Security analysis and provision of authentication protocol, based on peer-to-peer structure in IOT platform

Abstract

The Internet of Things technology allows you to transfer data to any asset through communication networks such as the Internet or intranet. One of the most important problems is security, which includes confidentiality, authenticity validation, and completeness. Among security problems, the most important ones are ensuring authenticity and protecting privacy. The elliptic curve encryption algorithm is famous protocols for maintaining privacy and verifying authenticity. Despite the capabilities they have and provide good security against network attacks, they face challenges such as response time, delay in the authentication and authentication process, as well as the amount of memory consumed. Two-factor authentication (2FA) is a security process where users provide two different authentication factors to authenticate themselves. It also provides a higher level of security than authentication methods that rely on single-factor authentication (SFA). This article presents two novel methods for 2FA that use encryption techniques, scramble architecture, and chord architecture. These methods allow for the authentication process to be initiated by both parties involved in the communication. Three scenarios and three metrics of reaction time, memory usage, and connection latency were used to assess the efficacy of the suggested approach. The average reaction time is improved and is 0.012 s when the lengths of the finger table in the chord structure are increased to 10, 15, and 20. The findings show that the suggested approach reduces computing complexity, reaction time, and communication latency. It has an average reaction time of 0.016 s, an average memory use of 30,360 kb, and an average connection latency of 0.042 s.

Introduction

The Internet of Things (IoT) is a network of interconnected electrical devices and sensors that communicate information with each other. This network enables remote measurement and control^1–3. This network enables the transmission of data across communication networks, such as the Internet or intranet, for any entity or object. It also allows for the automated and intelligent control and management of a group of systems by current applications. The Internet of Things incorporates several technologies like wireless sensor networks, machine-to-machine communication, robotics, Internet technologies, and smart gadgets^4,5. The Internet of Things is a technical innovation that enables any object to transmit data over communication networks, such as the Internet or intranet^6,7. An essential benefit of the proliferation of the Internet of Things is the capacity to link various things and gadgets to the virtual realm. This means that all objects, even non-living ones, own their unique digital identity, enabling computers to arrange and control them^8,9. Confidentiality, integrity, authentication, and access control techniques are necessary for the functioning of the Internet of Things^10,11. The proliferation of Internet connectivity in billions of products and instruments poses a significant rise in possible security vulnerabilities in the virtual realm. This is due to the fact that in the next years, many home appliances, automobiles, doors, and other items will be integrated into the information technology sector. The Internet of Things poses significant challenges in terms of security and privacy, necessitating proactive efforts by authorities to address these issues^12–14.

The security of the Internet of Things refers to the measures used to protect the information that is communicated over the global Internet network. These measures aim to prevent eavesdropping, infiltration, and purposeful or systematic manipulation of data between command servers and different devices^15–17. The Internet of Things (IoT) consists of three physical levels: the network layer, the application layer, and the security layer, which should be implemented across all these layers. Security must be thoroughly observed at the physical layer for data collection, in the network layer for routing and transmission, and in the application layer for preserving secrecy and confirming authenticity^18,19. Security is a crucial aspect that must be taken into account in the majority of applications in the Internet of Things²⁰. However, it is important to note that traditional security measures cannot be directly implemented in Internet of Things technologies because of the varying standards and large quantities of current connections^21,22. Simultaneously, the large quantity of interconnected devices presents the obstacle of scalability²³. The topic of Internet of Things security focusses on concerns related to secrecy, authentication, and integrity^24,25. IoT device manufacturers prioritize cost, size, and efficiency above security, resulting in a disregard of security measures. Consequently, the devices become vulnerable and convenient for potential attackers. The sensors used in physical objects for data gathering are constrained by restricted resources and computing capabilities, mostly owing to their tiny size. Additionally, they face challenges from standard security methods and intricate computational algorithms²⁶. The emergence of smart cities and home settings has led to a transformation in the idea of user authentication. It now encompasses more than just preventing unauthorized access to linked information. It also offers possibilities for tailored services that cater to individual users. Biometric technology, such as biometrics, is used for authentication. Its primary objective is to provide alternative techniques or substitutes for existing access control systems. These approaches are employed to safeguard personal information and organizational assets²⁷. The primary concern with security in the Internet of Things is the challenge of validating the genuineness and completeness of data. Implementing authentication in the Internet of Things is challenging due to the need for servers and suitable infrastructures to facilitate communication between various components. In the present period, as smart cities and smart homes continue to advance, the notion of user authentication serves the dual purpose of preventing unauthorized access to sensitive information and facilitating the provision and advancement of services⁸. This article presents a protocol for authentication in the Internet of Things platform, which is based on a peer-to-peer structure. The authentication process is of great relevance in the Internet of Things and its applications in service creation. The research aims to achieve several key objectives, including accurate and correct authentication, ensuring non-repudiation of communication, enhancing the protocol’s security and computing capabilities, providing resistance against various attacks, optimizing memory consumption, minimizing communication delay time, and reducing response time to requests.

The primary obstacle in establishing a secure connection inside the Internet of Things, as the number of sensors grows, is the need to authenticate their validity. When considering these factors, it is important to take into account anonymity, low weight, and dependability²⁸. A suggested authentication mechanism for Internet of Things devices, as described in²⁹, utilizes hashing techniques for the encryption process. However, a more efficient approach is suggested in³⁰ for this objective, using XOR-based encryption instead of more intricate encryption methods like hashing functions. A novel authentication system is presented in³¹ for Internet of Things (IoT) devices in a distributed environment. The protocol is designed to be lightweight, making it suitable for resource-constrained devices. The results demonstrate that the suggested protocol effectively safeguards the network environment from all potential security risks. The research and comparison of its performance further substantiates that the suggested procedure outperforms comparable techniques across many parameters. A novel authentication system is presented in³¹ for Internet of Things (IoT) devices in a distributed environment. The protocol is designed to be lightweight, making it suitable for resource-constrained devices. A novel remote authentication system is introduced in³² for secure communication in the Internet of Things, using elliptic curve encryption. The study of the data has assessed the security of the suggested protocol in addition to proving its lightweight nature.

This article is a text document organized in 6 sections. Section 1 deals with the introduction and overview of related topics. In Sect. 2, the relevant literature is discussed, and recent relevant previous works are reviewed. The basic method is introduced in Sect. 3. In Sect. 4, a new proposed method for improving authentication in the Internet of Things is presented. Section 5 of the proposed method is implemented in a suitable software environment and after performing simulations, the findings are analyzed. Finally, in Sect. 6, the findings are discussed and suggestions for future work are presented.

Related works

A light authentication strategy for heterogeneous wireless sensor networks (WSNs) in the Internet of Things was presented in³³. The technique relies on a unidirectional scrambling function, which needs little computations and is very energy-efficient. Nevertheless, it necessitates several security vulnerabilities. Subsequently, the authors³⁴ introduced a sophisticated and enhanced protocol to rectify vulnerabilities in security³³. Although³⁵ discovered that the design used in³⁴ is inefficient, it is also vulnerable to a range of assaults. Two authentication and key agreement procedures have been developed for body area wireless networks to provide secure communication in health care programs inside the Internet of Things³⁶. Both protocols use elliptic curve encryption and message confirmation code to ensure the secrecy, integrity, and authenticity of information. Nevertheless, they are susceptible to a wide range of security threats, including spoofing, reflection, and sensor node assaults. In their study, the authors focused on identifying and addressing several vulnerabilities in the authentication system proposed in³⁸. They also introduced an enhanced scheme for Wireless Sensor Networks (WSNs) that is built upon the principles of the Internet of Things. The suggested solution utilizes elliptic curve encryption and one-way hashing algorithm, offering several security measures for the Internet of Things. Nevertheless, reciprocal authentication is not cost-effective in terms of communication expenses. Furthermore³⁹, demonstrated that³⁸ is susceptible to both DDOS attacks and identity spoofing. The authors in⁴⁰ highlighted that the authentication technique discussed in⁴¹ is vulnerable to many forms of security attacks. The authors suggested a robust authentication mechanism for enabling multimedia communication in Wireless Sensor Networks (WSN) with Internet of Things (IoT) capabilities. The suggested protocol fulfils several security requirements. Nevertheless, it is not lightweight in terms of transmission and storage expenses. The article in⁴² highlights the lack of security in the authentication procedures described in⁴³ and ⁴⁴. It proposes an enhanced protocol based on elliptic bending as a solution. It has successfully resolved all security issues⁴³ and ⁴⁴. Nevertheless, due to the substantial computational and communication expenses involved, it is unsuitable for Internet of Things applications. An analysis of prior studies reveals that although the current solutions have offered sufficient security for the Internet of Things (IoT) environment, there is still potential for improvement. Simultaneously, they also experience a significant level of computational complexity. Developing an authentication protocol with sufficient security against different attacks is considered a challenge, since it requires improving efficiency and reducing computing complexity compared to earlier techniques.

Basic method

CoAPMicro is a compact and efficient library created to improve the security and effectiveness of Internet of Things (IoT) connections, particularly using the Constrained Application Protocol (CoAP). It works on top of CoAPSharp and offers a secure authentication system that is both strong and efficient in terms of resources. This library is renowned for its exceptional ability to minimize memory use, shorten handshake length, decrease average reaction time, and lower computation cost. As a result, it is highly ideal for IoT contexts where these variables are of utmost importance. Because every physical component in the Internet of Things architecture has distinct qualities of its own, security is a difficult problem to solve. With this technology, it is necessary to verify the legitimacy of each person, item, and system that is linked to the Internet. Weak or abandoned authentication procedures allow hackers to enter the network and carry out a variety of destructive tasks, each of which might have a different outcome, such delivering incorrect data or turning off the security system. Many of the IoT devices in use today lack security features, in spite of all these concerns. As a result, we might claim that the things we use are susceptible to various security flaws. These gadgets introduce consumers to a new age of cybercrimes rather than making life better. A lightweight mutual authentication strategy employing payload encryption is presented in⁴⁵ in response to the aforementioned situations and the significance of Internet of Things authentication. CoAP is used as the foundational protocol to meet the criteria of the scheme. Four fundamental components make up the basic design that is examined in this article: (1) The fundamental design verifies the communication partner’s and client’s legitimacy. This technique is robust and easy to compute, requiring minimal communication at all. Additionally, the client and server authenticate one another by generating encrypted payloads as a challenge. (2) The fundamental plan exchanges resources between the client and the server using the characteristics of the CoAP style. The Internet of Things’ transmission layer is where security is addressed, and DTLS is used there. The DTLS technique is computationally demanding and uses a lot of network resources (remember that the Internet of Things is not resource-rich). All the capabilities required to guarantee security in the Internet of Things are included in the basic plan, which takes the role of DTLS. In contrast to DTLS, the fundamental approach aims to lower communication and processing costs without adding a new layer. (3) The basic approach lessens the likelihood of resource depletion and denial-of-service attacks by limiting malevolent nodes by allowing them to establish many connections with the server at once. In the basic architecture, each client is permitted to create a single connection with the server in order to equitably utilize the limited resources. (4) The fundamental architecture not only lessens DoS assaults but also withstands recurrent ones. The fundamental plan uses the notification server’s subsequent updates as a way to identify re-attacks.

For secure resource sharing, almost all Internet of Things systems using CoAP depend on DTLS. An extra protocol layer for security is provided by the CoAP stack with DTLS capabilities. The cost of calculation and communication will go up as a result. This is true even if the basic design does not include any new layers. The security of the communications sent back and forth between the client and the server is unaffected by the lack of an extra protocol layer. The session key is sent in the message payload that is sent over during the request-response exchange between the client and the server to enable authentication. Security elements have been introduced to CoAP to make it reliable, effective, and safe from hostile activity in order to guarantee authenticity in the fundamental architecture. The fundamental method of authentication, in contrast to DTLS encryption approaches, involves the client and server exchanging the message content while the client and server challenge one another. There are four messages used throughout the whole procedure. The fundamental plan’s characteristics are: A maximum of 256 bits are retained in the payload of every communication. 2. The CBC working mode of the AES encryption protocol is used. 3. For Internet of Things nodes with constrained resources, an 18-bit key will do. Generally speaking, there are four fundamental phases in the basic strategy. These steps are: 1- Beginning of communication 2- Server challenge 3- Client response and challenge 4- Server response.

In accordance with the Internet threat model, it is presumed that end devices are secure in order to prevent violating security standards. In production, the client has the original key (pre-shared) as a hard code. An alert alerting the user to the security vulnerability will be created if an attacker attempts to modify the client. Every client sends a Hello Message request to the server to start the session. It is appropriate for the server to initiate the session with this message at the outset of the task. There’s a token for every communication to get a response. The payload of this message contains the client ID, and the request has two options, Auth and Auth-MsgType, the values of which indicate what action should be taken on the received packets on the server. The authentication resource is generated on the server upon the client ID’s successful authentication. A request to initiate a session with the server is indicated by the combination of Auth = true and Auth-MsgType = 0. In the first stage, the values of λi and µkey are subjected to the XOR operator using Eq. 1. The XOR function is widely used in complicated cryptography and is inexpensive to compute. Furthermore, the original text may be recovered twice thanks to this process, which yields no information about the original text; in this instance, the plaintext serves as the session key.

1

Applying Eq. 2, the 128-bit ψresultant is added to ηserver and encrypted with λi. In this way, the 256-bit payload of the server challenge is completed and ready to be sent to the client.

2

During the client and challenge phase, the client decrypts the γserver-payload and obtains the session key µkey. Upon successful completion of this task, the client will ultimately get the accurate ηserver and µkey. The server and key are only known to the server, whereas λi is associated with a particular client. Only the authorized client may decode this challenge. The attacker is limited to intercepting communications on the ηserver and µkey. Customers use the parameter λi to decipher the data packets. Upon successful decryption, the customer has verified its legitimacy. As mutual authentication necessitates the verification of both parties, the server must likewise validate its authenticity. The client generates a new encrypted payload that is comparable to the server’s payload and performs an XOR operation between ηserver and λi, as specified in Eq. 3.

3

Applying Eq. 4, 128-bit ψresultant is added to ηclien and encrypted with µkey. In this way, the 256-bit payload of the client challenge is completed and ready to be sent to the server.

4

During the server and challenge phase, the server deciphers the γclient-payload and looks for the ηserver. If the search yields a favorable outcome, it is determined that the client has successfully verified its authenticity. The server then appends the µkey to the client by obtaining ηclient and constructs an encrypted payload based on Eq. 5. Subsequently, it transmits the payload, which has been encrypted using a 256-bit encryption algorithm, to the client.

5

By following the aforementioned procedure, the server’s resource status is changed to “Authenticated” as a result of the client successfully authenticating themselves. The client has not yet validated the legitimacy of the server. Consequently, the client decrypts the γserver-payload and looks for ηclient. If the search result is affirmative, the verification of the server’s legitimacy is confirmed. Thus, the client and the server mutually authenticate each other’s legitimacy.

Proposed method

Through the conducted research on the fundamental design, it has been established that the CoAPMicro technique involves four phases of message transmission and several stages of encryption using distinct encryption keys, namely the temporary key Mkey and the fixed key. Furthermore, this authentication technique is initiated by the service provider, with each side of the connection acting as the initiator of the authentication process by submitting a plan. Authenticity verification may be achieved across several items in the Internet of Things ecosystem without relying on a central server. In the proposed strategy, two novel methods are suggested to enhance the two-way authentication process in the Internet of Things. These methods allow both sides to initiate the authentication mechanism. The suggested strategy consists of two primary methods: the first approach involves authentication by hashing, while the second approach involves authentication utilizing the Chord architecture.

First proposed approach

In this method, a message identifier Pid is generated by using a linear formula with predefined coefficients. The algorithm takes into account the size of the packets and the time at which it is transmitted, as provided in the message Payload. The coefficients are mutually agreed upon by the authorized nodes of the network.

6

where Pt is the time at which the packet is sent and Pv represents the size of the packet. The constants α and β are preprogrammed into the Internet of Things nodes to safeguard them against modifications in the event of a physical breach by an attacker⁴⁶. Figure 1 depicts the block diagram of the first suggested technique.

Fig. 1.

Block diagram of the first proposed approach.

Based on the illustration, every network node that want to initiate communication determines the size of the packet (v) it transmits by generating a Pid using Eq. 6. The message objectively includes both Pt and Pv. Any node that incorrectly gets the encryption key three times and transmits a message encrypted with the incorrect encryption key will be forbidden from requesting and communicating for a certain duration of time. The authentication process between the service provider and the service receiver consists of three communication phases. Firstly, node A generates a message identification titled Pid by applying a linear formula using the approved coefficients α and β, if they are available and authorized. 2- The message identification undergoes a hashing process, and the resulting value of key1 is used as the encryption key. Node A considers a communication identification called µ and encrypts it using the key µ. Subsequently, the encrypted message is sent to node B, along with the payload that includes the packet details, such as the time and volume.

7

Node B at the destination verifies the received packet and extracts the payload, which includes the time and volume of the packet. After that, if it is permitted and contains α and β coefficients, it creates a message identification and uses hashing to get the decryption key. The identical password key1 has to be created if the user is authorized and has produced the right ID. Using this key, decryption, and the µ value, the message delivered by node A may be recovered.

8

Node B generates λ, the connection identification, at the destination. Then, using hashing key2, a Pid is produced depending on the packet sending time and volume. As the content of the new packet, the identification µ is appended to the identifier λ, encrypted using key2, and sent to node A.

9

In node A, the time and volume of the delivered packet are retrieved from the payload content. Based on this information, a Pid is produced by rearranging the Key2. Now that node B’s packet has been decoded, λ and µ may be found.

10

If µ represents the accurate value, then node B is verified as legitimate for node A. Node A now generates a Pid by shuffling Key3, taking into account the timing of packet transmission and its volume. Node A creates a session key, denoted as skey, with a length of 128 bits. Node A concatenates the session key with ID λ and encrypts the resulting value using key3. It then transmits the encrypted value to node B in a message packet.

11

In node B, the time and volume of the transmitted packet are retrieved from the payload content of the message. Based on this information Pid, a is formed by hashing the key3, and it is used to decode the message. The program retrieves the lambda identifier and compares it with the identifier. If the values are same, node A is verified, and sKey is used as the session key.

12

13

Second proposed approach

In this method, a message identifier Pid is generated when a packet is sent. The identifier is created using a linear formula with predetermined coefficients, which are agreed upon by the authorized nodes. The formula takes into account the size of the packet and the time of sending, specified in the message Payload.

14

Here, Pt represents the time it takes to transmit a packet, and Pv represents the size of the packet. The constants α and β are preprogrammed into the Internet of Things nodes, and measures are taken to safeguard them against modification in the event of physical access by an attacker to the Internet of Things node³⁴. Figure 2 depicts the block diagram of the second suggested technique.

Fig. 2.

Block diagram of the first proposed approach.

At this point, using the highest ID value of the created message, a random selection of IDs is designated as warehouse IDs. These IDs are chosen from a range of 1 to 10,000, for instance. A Chord structure is established, with each warehouse’s ID being identical to the produced IDs. The IDs produced at this stage pertain to the warehouses inside the Chord structure, each of which has a code key. This structure undergoes regular updates and is sent to all authorized nodes of the network. This update may be performed on either the keys, values, or both. The procedure of upgrading this structure involves encrypting the new structure using a predetermined key, such as γ, which is hardcoded in the authorized nodes. It is openly disseminated on the network. Authorized nodes use and utilize this framework to ensure decentralization and eliminate any single point of vulnerability in the plan. The encryption keys are generated by a process of hashing using a hashing algorithm. Figure 3 displays the password keys found inside the compromised storage units of the Chord structure.

Fig. 3.

Stores on the Chord structure to hold encryption keys.

Every network node that wishes to commence communication produces a Pid using Eq. 14, by sending a packet with certain sizes (v). The message objectively includes both Pt and Pv. Conversely, if a user enters an incorrect password three times and attempts to send a message using an incorrect password key, they will be forbidden from making further requests or interacting for a certain duration of time. The crucial factor is the impact of several warehouses on the system’s scalability and the acceptability of the search time required to locate the relevant warehouse ID. Therefore, the suggested strategy would use the Chord structure, which enables rapid data retrieval as needed. The IDs in this structure are derived from the process of hashing. In the suggested approach, the transmitting ID is associated with the Chord structure, allowing us to locate a specific warehouse that has the password key by searching for the corresponding key inside this structure. Due to the data amount and time of transmission, the node (provider or receiver) is assigned to a certain warehouse. Consequently, the packet supplied by the user may be reassigned to a different warehouse at a later time. In the Chord system, the nodes are arranged in a ring structure, and a data item with a key k is assigned to the nodes with the lowest ID that meets the criterion k= (see Fig. 4).

Fig. 4.

ID ring in the proposed method.

Each node in the Chord protocol simply requires knowledge of how to interact with its immediate successor node. Surrogate pointers may be used to go around the loop and find queries for a given key ID. This procedure continues until the queries come across a pair of nodes where the desired key ID is situated between them. The second node in this pair is the node that provides the response to this query. Figure 5 depicts an instance where node 8 does a search operation for the key value of 54. Node 8 invokes the procedure for key 54, which ultimately yields the substitution of that key (i.e., node 56). This query traverses each node in the ring, starting with node 8 and ending at node 56. The query’s response is returned to node 8 following the reverse route of the query’s journey.

Fig. 5.

The path taken by the query from node 8 for key 54.

In the given search strategy, the quantity of messages sent to the nodes increases proportionally with the expansion of the node count. Chord holds supplementary route information to enhance the speed of search operations. This further information is not essential for the accuracy of the system, but the process will continue until each node discovers its accurate substitute. Let’s assume that m represents the number of bits in the node/key ID, and each node keeps a finger table with a minimum of 1 entry. The table referred to as the index table includes information on the nodes in a network. Specifically, the i-th item of the table for node n indicates the identity of the first node (s) that can be found on the ID ring, with a minimum identifier distance of 2−1 from node n. During this procedure, node S is referred to as the i-th finger of node n.

Protocol security

In the suggested methods, a malicious individual may intercept the wireless connection between the authorized nodes of the network and get the unencrypted details of the message, including the packet size and the time of transmission. However, the network nodes do not provide access to the keys α and β, which are fixed coefficients in the linear formula. As a result, it is prohibited from generating Pid. Meanwhile, the password key is derived from the abstract Pid. Consequently, the key becomes inaccessible, rendering the attacker unable of deciphering the intercepted packets, and hence, the desired outcome is not achieved via this endeavor. If an assailant attempts to establish communication with an authorized node by imitating a packet. Due to the attacker node’s lack of knowledge about the constants and linear formula, it is unable to produce the key used to encrypt its message content by the authorized node. Therefore, the generation of the suitable parameters will not occur, rendering the continuation of the protocol procedure unfeasible.

In the proposed protocol, any user who obtains the erroneous encryption key three times and the message he sends is encrypted with the wrong encryption key would be prevented from re-requesting and communicating for twait. So, as a consequence, an attacker cannot keep an authorized node busy all the time. To prevent 1Dos attacks, the network architecture packet and the current communication protocols ID should be put to the blacklist. The keys needed to encrypt the message content are obtained using the Chord architecture in the second suggested method. This structure has a lot of search space since it contains a lot of warehouses. To preserve the system’s relative security, only authorized network nodes may access this structure. The Chord structure is redesigned in rounds to optimize security. You may change the values, the keys, or both. The new structure is encrypted using a pre-agreed key, like μ, which is hardcoded in the authorized nodes, in order to update this structure. The network broadcasts it to the general audience. This structure is received by authorized nodes, who then swap it out for the existing one. Thus, it may be said that in this instance, the attacker’s potential for vandalism is diminished and the system’s harm is minimized.

The suggested protocol is structured in a peer-to-peer style, enabling authorized nodes to authenticate each other and exchange session keys without relying on a centralized server. The absence of a server results in the lack of storage for the keys. Therefore, the occurrence of flood requests, as described in assaults such as DoS, will be completely eliminated. Thus, the design’s resilience to network assaults has been shown. In addition to DoS and man-in-the-middle attacks, the suggested protocol is also immune to replay assaults. Since the message body of the proposed protocol includes time as a parameter to generate Pid, it is ensured that Pid is a distinct value at each instance. The suggested protocol’s design presents a challenge for potential attackers attempting to execute a replay assault. If the time is altered as well, the resulting key will be incorrect and the process of decryption will become challenging. The first suggested method used the hashing result Pid as the encryption key. In the second suggested solution, the hashing result of Pid is converted into a chord structure and the output of the chord structure search function is used as an encryption key throughout the protocol’s execution. Therefore, the second recommended solution is more secure than the first proposed one. Upon the completion of a successful mutual authentication process, both the receiver and the server are granted permission to use the shared key (sKey) for the purpose of exchanging data packets.

Implementation of proposed method

The suggested approach has been implemented using the MATLAB programming language in this study. MATLAB is a sophisticated software system designed for doing mathematical and technical calculations. It utilizes a high-level programming language that incorporates a vast array of instructions specifically tailored for mathematical computations. MATLAB offers a wide range of application libraries in many domains like signal processing, artificial intelligence, neural networks, machine learning, and solving differential and Fourier equations. Additionally, users have the ability to design and include their own distinct libraries. Furthermore, plugins may also be used. The data necessary for analyzing the outcomes has been generated using a random and simulation-based technique. This collection has been used as input for both the suggested design and the basic design under exactly the same circumstances. The suggested technique is a hybrid strategy that aims to guarantee lightweight authenticity in the Internet of Things. This is achieved by using the Chord architecture to enhance the outcomes. Therefore, during the simulation phase, it is essential to precisely and properly measure certain factors that have an impact on the outcomes. Table 1 presents the key parameters that are specified and valued throughout the implementation process.

Table 1.

Introduction and setting of simulation parameters.

Variable	Simulation scenario	Value
Number of warehouses in structure of cord	Yes	20,30,40
Number of hashing times	No	3
Equivalent string length for hashing	No	4
Size of fingerboard length	Yes	10,15,20
Range of random identifiers	No	1-1000
Number of packets	Yes	20,40,60,80,100

To assess the effectiveness of the suggested approach, three metrics have been employed: overall delay in the communication process, response time, and memory utilization. Table 2 presents these indicators.

Table 2.

Introduction of evaluation indicators.

Indicator	Indicator description
Handshaking delay	In computer science, handshaking is the phrase used to describe the process of initiating communication between two systems. When one system transmits a message to another system over a communication channel Upon receiving the request, a connection is established, initiating the handshaking activity. Subsequently, the two systems engage in a series of message exchanges to mutually agree on establishing a connection and defining the communication protocol to be used. This assessment index examines the overall delay rate of the communication process in both the suggested and fundamental methods.
Response time	This indicator measures how long it takes to submit a user request and obtain the first response. The suggested technique and the basic plan were evaluated in terms of efficiency using this index.
Memory consumption	This index examines and evaluates the amount of memory used by the fundamental and suggested research designs in order to verify mutual legitimacy.

The efficiency of the suggested technique has been assessed in this part, using the scenarios outlined in Table 3 and the assessment indicators presented in Table 2. The results have been compared with those of the basic plan.

Table 3.

Introduction of scenarios designed for simulation execution.

Scenario title	Description
Checking effect of number of packets	This study examines the efficacy of the suggested approach in three assessment metrics: overall delay in the communication process, reaction time, and memory consumption. The examination includes 20, 40, 60, 80, and 100 packets, and the results are then reviewed.
Investigating the effect of number of warehouses in chord architecture	This study examines the efficacy of the suggested technique in three assessment metrics: overall delay in the communication process, reaction time, and memory consumption. The investigation focusses on 20, 25, and 30 stores using the Chord architecture. The results are then analyzed.
Investigating the effect of fingerboard length	This review examines the efficacy of the suggested approach in measuring three evaluation metrics: response time, total delay in communication, and amount of memory used with various fingerboard lengths (10, 15, and 20) in the chord structure. The results are then discussed.

First scenario

This study examines the impact of different fingerboard lengths (10, 15, and 20) on the efficiency of the proposed approach. The assessment focusses on three indicators: overall delay in the communication process, reaction time, and memory consumption. This study assesses the duration between the user initiating a request and receiving the first response for the suggested plan. Accordingly, the simulation was conducted 30 times for varying numbers of packets (20, 40, 60, 80, 100), and the average outcomes were computed. Upon analyzing the results for the lengths of 10, 15, and 20 fingerboards in the Chord structure, it is evident that the average reaction time for the suggested approach is 0.012 s (Fig. 6).

Fig. 6.

Investigating the effect of the length of the fingerboard on the efficiency of the proposed method in the response time index in seconds.

Additional results from this research suggest that as the length of the fingerboard in the Chord structure increases, the reaction time reduces. Hence, it can be said that there exists a substantial correlation between the augmentation of the finger table’s length and the reaction time in the suggested approach. This assessment index assesses the memory consumption of the suggested approach. Consequently, the simulation was executed 30 times for varying numbers of packets (20, 40, 60, 80, 100) and the mean outcomes were computed. Upon analyzing the data shown in Fig. 7, it is evident that the average memory use for the suggested strategy amounts to 30,660 kilobytes.

Fig. 7.

Investigating the effect of the length of the index table on the efficiency of the proposed method in the memory consumption index in terms of kilobytes.

Additional research results show that the suggested method’s memory usage reduces as the fingerboard length in the Chord structure lengthens. It is thus possible to assert that there is a clear and significant correlation between the quantity of memory used and the growth in the finger table’s length. The communication delay rate for the suggested approach has been examined in this indication. On the basis of this, the simulation was run thirty times for each of the following number of packets (20, 40, 60, 80, and 100), and the average outcomes were determined. Analyzing the results in Fig. 8, it can be seen that the suggested solution has an average transmission latency of 0.035 s.

Fig. 8.

Investigating the effect of the length of the index table on the efficiency of the proposed method in the index of delay between connections in seconds.

Other results from this research show that extending the length of the finger chart in the Chord structure reduces transmission latency. As a result, it is possible to conclude that there is a substantial association between increased finger table length and communication delay rate.

Second scenario

This study examines the impact of the number of warehouses in the Chord architecture (20, 30, 40) on the efficiency of the suggested technique. The investigation focusses on three evaluation indicators: overall latency in the communication process, response time, and memory consumption. This study assesses the duration between the user initiating a request and receiving the first response for the suggested plan. Consequently, the simulation was executed 30 times for varying packet quantities (20, 40, 60, 80, 100), and the mean outcomes were computed. Upon analyzing the data shown in Fig. 9, it is evident that the average reaction time for the suggested technique is 0.016 s while considering 20, 30, and 40 warehouses in the Chord structure.

Fig. 9.

Investigating the effect of the number of warehouses of the Chord architecture on the efficiency of the proposed method in the response time index in seconds.

Additional results from this research suggest that when the number of warehouses in the Chord structure grows, there is a corresponding rise in communication latency. Hence, it can be said that there exists a clear and substantial correlation between the rise in warehouse quantity and the rate of communication delays. This assessment index focusses on assessing the memory consumption of the suggested approach. The suggested design has undergone evaluation. Consequently, the simulation was executed 30 times for varying numbers of packets (20, 40, 60, 80, 100), and the average outcomes were computed. The analysis of the results obtained from 20, 30, and 40 warehouses in the Chord structure reveals that the average memory use for the suggested approach is 30,360 kilobytes (as seen in Fig. 10).

Fig. 10.

Investigating the effect of the number of storages of Chord architecture on the efficiency of the proposed method in the memory consumption index in terms of kilobytes.

Additional results from this research demonstrate that the expansion of Chord architecture’s warehouse count directly impacts memory use in the suggested approach. Moreover, as the number of warehouses increases, so does the quantity of memory consumed. This assessment index examines the rate of communication latency for the suggested approach. An analysis of the results obtained by varying the number of Chord architecture warehouses (40, 30, 20) and the number of data packets (20, 40, 60, 80, 100) reveals that the average communication latency for the proposed technique is 0.042 s (refer to Fig. 11).

Fig. 11.

Investigating the effect of the number of warehouses of Chord architecture on the efficiency of the proposed method in the index of delay between connections in seconds.

Additional research results show that the reaction time rises as the number of warehouses in the Chord structure grows. Thus, it can be said that there is a clear and substantial correlation between the number of warehouses growing and the pace at which communications are delayed.

Third scenario

In this case study, the impact of varying the number of data packets (20, 40, 60, 80, 100) on the effectiveness of the suggested approach is examined based on three assessment metrics: response time, total communication delay, and memory consumption. The results are then juxtaposed with those of the base design. In this research, the basic and recommended designs’ 30-time implementation of the time it takes to make a user request and get a response was examined, and the results were assessed. Analyzing the results for the quantity of distinct data packets (20, 40, 60, 80, and 100) reveals that the average reaction time for the basic plan is 0.026 s, while the suggested method’s average is 0.016 s. The study’s findings are shown in Fig. 12.

Fig. 12.

Evaluation of the efficiency of the proposed method and the basic design in response time index in seconds.

It is deemed an acceptable design with a lower reaction time in this study. As a result, it may be said that the suggested approach performs better than the original concept. The whole communication process delay in the suggested approach and the fundamental design has been calculated and is included in this assessment index. Based on this, the simulation for the standard and suggested designs has been run thirty times in identical settings, and the results have been assessed. The results indicate that the average delay of all communications in the suggested technique is 0.042 s, and for the basic plan it is 0.094 s (Fig. 13). This is due to the examination of the findings for the number of distinct data packets (20, 40, 60, 80, 100).

Fig. 13.

Evaluating the effectiveness of the proposed method and the basic design in the index of delay between communications in seconds.

It is ideal to have a workable strategy with minimum delay for this evaluation. As a result, it may be said that the suggested approach performs better than the original concept. The amount of memory used by the basic plan under consideration and the suggested approach is calculated in this index. The memory consumption in the suggested way is 30,360 (when the number of storages is 30 and the length of the finger table is 10) and 32,640 for the basic design when the results are examined for the number of distinct data packets (20, 40, 60, 80, and 100). Kilobytes is the unit of measurement. Figure 14 presents the findings from this study.

Fig. 14.

Evaluating the efficiency of the proposed method and the basic plan in the memory consumption index in terms of kilobytes.

The efficacy of the suggested methodology was assessed for three metrics: connection latency, response time, and memory use. The results were then compared to those of the CoAPMicro base design under identical settings. Table 4 displays the performance comparison outcomes between the suggested design and the basic design.

Table 4.

Comparison of the performance of the proposed plan with the basic plan in different evaluation indices.

Index / method	Response time	Consumable memory	Delay between connections
CoAPMicro	0.026(s)	32,640(KB)	0.094(s)
proposed method	0.016(s)	30,360(KB)	0.042(s)

The results reveal that the suggested strategy outperforms the basic design across all evaluation indices. As a result, it may be concluded that it will be a dependable method of verifying authenticity in the Internet of Things.

Conclusion

This study presents a novel and combined method for two-factor authentication, where both the sender and receiver of the message may initiate the authentication process. Prior to establishing a connection and transmitting data, the authenticity of both the sender and recipient of the message is verified. In other words, the legality of the communication is verified on both ends. The use of chord architecture has also been employed to mitigate the computational complexity. The average reaction time using the suggested technique is 0.016 s, compared to 0.026 s for the basic design. The average memory use in the suggested plan is 30,360 KB, however the findings of this investigation for the basic plan indicate a figure of 32,640 KB. The suggested strategy has an average delay of 0.042 s between connections, whereas the basic plan has an average delay of 0.094 s. Upon analyzing the results, it was seen that the computational complexity was lowered, resulting in a decrease in both reaction time and connection latency.

Author contributions

All authors wrote the main manuscript text. All authors reviewed the manuscript.

Funding

This work was supported by Natural Science Foundation of Anhui Province (No.2008085MF215), Natural Science Research Project for Anhui Universities (No. KJ2018A0336, No. KJ2021A0682, No. J2021A0662, No. 2023AH050403, No. 2023AH050406), Anhui Province Quality Engineering Project(2023jyxm0528).

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Competing interests

The authors declare no competing interests.