Secure communication system and method for generative artificial intelligence-enabled Internet of Things

Abstract

The invention belongs to the technical field of artificial intelligence, and discloses a safety communication system and method of a generative artificial intelligence enabled internet of things, and the system comprises a registration authority which is responsible for registering an internet of things user and internet of things equipment; the Internet of Things user registers with the registration authority by using the smart phone, interacts with the edge artificial intelligence after logging in, and controls the remote intelligent device through the edge artificial intelligence; the intelligent device is used for collecting data in the environment and transmitting the data to the edge artificial intelligence and the Internet of Things user after mutual authentication; the edge artificial intelligence is an edge server close to the intelligent equipment, and an artificial intelligence application is deployed in the edge server and is fusion of edge calculation and artificial intelligence. According to the invention, the secure communication between the Internet of Things user and the remote intelligent device is mainly realized, and a weak physical unclonable function is embedded in the remote intelligent device in order to protect the physical security of the remote intelligent device.

Description

Technical Field

[0001] This invention belongs to, but is not limited to, the field of artificial intelligence technology, and particularly relates to a secure communication system and method for enabling the Internet of Things with generative artificial intelligence. Background Technology

[0002] Generative AI, a branch of artificial intelligence, not only processes data but also learns the inherent laws of things, autonomously creating new content. Generative AI-enabled IoT refers to integrating generative AI technology into the Internet of Things (IoT), enabling a comprehensive upgrade of IoT intelligence. Generative AI-enabled IoT generates and transmits large amounts of data, posing significant challenges to communication security in IoT scenarios. Authentication protocols are crucial for protecting secure communication between entities in IoT scenarios. In the past, numerous authentication protocols have been proposed for various IoT application scenarios. However, many protocols claim superior performance and resistance to known attacks. Unfortunately, some authentication protocols have been found to have security vulnerabilities shortly after their introduction. Even worse, some improved authentication protocols have been found to be insecure. Designing a secure authentication protocol has attracted widespread attention. Generative AI-enabled IoT scenarios present even more serious security threats; attackers can not only completely control communication channels but also capture smart devices and implement AI-based attack methods based on the obtained data. Therefore, in order to ensure secure communication in generative AI-enabled IoT scenarios, it is necessary to design efficient and secure authentication protocols to prevent various attacks on generative AI-enabled IoT.

[0003] Currently, numerous authentication protocols have been proposed for IoT scenarios such as Industrial Internet of Things (IIoT), Vehicle Internet of Things (VIoT), Smart Grid, Smart Medical Systems, Wireless Body Area Networks (WBN), Unmanned Aerial Vehicle (UAV) Networks, Agricultural IoT, and Smart Home. However, many protocols are found to have security vulnerabilities shortly after their proposal, and "design-patch-redesign" authentication protocols may still be insecure. The shortcomings of existing technologies are, firstly, that they do not fully consider the attack capabilities of attackers, especially the attack techniques based on artificial intelligence, which makes existing authentication technologies secure under weak attacker capabilities but insecure under strong attacker capabilities. Secondly, existing technologies use computationally intensive cryptographic primitives to enhance security, making these technologies unsuitable for resource-constrained IoT scenarios. This invention solves these currently unsolvable problems: (1) The authentication protocol can provide security even when facing the strongest attackers. (2) While providing strong security, the communication scheme is also lightweight, making it suitable for application in IoT scenarios enabled by generative artificial intelligence. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides a secure communication system and method for enabling the Internet of Things with generative artificial intelligence.

[0005] This invention is implemented as follows: a secure communication system for the Internet of Things enabled by generative artificial intelligence, the system comprising:

[0006] The registration authority is responsible for registering IoT users and IoT devices. It is a trusted server that is responsible for registering entities in the IoT.

[0007] IoT users connect with a registration authority. IoT users register with the registration authority using their smartphones. After logging in, they interact with edge AI and control remote smart devices through edge AI.

[0008] Smart devices are used to collect data from the environment and transmit the data to edge AI and IoT users after mutual authentication.

[0009] Edge AI connects with IoT users and smart devices. Edge AI consists of edge servers located close to smart devices, where AI applications are deployed. It is a fusion of edge computing and AI. Edge AI can process data transmitted from smart devices and analyze and make decisions based on the collected data. When necessary, it is also responsible for transmitting certain data to cloud AI. Edge AI also assists IoT users and smart devices in performing mutual authentication.

[0010] Another objective of this invention is to provide a secure communication method for a generative artificial intelligence-enabled Internet of Things (IoT) based on the aforementioned secure communication system, the method comprising: a registration phase and an authentication phase;

[0011] The registration phase involves registration by a registration authority for IoT users and smart devices; the authentication phase involves mutual authentication between IoT users and smart devices to achieve secure communication, specifically including:

[0012] (1) User registration:

[0013] S1: The user sends a registration request message Regi to the registration authority;

[0014] S2: After receiving the user's registration request, the registration authority generates a response message Resi and sends (Regi, Resi) to the user;

[0015] S3: The user selects username IDi and password PWi, and selects two random numbers ai and bi. Calculate RIDi = h(IDi||ai), RPWi = h(IDi||PWi||ai)⊕bi, Proci = RIDi⊕Resi, and then send {Proci,RPWi} to the registration authority.

[0016] S4: After receiving the user's registration request, the registration authority generates a random number ri for the user, calculates RIDi=Proci⊕Resi,Ai=h(ri||x)⊕RPWi,Bi=h(x||y)⊕h(RPWi), transmits Ai and Bi to the user, and stores the user's registration information {RIDi,ri} in the edge AI.

[0017] S5: The user receives the message transmitted by the registration authority. The mobile device calculates h(ri||x)=Ai⊕RPWi,h(x||y)=Bi⊕h(RPWi),Ci=h(IDi||PWi)⊕ai,Di=h(ri||x)⊕h(RIDi||PWi),Ei=h(x||y)⊕h(RIDi||IDi||PWi),Fi=h(h(ri||x)||h(x||y)||RIDi||PWi||ai). Finally, {Ci,Di,Ei,Fi} are stored in the mobile phone.

[0018] (2) Smart device registration: A weakly physically unclonable function (WPUF) is embedded in each smart device; the registration authority selects an identity IDj for the smart device, generates a challenge Cj and the current registration timestamp Tj, and calculates Sj = h(IDj||x||Tj) and Rj = WPUF(Cj); {IDj, h(Rj) ⊕ h(x||Tj), Tj} is stored in the edge AI, and {IDj, Cj, Sj} is stored in the smart device;

[0019] The authentication phase includes login and mutual authentication; the user first logs in using a smartphone. After successful login, mutual authentication is performed with a remote smart device through edge artificial intelligence, and a secure session key for subsequent communication of the user is generated.

[0020] (1) Login: The user enters the username IDi and password PWi. The smartphone calculates ai=h(IDi||PWi)⊕Ci,RIDi=h(IDi||ai),h(ri||x)=Di⊕h(RIDi||PWi),h(x||y)=Ei⊕h(RIDi||IDi||PWi), and then calculates Fi=h(h(ri||x)||h(x||y)||RIDi||PWi||ai). If the Fi is equal to the stored Fi, the user logs in successfully.

[0021] (2) Mutual authentication: After successful login, the user and the remote smart device perform mutual authentication. The authentication steps are as follows:

[0022] The mobile device selects the smart device IDj to access, generates a random number n1, the current timestamp t1, and calculates M1=n1⊕h(h(ri||x)||t1),TIDi=RIDi⊕h(ri||x),Auth1=h(h(ri||x)||RIDi||TIDi||n1||t1),M2=IDj⊕h(Auth1||t1),M3=h(RIDi||PWi||n1||t1)⊕h(h(ri||x)||IDj||n1), and finally passes Meg1={M1,TIDi,M2,M3,Auth1,t1} to the edge AI.

[0023] After receiving a message, the edge AI checks its freshness, that is, whether the difference between the timestamp of the received message and the timestamp of the sent message is less than the preset maximum transmission delay. If the message is fresh, the edge AI calculates RIDi = TIDi ⊕ h(x||y), checks whether RIDi exists in its own database, if it exists, finds the ri corresponding to RIDi, and calculates n1 = M1 ⊕ h(h(ri||x)||t1), Auth1* = h(h(ri||x)||RIDi||TIDi||n1||t1), and checks whether Auth1* is the same as the received Auth1. If they are the same, the edge AI authenticates the user as real and calculates IDj = M2 ⊕ h(Auth1||t1), h(RIDi||PWi||n1||t1) = M3 ⊕ h(h(ri||x)||IDj||n1).

[0024] Edge AI checks if IDj exists in the database. If it does, it finds the corresponding h(Rj)⊕h(x||Tj),Tj, and recovers h(Rj).

[0025] Edge AI generates a random number n2, the current timestamp t2, and calculates Sj = h(IDj||x||Tj), M4 = n2⊕h(Sj||h(Rj)||t2), M5 = h(RIDi||PWi||n1||t1)⊕h(IDj||h(Rj)||n2||t2), M6 = h(x||n2)⊕h(Sj||h(Rj)||n2), Auth2 = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||h(Rj)||n2||t2); finally, the message Meg2 = {M4, M5, M6, Auth2, t2} is transmitted to the smart device.

[0026] After receiving a message, the smart device checks its freshness. If the message is fresh, the smart device calculates Rj = WPUF(Cj) based on Cj in memory, then calculates n2 = M4⊕h(h(IDj||x||Tj)||h(Rj)||t2),h(RIDi||PWi||n1||t1) = M5⊕h(IDj||h(Rj)||n2||t2),h(x||n2) = M6⊕h(Sj||h(Rj)||n2),Auth2* = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||h(Rj)||n2||t2), and checks whether Auth2* is equal to Auth2. If they are equal, it means that the smart device has authenticated edge AI.

[0027] The smart device generates a random number n3 and the current timestamp t3, and calculates M7 = n3 ⊕ h(h(RIDi||PWi||n1||t1)||IDj||t3), M8 = h(x||n2) ⊕ h(h(RIDi||PWi||n1||t1)||n3), SK = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||n3||t3), Auth3 = h(h(RIDi||PWi||n1||t1)||SK||IDj||t3); then, the smart device sends the message Meg3 = {M7, M8, Auth3, t3} to the user.

[0028] After receiving a message from a smart device, the user checks the freshness of the message. If the received message is fresh, the user calculates n3 = M7⊕h(h(RIDi||PWi||n1||t1)||IDj||t3), h(x||n2) = M8⊕h(h(RIDi||PWi||n1||t1)||n3), SK = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||n3||t3), Auth3* = h(h(RIDi||PWi||n1||t1)||SK||IDj||t3), and checks whether Auth3* is equal to the received Auth3. If they are equal, the user has authenticated the smart device. At this point, the user and the smart device have performed mutual authentication and generated a shared session key SK for subsequent secure communication.

[0029] Another object of the present invention is to provide a computer device including a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the steps of the secure communication method for enabling the Internet of Things with generative artificial intelligence.

[0030] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the secure communication method for enabling the Internet of Things with generative artificial intelligence.

[0031] Another objective of this invention is to provide an information data processing terminal for implementing the secure communication system of the generative artificial intelligence-enabled Internet of Things.

[0032] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:

[0033] First, this invention solves the problem of providing security guarantees even when facing the strongest attackers; while providing strong security guarantees, the communication scheme is also lightweight, making it suitable for application in IoT scenarios enabled by generative artificial intelligence.

[0034] This invention primarily enables secure communication between IoT users and remote smart devices. Its working principle is as follows: The first stage is the registration stage, where both the IoT user and the remote smart device need to register with a registration authority before use. The second stage involves mutual authentication between the IoT user and the remote smart device via a smartphone, where both parties negotiate a secure session key for communication. The user logs in to the smartphone using a username and password, and after successful login, performs mutual authentication with the remote smart device. To protect the physical security of the remote smart device, a weakly physically unclonable function is embedded within the process.

[0035] Second, the expected benefits and commercial value of the technical solution of this invention after transformation are as follows:

[0036] Generative AI-enabled IoT aims to seamlessly connect the physical world and digital intelligence by integrating advanced AI technologies to analyze and transform data from IoT devices in real time into valuable information. It not only expands the traditional data collection and transmission functions of IoT but also introduces advanced capabilities such as deep learning, predictive analytics, and automated decision-making, thus driving the evolution of IoT from simple data collection to intelligent decision-making and autonomous action. Based on this, generative AI-enabled IoT is considered a new stage in IoT development, indicating that it will profoundly change IoT application scenarios and service models, leading the next generation of IoT systems towards a more intelligent and autonomous operating realm. The application scope of generative AI-enabled IoT is wide-ranging, with manufacturing being one of the most important and closely watched areas. In manufacturing, generative AI-enabled IoT technology can be applied to the intelligent monitoring and optimization of production lines. By analyzing production data in real time, it can predict equipment failures, optimize production processes, and improve production efficiency and product quality. Simultaneously, combined with cloud computing and edge computing technologies, generative AI-enabled IoT technology can support the rapid processing and analysis of large-scale data, providing strong technical support for intelligent manufacturing and realizing the digitization and intelligent interconnection of everything. Generative artificial intelligence (AI) enables the generation and transmission of massive amounts of data in the Internet of Things (IoT). The high speed and low latency of this data transmission ensure the real-time performance and accuracy of applications, while secure data communication prevents privacy breaches and financial losses. This invention primarily aims to protect secure communication processes in AI-enabled IoT environments. Specifically, it provides security against even the most powerful attackers who can completely control communication channels, capture smart devices, and implement AI-based attacks based on various data. This security is achieved using only lightweight primitives, a feat many solutions cannot accomplish. Therefore, the technical solution of this invention, once commercialized, will generate significant expected benefits and commercial value.

[0037] Third, this invention addresses the problems existing in IoT secure communication. Traditional IoT communication architectures lack effective authentication and secure communication mechanisms between devices, users, and servers, making them vulnerable to attacks such as data theft, identity forgery, and unauthorized access, resulting in low overall system security. In practical industrial applications, these problems significantly limit the reliability and scalability of the IoT. To address this, this invention designs a multi-layered authentication system based on generative artificial intelligence. By registering an authority, it authenticates users and devices, ensuring that only legitimate users and devices can access the system, significantly improving communication security.

[0038] This invention employs edge artificial intelligence as a core component, performing data analysis and processing close to the device, thereby reducing the frequency and latency of data transmission to the cloud. In traditional IoT systems, large amounts of data need to be transmitted to the cloud for processing, resulting in low communication efficiency, especially in scenarios with high real-time response requirements. This invention, by integrating artificial intelligence technology into edge servers, enables most data processing and decision-making to be completed at the edge, significantly reducing response time and improving data processing efficiency, making it particularly suitable for industrial applications with high real-time response requirements.

[0039] The edge AI module in this invention not only handles data processing but also assists in mutual authentication between users and devices. Traditional IoT systems often lack edge authentication functionality, making device and user authentication overly reliant on central servers, which can easily create single points of failure. This invention, by introducing edge AI, offloads the authentication workload. Even when the central server is inaccessible, the edge server can still effectively authenticate the identities of devices and users, improving the system's reliability and availability, making it more suitable for complex communication environments in industrial settings.

[0040] Furthermore, this invention enhances data security and intelligence through the application of generative artificial intelligence. In the system, edge AI can intelligently analyze data based on its characteristics and behavioral patterns, identify abnormal data or potential threats, and automatically take corresponding protective measures. This intelligent analysis function typically relies on manual intervention in traditional IoT communication systems, while this invention achieves a high degree of automation and intelligence in data processing, resulting in significant technological advancements in industrial applications, improved data security, and a more efficient, secure, and intelligent communication solution for industrial scenarios. Attached Figure Description

[0041] Figure 1 This is a structural diagram of a secure communication system for enabling the Internet of Things using generative artificial intelligence, provided by an embodiment of the present invention.

[0042] Figure 2 This is a flowchart of the user registration method provided in an embodiment of the present invention;

[0043] Figure 3 This is a flowchart of the authentication process between a user and a smart device through edge artificial intelligence, provided in an embodiment of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0045] like Figure 1 As shown, this embodiment of the invention provides a secure communication system for the Internet of Things (IoT) enabled by generative artificial intelligence. The system includes:

[0046] The registration authority is responsible for registering IoT users and IoT devices. It is a trusted server that is responsible for registering entities in the IoT.

[0047] IoT users connect with a registration authority. IoT users register with the registration authority using their smartphones. After logging in, they interact with edge AI and control remote smart devices through edge AI.

[0048] Smart devices are used to collect data from the environment and transmit the data to edge AI and IoT users after mutual authentication.

[0049] Edge AI connects with IoT users and smart devices. Edge AI consists of edge servers located close to smart devices, where AI applications are deployed. It is a fusion of edge computing and AI. Edge AI can process data transmitted from smart devices and analyze and make decisions based on the collected data. When necessary, it is also responsible for transmitting certain data to cloud AI. Edge AI also assists IoT users and smart devices in performing mutual authentication.

[0050] The generative AI-enabled IoT secure communication system provided in this invention ensures secure communication between IoT users and smart devices through multi-layered module collaboration. First, the registration authority module, acting as a trusted server in the system, is responsible for registering entities within the IoT. Before connecting, IoT users and smart devices must first authenticate and register with the registration authority to ensure the legitimacy of each entity. This step effectively prevents unauthorized devices and users from accessing the system, ensuring the trustworthiness of the network environment.

[0051] After registering and logging in with an authoritative entity via a smartphone, the IoT user module interacts with edge AI. Users can then use their smartphones to control remote smart devices, enabling monitoring and management. Once logged in, IoT users can submit control requests to smart devices through edge AI, effectively achieving remote device management and control.

[0052] In the smart device module, devices are used to collect data from the environment in real time. After mutual authentication, the smart devices transmit the collected data to edge AI and IoT users. To ensure the security of data transmission, mutual authentication is used between the smart devices and IoT users to prevent unauthorized devices and users from exchanging data. The close collaboration between smart devices and edge AI enhances the security and real-time performance of data transmission.

[0053] The edge AI module is deployed on edge servers close to smart devices and connects to IoT users and these devices. Edge AI possesses data processing and decision analysis capabilities, enabling it to immediately analyze and process data received from smart devices and control or provide feedback to them. Simultaneously, edge AI assists in mutual authentication between IoT users and smart devices, and transmits data to cloud AI when necessary. This module's configuration combines AI and edge computing, fully leveraging the low latency of edge computing to ensure system data security and communication efficiency.

[0054] This invention provides a secure communication method for a generative artificial intelligence-enabled Internet of Things (IoT) based on the aforementioned secure communication system, the method comprising: a registration phase and an authentication phase;

[0055] The registration phase involves registration by a registration authority for IoT users and smart devices; the authentication phase involves mutual authentication between IoT users and smart devices to achieve secure communication, specifically including:

[0056] (1) As Figure 2 As shown, user registration:

[0057] S1: The user sends a registration request message Regi to the registration authority;

[0058] S2: After receiving the user's registration request, the registration authority generates a response message Resi and sends (Regi, Resi) to the user;

[0059] S3: The user selects username IDi and password PWi, and selects two random numbers ai and bi. Calculate RIDi = h(IDi||ai), RPWi = h(IDi||PWi||ai)⊕bi, Proci = RIDi⊕Resi, and then send {Proci,RPWi} to the registration authority.

[0060] S4: After receiving the user's registration request, the registration authority generates a random number ri for the user, calculates RIDi=Proci⊕Resi,Ai=h(ri||x)⊕RPWi,Bi=h(x||y)⊕h(RPWi), transmits Ai and Bi to the user, and stores the user's registration information {RIDi,ri} in the edge AI.

[0061] S5: The user receives the message transmitted by the registration authority. The mobile device calculates h(ri||x)=Ai⊕RPWi,h(x||y)=Bi⊕h(RPWi),Ci=h(IDi||PWi)⊕ai,Di=h(ri||x)⊕h(RIDi||PWi),Ei=h(x||y)⊕h(RIDi||IDi||PWi),Fi=h(h(ri||x)||h(x||y)||RIDi||PWi||ai). Finally, {Ci,Di,Ei,Fi} are stored in the mobile phone.

[0062] (2) Smart device registration: A weakly physically unclonable function (WPUF) is embedded in each smart device; the registration authority selects an identity IDj for the smart device, generates a challenge Cj and the current registration timestamp Tj, and calculates Sj = h(IDj||x||Tj) and Rj = WPUF(Cj); {IDj, h(Rj) ⊕ h(x||Tj), Tj} is stored in the edge AI, and {IDj, Cj, Sj} is stored in the smart device;

[0063] The authentication phase includes login and mutual authentication; the user first logs in using a smartphone. After successful login, mutual authentication is performed with a remote smart device through edge artificial intelligence, and a secure session key for subsequent communication of the user is generated.

[0064] (1) Login: The user enters the username IDi and password PWi. The smartphone calculates ai=h(IDi||PWi)⊕Ci,RIDi=h(IDi||ai),h(ri||x)=Di⊕h(RIDi||PWi),h(x||y)=Ei⊕h(RIDi||IDi||PWi), and then calculates Fi=h(h(ri||x)||h(x||y)||RIDi||PWi||ai). If the Fi is equal to the stored Fi, the user logs in successfully.

[0065] (2) Mutual authentication: After successful login, the user and the remote smart device perform mutual authentication. The authentication steps are as follows:

[0066] The mobile device selects the smart device IDj to access, generates a random number n1, the current timestamp t1, and calculates M1=n1⊕h(h(ri||x)||t1),TIDi=RIDi⊕h(ri||x),Auth1=h(h(ri||x)||RIDi||TIDi||n1||t1),M2=IDj⊕h(Auth1||t1),M3=h(RIDi||PWi||n1||t1)⊕h(h(ri||x)||IDj||n1), and finally passes Meg1={M1,TIDi,M2,M3,Auth1,t1} to the edge AI.

[0067] After receiving a message, the edge AI checks its freshness, that is, whether the difference between the timestamp of the received message and the timestamp of the sent message is less than the preset maximum transmission delay. If the message is fresh, the edge AI calculates RIDi = TIDi ⊕ h(x||y), checks whether RIDi exists in its own database, if it exists, finds the ri corresponding to RIDi, and calculates n1 = M1 ⊕ h(h(ri||x)||t1), Auth1* = h(h(ri||x)||RIDi||TIDi||n1||t1), and checks whether Auth1* is the same as the received Auth1. If they are the same, the edge AI authenticates the user as real and calculates IDj = M2 ⊕ h(Auth1||t1), h(RIDi||PWi||n1||t1) = M3 ⊕ h(h(ri||x)||IDj||n1).

[0068] Edge AI checks if IDj exists in the database. If it does, it finds the corresponding h(Rj)⊕h(x||Tj),Tj, and recovers h(Rj).

[0069] Edge AI generates a random number n2, the current timestamp t2, and calculates Sj = h(IDj||x||Tj), M4 = n2⊕h(Sj||h(Rj)||t2), M5 = h(RIDi||PWi||n1||t1)⊕h(IDj||h(Rj)||n2||t2), M6 = h(x||n2)⊕h(Sj||h(Rj)||n2), Auth2 = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||h(Rj)||n2||t2); finally, the message Meg2 = {M4, M5, M6, Auth2, t2} is transmitted to the smart device.

[0070] After receiving a message, the smart device checks its freshness. If the message is fresh, the smart device calculates Rj = WPUF(Cj) based on Cj in memory, then calculates n2 = M4⊕h(h(IDj||x||Tj)||h(Rj)||t2),h(RIDi||PWi||n1||t1) = M5⊕h(IDj||h(Rj)||n2||t2),h(x||n2) = M6⊕h(Sj||h(Rj)||n2),Auth2* = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||h(Rj)||n2||t2), and checks whether Auth2* is equal to Auth2. If they are equal, it means that the smart device has authenticated edge AI.

[0071] The smart device generates a random number n3 and the current timestamp t3, and calculates M7 = n3 ⊕ h(h(RIDi||PWi||n1||t1)||IDj||t3), M8 = h(x||n2) ⊕ h(h(RIDi||PWi||n1||t1)||n3), SK = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||n3||t3), Auth3 = h(h(RIDi||PWi||n1||t1)||SK||IDj||t3); then, the smart device sends the message Meg3 = {M7, M8, Auth3, t3} to the user.

[0072] After receiving a message from a smart device, the user checks the freshness of the message. If the received message is fresh, the user calculates n3 = M7⊕h(h(RIDi||PWi||n1||t1)||IDj||t3), h(x||n2) = M8⊕h(h(RIDi||PWi||n1||t1)||n3), SK = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||n3||t3), Auth3* = h(h(RIDi||PWi||n1||t1)||SK||IDj||t3), and checks whether Auth3* is equal to the received Auth3. If they are equal, the user has authenticated the smart device. At this point, the user and the smart device have performed mutual authentication and generated a shared session key SK for subsequent secure communication.

[0073] like Figure 3 As shown, ① the user logs in using a smartphone and initiates an authentication request to the edge AI; ② the edge AI verifies the freshness of the message and authenticates the user's authenticity. After successful authentication, it initiates an authentication request to the smart device; ③ the smart device verifies the freshness of the message and authenticates the edge AI. After successful authentication, it initiates an authentication request to the user; ④ the user verifies the freshness of the message and authenticates the smart device. After successful authentication, a secure communication key is generated between the user and the smart device.

[0074] This invention provides a computer device, which includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the secure communication method for enabling the Internet of Things using generative artificial intelligence.

[0075] This invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the secure communication method for enabling the Internet of Things using generative artificial intelligence.

[0076] This invention provides an information data processing terminal, which is used to implement the secure communication system of the generative artificial intelligence-enabled Internet of Things.

[0077] The technical solution of this invention can be applied to generative artificial intelligence-enabled IoT systems to ensure communication security. A typical application example is intelligent predictive maintenance. Various sensors (such as temperature sensors and pressure sensors) or intelligent devices (such as CNC machine tools) within the factory collect real-time operating data and transmit it to generative artificial intelligence models for analysis via edge computing and a central cloud platform. These models can predict equipment failure trends and maintenance needs, storing the results in an encrypted database. Authorized maintenance personnel can remotely access these predictions and promptly schedule maintenance tasks, thereby avoiding production interruptions and equipment damage. This reduces maintenance costs and improves production efficiency and equipment reliability. If the communication security of the intelligent manufacturing IoT system cannot be guaranteed, the prediction results may be leaked or tampered with, potentially even affecting production safety and product quality.

[0078] In terms of security, the present invention can simultaneously achieve the following security attribute objectives:

[0079] Mutual authentication and key establishment, resistance to replay attacks, resistance to man-in-the-middle attacks, resistance to impersonation attacks, resistance to offline password guessing attacks, session key security, perfect forward secrecy, anonymity, resistance to desynchronization attacks, and resistance to the strongest combination of privacy attacks included under these security attributes.

[0080] Regarding communication costs, this invention requires relatively low communication costs. To fairly compare communication costs, the data size involved is uniformly defined. It is assumed that the bit length of various identities and temporary interaction numbers is 128 bits, the timestamp bit length is 32 bits, the block size for symmetric encryption / decryption is 128 bits, the hash function digest bit length is 256 bits, and the elliptic curve point length is 320 bits (this is because the security strength of 160-bit ECC is equivalent to the security strength of 1024-bit RSA). This invention requires the transmission of 3 messages, totaling 3040 bits. In other similar communication systems, the protocol of Srinivas-TDSC et al. requires the exchange of 3 messages, with a total communication cost of 2912 bits. The protocol of Wazid et al. requires the exchange of 4 messages, with a total communication cost of 3840 bits. The protocol of Srinivas-ITJ et al. requires the exchange of 3 messages, with a total communication cost of 3424 bits. The protocol of Chen et al. requires the exchange of 4 messages, with a total communication cost of 4480 bits. Zhang et al.'s protocol exchanged four messages, with a total communication cost of 4608 bits.

[0081] In terms of computational cost, this invention also has advantages. To facilitate comparison of the computational costs of different systems, T is defined. h T represents the hash function operation time. WPUF T represents the runtime of a physically unclonable function. ECC T represents the ECC dot product operation time. cm T represents the Chebyshev mixture mapping operation time. ed T represents the time required for symmetric encryption / decryption operations. fe This represents the operation time of the biometric fuzzy extractor. The experimental measurement used is: at the user device end, T h ≈T WPUF ≈0.00216, T ECC ≈T fe ≈3.3309ms, T cm ≈2.3891ms, T ed ≈0.4834ms; On the server side, T h ≈T WPUF ≈0.000038ms, T ECC ≈T fe ≈0.5078ms, T cm ≈0.3118ms, T ed ≈0.0082ms; On the IoT device side, T h ≈T WPUF ≈0.05835ms, T ECC ≈T fe ≈73.9319ms, T cm ≈45.4681ms, T ed ≈2.9556ms. The computational cost of this invention on mobile devices is 9T. h The computation time is approximately 0.01944ms, and the server-side computation cost is 13T. h ≈0.000494ms, the computational cost on the IoT device side is 8T. h +T WPUF The computation time was approximately 0.52515ms, with a total computation cost of 0.5451ms. In comparison, the total computation cost of the protocol proposed by Srinivas-TDSC et al. was 99.3118ms; the total computation cost of the protocol proposed by Wazid et al. was 7.1541ms; the total computation cost of the protocol proposed by Srinivas-ITJ et al. was 317.2325ms; the total computation cost of the protocol proposed by Chen et al. was 224.015ms; and the total computation cost of the protocol proposed by Zhang et al. was 7.1154ms.

[0082] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0083] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A secure communication system for the Internet of Things enabled by generative artificial intelligence, characterized in that, The system includes: The registration authority module is used for registration of IoT users and smart devices. It is a trusted server that registers entities in the IoT. The IoT user module connects to the registration authority module. IoT users register with the registration authority module using their smartphones. After logging in, they can control remote smart devices through edge artificial intelligence. The smart device module is used to collect data from the environment. After mutual authentication with IoT users, the smart device transmits the data to edge AI and IoT users. The edge AI module, connected to the IoT user module and the smart device module, is an edge server located close to the smart device. It deploys AI applications, processes data transmitted by the smart device, performs data analysis and decision-making, and assists IoT users and smart devices in performing mutual authentication.

2. The secure communication system for the Internet of Things enabled by generative artificial intelligence according to claim 1, characterized in that, The registration authority module is configured to perform operations during the registration and authentication phases, wherein: The registration phase includes user registration and smart device registration. The registration authority module stores identity registration and authentication information for IoT users and smart devices. The authentication phase includes user login and mutual authentication between IoT users and smart devices to ensure the security of subsequent communications.

3. The secure communication system for the Internet of Things enabled by generative artificial intelligence according to claim 1, characterized in that, The edge AI module further includes: The data processing unit is used to process and analyze data transmitted from smart devices and make decisions based on the analysis results; The authentication assistance unit is used to assist in mutual authentication between IoT users and smart devices, and generates a secure session key after successful authentication to enable secure communication between users and smart devices.

4. The secure communication system for the Internet of Things enabled by generative artificial intelligence according to claim 1, characterized in that, The intelligent device module includes: The data acquisition unit is used to collect data from the environment; The Weak Physically Unclonable Function (WPUF) module is used to generate response values ​​during the smart device registration phase to enhance the uniqueness and resistance to attacks of the device; The authentication unit is used to mutually authenticate with IoT users and edge AI modules during the authentication phase, ensuring data security during transmission.

5. A secure communication method for a generative artificial intelligence-enabled Internet of Things (IoT) based on the secure communication system of the generative AI-enabled IoT as described in claim 1, characterized in that, This method includes: a registration phase and an authentication phase; The registration phase involves registration by a registration authority for IoT users and smart devices; the authentication phase involves mutual authentication between IoT users and smart devices to achieve secure communication, specifically including: (1) User registration: S1: The user sends a registration request message Regi to the registration authority; S2: After receiving the user's registration request, the registration authority generates a response message Resi and sends (Regi, Resi) to the user; S3: The user selects username IDi and password PWi, and selects two random numbers ai and bi. Calculate RIDi = h(IDi||ai), RPWi = h(IDi||PWi||ai)⊕bi, Proci = RIDi⊕Resi, and then send {Proci,RPWi} to the registration authority. S4: After receiving the user's registration request, the registration authority generates a random number ri for the user, calculates RIDi=Proci⊕Resi,Ai=h(ri||x)⊕RPWi,Bi=h(x||y)⊕h(RPWi), transmits Ai and Bi to the user, and stores the user's registration information {RIDi,ri} in the edge AI. S5: The user receives the message transmitted by the registration authority. The mobile device calculates h(ri||x)=Ai⊕RPWi,h(x||y)=Bi⊕h(RPWi),Ci=h(IDi||PWi)⊕ai,Di=h(ri||x)⊕h(RIDi||PWi),Ei=h(x||y)⊕h(RIDi||IDi||PWi),Fi=h(h(ri||x)||h(x||y)||RIDi||PWi||ai). Finally, {Ci,Di,Ei,Fi} are stored in the mobile phone. (2) Smart device registration: A weakly physically unclonable function (WPUF) is embedded in each smart device; the registration authority selects an identity IDj for the smart device, generates a challenge Cj and the current registration timestamp Tj, and calculates Sj = h(IDj||x||Tj) and Rj = WPUF(Cj); {IDj, h(Rj) ⊕ h(x||Tj), Tj} is stored in the edge AI, and {IDj, Cj, Sj} is stored in the smart device; The authentication phase includes login and mutual authentication; the user first logs in using a smartphone. After successful login, mutual authentication is performed with a remote smart device through edge artificial intelligence, and a secure session key for subsequent communication of the user is generated. (1) Login: The user enters the username IDi and password PWi. The smartphone calculates ai=h(IDi||PWi)⊕Ci,RIDi=h(IDi||ai),h(ri||x)=Di⊕h(RIDi||PWi),h(x||y)=Ei⊕h(RIDi||IDi||PWi), and then calculates Fi=h(h(ri||x)||h(x||y)||RIDi||PWi||ai). If the Fi is equal to the stored Fi, the user logs in successfully. (2) Mutual authentication: After successful login, the user and the remote smart device perform mutual authentication. The authentication steps are as follows: The mobile device selects the smart device IDj to access, generates a random number n1, the current timestamp t1, and calculates M1=n1⊕h(h(ri||x)||t1),TIDi=RIDi⊕h(ri||x),Auth1=h(h(ri||x)||RIDi||TIDi||n1||t1),M2=IDj⊕h(Auth1||t1),M3=h(RIDi||PWi||n1||t1)⊕h(h(ri||x)||IDj||n1), and finally passes Meg1={M1,TIDi,M2,M3,Auth1,t1} to the edge AI. After receiving a message, the edge AI checks its freshness, that is, whether the difference between the timestamp of the received message and the timestamp of the sent message is less than the preset maximum transmission delay. If the message is fresh, the edge AI calculates RIDi = TIDi ⊕ h(x||y), checks whether RIDi exists in its own database, if it exists, finds the ri corresponding to RIDi, and calculates n1 = M1 ⊕ h(h(ri||x)||t1), Auth1* = h(h(ri||x)||RIDi||TIDi||n1||t1), and checks whether Auth1* is the same as the received Auth1. If they are the same, the edge AI authenticates the user as real and calculates IDj = M2 ⊕ h(Auth1||t1), h(RIDi||PWi||n1||t1) = M3 ⊕ h(h(ri||x)||IDj||n1). Edge AI checks if IDj exists in the database. If it does, it finds the corresponding h(Rj)⊕h(x||Tj),Tj, and recovers h(Rj). Edge AI generates a random number n2, the current timestamp t2, and calculates Sj = h(IDj||x||Tj), M4 = n2⊕h(Sj||h(Rj)||t2), M5 = h(RIDi||PWi||n1||t1)⊕h(IDj||h(Rj)||n2||t2), M6 = h(x||n2)⊕h(Sj||h(Rj)||n2), Auth2 = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||h(Rj)||n2||t2); finally, the message Meg2 = {M4, M5, M6, Auth2, t2} is transmitted to the smart device. After receiving a message, the smart device checks its freshness. If the message is fresh, the smart device calculates Rj = WPUF(Cj) based on Cj in memory, then calculates n2 = M4⊕h(h(IDj||x||Tj)||h(Rj)||t2),h(RIDi||PWi||n1||t1) = M5⊕h(IDj||h(Rj)||n2||t2),h(x||n2) = M6⊕h(Sj||h(Rj)||n2),Auth2* = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||h(Rj)||n2||t2), and checks whether Auth2* is equal to Auth2. If they are equal, it means that the smart device has authenticated edge AI. The smart device generates a random number n3 and the current timestamp t3, and calculates M7 = n3 ⊕ h(h(RIDi||PWi||n1||t1)||IDj||t3), M8 = h(x||n2) ⊕ h(h(RIDi||PWi||n1||t1)||n3), SK = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||n3||t3), Auth3 = h(h(RIDi||PWi||n1||t1)||SK||IDj||t3); then, the smart device sends the message Meg3 = {M7, M8, Auth3, t3} to the user. After receiving a message from a smart device, the user checks the freshness of the message. If the received message is fresh, the user calculates n3 = M7⊕h(h(RIDi||PWi||n1||t1)||IDj||t3), h(x||n2) = M8⊕h(h(RIDi||PWi||n1||t1)||n3), SK = h(h(RIDi||PWi||n1||t1)||h(x||n2)||IDj||n3||t3), Auth3* = h(h(RIDi||PWi||n1||t1)||SK||IDj||t3), and checks whether Auth3* is equal to the received Auth3. If they are equal, the user has authenticated the smart device. At this point, the user and the smart device have performed mutual authentication and generated a shared session key SK for subsequent secure communication.

6. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the secure communication method for enabling the Internet of Things using generative artificial intelligence as described in claim 1.

7. A computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the steps of the secure communication method for enabling the Internet of Things using generative artificial intelligence as described in claim 1.

8. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the secure communication system for enabling the Internet of Things with generative artificial intelligence as described in claim 1.

Images