bits. The LFSR unit mainly confuses the external excitations by shifting, and inputs the confused excitations into the APUF unit. The APUF unit is a traditional APUF circuit with an n-level symmetrical switch block.

[0046] The structure proposed in the present invention is different from the traditional time-invariant fuzzy scheme that converts external excitation into fuzzy excitation with a static mapping function. Before the external excitation is obfuscated, several bits are extracted from the external excitation and applied to the mapping function. Then the mapping function of the fuzzy module is determined using the real-time specific external excitation. This is equivalent to converting a fixed mapping function into a mapping function that changes with the external excitation, converting the obfuscation mechanism from "time-invariant" to "time-varying", greatly reducing the ability of attackers to destroy it through modeling attacks;

[0047] The method of lightweight identity authentication protocol of novel LFSR-APUF and private Cover function proposed in the present invention realizes dynamic obfuscation scheme. The protocol model includes multiple devices, a server and a database. The protocol has two parts: registration phase and authentication phase:

[0048] The registration part has the following steps:

[0049] Step 1: After the two weak PUFs (WPUF1 and WPUF2) in the server are powered on, two server keys K are generated. s1 and K s2 ;

[0050] Step 2: The server assigns a unique ID to each device k in the authentication network. k ), the only primitive polynomial (p k ) and the only ultra-lightweight Cover function (Cover k );

[0051] Step 3: The server uses the encryption key K s1 p for device k k and Cover k The encrypted content is encrypted with Indicates, abbreviated as Enc k , where the symbol || is used to represent the connection of two vectors; then ID k The encrypted information is stored in the database as an index;

[0052] Step 4: The true random number generator (TRNG) in the server generates y sets of random numbers (n c ,Base); for the i-th round of device k authentication, (n c ,Base) is defined as (n ci ,Base i );n ci There are n bits, which is equivalent to the number of APUF circuit stages, and y determines the size of the CRP data set used for authentication with the corresponding device;

[0053] Step 5: For the i-th round of authentication of device k, receive (n ci ,Base i ) input to the LFSR k In the example, determine the LFSR k The seed value and shift count value of the i-th round of authentication generate the corresponding fuzzy stimulus, using n Lci express;

[0054] Step 6: LFSR in device k k Replace n in step 5 LciAs a seed value for incentive expansion, the incentive is used after expansion <c i > n express;

[0055] Step 7: APUF in device k k Use the extended <c i > n Generate multi-bit response r i , the response is sent to the server;

[0056] Step 8: The server uses the encryption key K s2 (defined as For the LFSR in device k k Related data and APUF k The response data is encrypted and stored in the reserved space under the corresponding index of the database.

[0057] The authentication part includes the following steps:

[0058] Step 1: Device k sends an identity token to the server for initial authentication communication; after receiving the identity token from device k, the server only needs to check its validity and send it to the database to request the encrypted message related to device k;

[0059] Step 2: The server receives the relevant encrypted information about device k from the database and uses the key K generated by WPUF1 s1 Decrypt it and configure device k;

[0060] Step 3: When the server completes the configuration in step 2, device k uses TRNG to generate (n d ,ind i1 ), where n d Cover k The random l bits, ind i1 Used to request the i-th round of encrypted messages in the database;

[0061] Step 4: Receive (n d ,ind i1 ), the server uses (ID k ,ind i1 ) Request the relevant encryption information Enc of device k from the database ki ; Then, the database sends the i-th round encrypted message about device k to the server;

[0062] Step 5: After the server receives the encrypted message from the database, it uses the key K generated by WPUF2 s2 Decrypt it and generate (n ci ||Base i ||ri ); Use Covers(n) on the server ci ||Base i ||r i ,n d ), generate n dc ; Finally, n dc Send to device k;

[0063] Step 6: Device k uses the Uncovers function to decrypt to obtain relevant information (n ci ||Base i ||r i ); Then, device k uses the received (n ci ||Base i )Configure LFSR k For fuzzy excitation n Lci , then use n Lci Used as a seed to expand the stimulus and send it to the APUF k Generates a multi-bit response r' i ; r' i With r i Compare them. If they are the same, the device successfully authenticates the server.

[0064] Step 7: After authenticating the server, the server uses its own TRNG to generate (n s , ind i2 ), similar to step 3; then, the server will n s Send to device k and send ind i2 Request relevant encrypted information from the database;

[0065] Step 8: The server receives Enc from the database kj Then, use the key K s2 Decrypt it and generate (nc j ||Base j ||r j ); then, the server (nc j ||Base j ) and n s Use the Covers function to generate n sc , and send it to device k;

[0066] Step 9: Device k receives n from the server in sequence s and n sc After that, use n s And Uncovers function for n sc Decrypt to obtain (n cj ||Base j ); then (ncj ||Base j ) Input LFSR-APUF generates multi-bit response r' j , and send it to the server;

[0067] Step 10: After receiving r' sent by device k j After that, the server compares it with the r extracted from the database j Compare; if they are the same, the server will successfully authenticate device k; the server and the device complete mutual identity authentication.

[0068] In order to make the purpose, content and advantages of the present invention more clear, the specific working process of the present invention is further described in detail below with reference to the accompanying drawings.

[0069] Figure 1 The figure shows the LFSR-APUF circuit structure designed by the present invention to resist modeling attacks. It includes four parts: an excitation generator, a traditional APUF unit, a LFSR unit, and a control unit. The specific workflow is summarized as follows: the excitation generator generates an excitation C, where C is divided into two parts, C1 and C2, which serve as inputs to the LFSR unit and the control unit, respectively. The LFSR unit receives C1 generated by the excitation generator and uses it as the initial value. Then, it waits for the control unit to send a shift signal S L The control unit receives C2 generated by the excitation generator and determines the shift signal S according to the value of C2. L The LFSR unit is based on S L After shifting the corresponding number of times, the output can be used as the C of the fuzzy excitation O Finally, when the control unit sends the shift signal S L After the specified number of times, it sends a pulse signal S to the APUF unit A The APUF receives the fuzzy stimulus C generated by the LFSR. O and control unit pulse signal S A When the pulse signal is transmitted from the beginning to the final D flip-flop through the two delay chains of the APUF, the APUF generates a response R one bit at a time.

[0070] According to the LFSR-APUF of the present invention, the external stimulus C is (n+m) bits, C1 is n bits, and C2 is m bits.

[0071] According to the LFSR-APUF of the present invention, both the LFSR and the APUF are of n stages.

[0072] According to the LFSR-APUF of the present invention, an integer Base corresponding to a shift count is predefined, which further improves the modeling attack resistance of the LFSR obfuscation scheme. The final shift count can be expressed as:

[0073] Shift count = ||C2||*Base

[0074] Where ||C2|| represents the value of m-bit C2, Base is a predefined integer value, and the shift count is the shift signal S L Controls the final number of times the LFSR is shifted.

[0075] Figure 2 The figure shows the registration phase of a lightweight authentication protocol proposed based on the proposed LFSR-APUF structure. The steps of the registration phase are as follows:

[0076] Step 1: After the two weak PUFs (WPUF1 and WPUF2) in the server are powered on, two server keys K are generated. s1 and K s2 ;

[0077] Step 2: The server assigns a unique ID to each device k in the authentication network. k ), the only primitive polynomial (p k ) and the only ultra-lightweight Cover function (Cover k );

[0078] Step 3: The server uses the encryption key K s1 p for device k k and Cover k The encrypted content is encrypted with Indicates, abbreviated as Enc k , where the symbol || is used to represent the connection of two vectors; then ID k The encrypted information is stored in the database as an index;

[0079] Step 4: The true random number generator (TRNG) in the server generates y sets of random numbers (n c ,Base); for the i-th round of device k authentication, (n c ,Base) is defined as (n ci ,Base i );n ci There are n bits, which is equivalent to the number of APUF circuit stages, and y determines the size of the CRP data set used for authentication with the corresponding device;

[0080] Step 5: For the i-th round of authentication of device k, receive (n ci ,Base i ) input to the LFSR k In the LFSR kThe seed value and shift count value of the i-th round of authentication generate the corresponding fuzzy stimulus, using n Lci express;

[0081] Step 6: LFSR in device k k Replace n in step 5 Lci As a seed value for incentive expansion, the incentive is used after expansion <c i > n express;

[0082] Step 7: APUF in device k k Use the extended <c i > n Generate multi-bit response r i , the response is sent to the server;

[0083] Step 8: The server uses the encryption key K s2 (defined as For the LFSR in device k k Related data and APUF k The response data is encrypted and stored in the reserved space under the corresponding index of the database.

[0084] The authentication part includes the following steps:

[0085] Step 1: Device k sends an identity token to the server for initial authentication communication; after receiving the identity token from device k, the server only needs to check its validity and send it to the database to request the encrypted message related to device k;

[0086] Step 2: The server receives the relevant encrypted information about device k from the database and uses the key K generated by WPUF1 s1 Decrypt it and configure device k;

[0087] Step 3: When the server completes the configuration in step 2, device k uses TRNG to generate (n d ,ind i1 ), where n d Cover k The random l bits, ind i1 Used to request the i-th round of encrypted messages in the database;

[0088] Step 4: Receive (n d ,ind i1 ), the server uses (ID k ,ind i1 ) Request the relevant encryption information Enc of device k from the database ki ; Then, the database sends the i-th round encrypted message about device k to the server;

[0089] Step 5: After the server receives the encrypted message from the database, it uses the key K generated by WPUF2 s2 Decrypt it and generate (n ci ||Base i ||r i ); Use Covers(n) on the server ci ||Base i ||r i ,n d ), generate n dc ; Finally, n dc Send to device k;

[0090] Step 6: Device k uses the Uncovers function to decrypt to obtain relevant information (n ci ||Base i ||r i ); Then, device k uses the received (n ci ||Base i )Configure LFSR k For fuzzy excitation n Lci , then use n Lci Used as a seed to expand the stimulus and send it to the APUF k Generates a multi-bit response r' i ; r' i With r i Compare them. If they are the same, the device successfully authenticates the server.

[0091] Step 7: After authenticating the server, the server uses its own TRNG to generate (n s , ind i2 ), similar to step 3; then, the server will n s Send to device k and send ind i2 Request relevant encrypted information from the database;

[0092] Step 8: The server receives Enc from the database kj Then, use the key K s2 Decrypt it and generate (nc j ||Base j ||r j ); then, the server (nc j ||Base j ) and n s Use the Covers function to generate n sc , and send it to device k;

[0093] Step 9: Device k receives n from the server in sequences and n sc After that, use n s And Uncovers function for n sc Decrypt to obtain (n cj ||Base j ); then (n cj ||Base j ) Input LFSR-APUF generates multi-bit response r' j , and send it to the server;

[0094] Step 10: After receiving r' sent by device k j After that, the server compares it with the r extracted from the database j Compare; if they are the same, the server will successfully authenticate device k; the server and the device complete mutual identity authentication.

[0095] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention.

Claims

1. A lightweight identity authentication protocol based on LFSR-APUF and private Cover function, characterized in that: It includes registration part and certification part; The registration part includes the following steps: Step 1: The two weak PUFs in the server include WPUF1 and WPUF2. After power-on, two server keys K are generated. s1 and K s2 ; Step 2: The server assigns a unique ID to each device k in the authentication network k , the only primitive polynomial p k And the only ultra-lightweight Cover function Cover k ; Step 3: The server uses the encryption key K s1 p for device k k and Cover k The encrypted content is encrypted with (p k || Cover k ) means, abbreviated as , where the symbol || is used to represent the connection of two vectors; then ID k The encrypted information is stored in the database as an index; Step 4: The true random number generator TRNG in the server generates y sets of random numbers (n c , Base); for the i-th round of authentication of device k, (n c , Base) is defined as (n ci , Base i );n ci There are n bits, which is equivalent to the number of APUF circuit stages, and y determines the size of the CRP data set used for authentication with the corresponding device; Step 5: For the i-th round of authentication of device k, receive (n ci , Base i ) input to the LFSR k In the example, determine the LFSR k The seed value and shift count value of the i-th round of authentication generate the corresponding fuzzy stimulus, using n Lci express; Step 6: LFSR in device k k Replace n in step 5 Lci As a seed value for incentive expansion, the incentive is used after expansion <c i > n express; Step 7: APUF in device k k Using the expanded < c i > n Generate multi-bit response r i , the response is sent to the server; Step 8: The server uses the encryption key K s2 For the LFSR in device k k Related data and APUF k The response data is encrypted and stored in the reserved space under the corresponding index of the database; The authentication part includes the following steps: Step 1: Device k sends an identity token to the server for initial authentication communication; after receiving the identity token from device k, the server only needs to check its validity and send it to the database to request the encrypted message related to device k; Step 2: The server receives the relevant encrypted information about device k from the database and uses the key K generated by WPUF1 s1 Decrypt it and configure device k; Step 3: When the server completes the configuration in step 2, device k uses TRNG to generate (n d , ind i1 ), where n d Cover k The random l bits, ind i1 Used to request the i-th round of encrypted messages in the database; Step 4: Receive from device k (n d , ind i1 ) after which the server uses (ID k , ind i1 ) Request the relevant encryption information Enc of device k from the database ki ;Then, the database sends the i-th round encrypted message about device k to the server; Step 5: After the server receives the encrypted message from the database, it uses the key K generated by WPUF2 s2 Decrypt it and generate (n ci || Base i || r i ); Use on the server Covers (n ci || Base i || r i, n d ), generate n dc ; Finally, n dc Send to Device k ; Step 6: Device k Usage Uncovers Function decryption to obtain relevant information, namely (n ci || Base i || r i Then, device k uses the received (n ci || Base i ) Configure LFSR k For fuzzy excitation n Lci , then use n Lci Used as a seed to expand the stimulus and send it to the APUF k Generates a multi-bit response r' i ; r' i With r i Compare them. If they are the same, the device successfully authenticates the server. Step 7: After authenticating the server, the server uses its own TRNG to generate (n s , ind i2 ), similar to step 3; then, the server will n s Send to device k and send ind i2 Request the relevant encrypted information from the database; Step 8: The server receives Enc from the database kj Then, use the key K s2 Decrypt it and generate (nc j || Base j || r j ); Then, the server (nc j || Base j ) and n s use Covers Function to generate n sc , and send it to device k; Step 9: Device k receives n from the server in sequence s and n sc After that, use n s and Uncovers Function for n sc Decrypt to obtain (n cj || Base j ); then (n cj || Base j ) Input LFSR-APUF generates a multi-bit response r' j , and send it to the server; Step 10: After receiving r' sent by device k j After that, the server compares it with the r extracted from the database j Compare; if they are the same, the server will successfully authenticate device k; the server and the device complete mutual identity authentication.

2. The LFSR-APUF and the private Cover A lightweight authentication protocol for functions, characterized in that: The LFSR-APUF consists of four parts: the stimulus generator, the traditional APUF unit, the LFSR unit, and the control unit; The specific workflow is summarized as follows: the stimulus generator generates stimulus C, where C is divided into two parts, C1 and C2, which serve as the input of the LFSR unit and the control unit respectively; the LFSR unit receives C1 generated by the stimulus generator and uses it as the initial value; then, it waits for the control unit to send a shift signal S L The control unit receives C2 generated by the excitation generator and determines the shift signal S according to the value of C2 L The number of times it should be sent; the LFSR unit is based on S L After shifting the corresponding number of times, the output can be used as the C of the fuzzy excitation O ; Finally, when the control unit sends a shift signal S L After the specified number of times, it sends a pulse signal S to the APUF unit A ; APUF receives the fuzzy stimulus C generated by LFSR O and pulse signal S A ;When the pulse signal is transmitted from the beginning to the last D flip-flop through the two delay chains of APUF, APUF generates a response R one bit at a time.

3. The lightweight identity authentication protocol of LFSR-APUF and private Cover function according to claim 2, characterized in that: In the LFSR-APUF, the external stimulus C is (n+m) bits, C1 is n bits, and C2 is m bits.

4. The lightweight identity authentication protocol of LFSR-APUF and private Cover function according to claim 2, characterized in that: In the LFSR-APUF, both the LFSR and the APUF are of n levels.

5. The lightweight identity authentication protocol of LFSR-APUF and private Cover function according to claim 2, characterized in that: The LFSR-APUF, wherein an integer Base corresponding to a shift count is predefined, further improves the modeling attack resistance of the LFSR obfuscation scheme, and the final shift count can be expressed as: ShiftCount = ||C2||*Base; Where ||C2|| represents the value of m-bit C2, Base is a predefined integer value, and the shift count is the shift signal S L Controls the final number of times the LFSR is shifted.

Images