Table 3.

Informal Security Analysis.

Serial Number	Propositions	Attack Description	Security Defense Explanation
P1	The proposed framework is resilient against impersonation attacks.	The attacker may undertake a vehicle impersonation attack by intercepting the login message and obtaining the secret values from the vehicle’s smart card in an unlawful manner to listen in on, intercept, and change any message in the public communication channel.	Assume the attacker intercepts the message and attempts to construct another acceptable message that the network will validate. It is never going to happen because each and every message is recorded in the ledger over the blockchain maintained among all the RSUs and controller nodes in the network. So, any unlawful activity may not get a positive response from the rest of the network. Furthermore, guessing all of the unknown restrictions in polynomial time is impossible. As a result, the attacker will not be able to construct or guess further legitimate messages in polynomial time. Further, it is also not feasible to impersonate the vehicles by altering the message. As a legitimate message cannot be computed by the attacker unless he or she has access to all of the parameters necessary to calculate it, including the nonce of the sink node.
P2	The proposed framework is secured against offline password guessing attacks.	A password guessing attack is one in which an attacker attempts to impersonate a user by repeatedly guessing his password or other login information. Password guessing attacks may be carried out online by connecting to a server and trying to guess a user’s password. There are no limits on how many times an adversary may try to login in this version of the attack, unlike the offline version, which does. In the offline variant, an adversary obtains a user’s password-related data (e.g., a hashed password) and then repeatedly attempts to guess a password while comparing the hashed version to the intercepted one.	If the attacker wants to launch an offline password guessing attack on the vehicle, the attacker must first get the stored settings from the smart card. Two scenarios have been presumptively considered. One in which the attacker has stolen the vehicle’s smart card, and the other in which it is presumed that most users utilize low entropy IDs and passwords for memorizing purposes, which can be readily guessed in polynomial time. If an attacker manages to get its hands on the vehicle’s secret information, the attacker will still be unable to guess the proper password simultaneously in polynomial time. This is because no one can know a vehicle’s true identify except the vehicle itself through its smart card or messages communicated over the public channel since our approach uses blockchain to establish identity protection and passwords are safeguarded by a non-invertible one-way hash function. Hence, the attacker is unable to get access to the user’s personal information.
P3	Replay attacks can be protected by the proposed framework.	As the name suggests, this kind of attack involves an attacker intercepting and then fraudulently delaying or resending an already intercepted secure network communication. It is common for an attacker to re-transmit previously delivered communications in order to verify that a certain message was sent by the intended sender, hoping that this time the recipient would make a mistake and do what the attacker wants.	It is possible that a malicious actor may attempt to replay previous messages. Every time a message is sent using the proposed framework, it generates a new random number. Assume another scenario in which an antagonist has obtained one of the CRPs for a PUF and is attempting to reuse a prior challenge. This is why a CRP should never be reused. By allowing PUFs to be reprogrammed after each CRP, the system will be impervious to replay attacks. As a result, replay attacks are not a concern for the proposed methodology under development.
P4	The proposed framework is guarded from linking attacks.	In such kinds of attacks, to discover the real world identification of an incognito node, an adversary, who might be a controller node or cloud storage node, connects various data in the cloud or blockchain transactions with the same ID.	Each transaction in layer 3 is assigned a unique public key by the overlay nodes. Separate cloud accounts are used to verify the identity of each connected device. This makes it impossible for an attacker to connect data from various devices belonging to the same user.
P5	The proposed framework can protect against man-in-the-middle attacks.	An adversary node breaches the communication between two nodes in the network and obtains or compromises the information they communicate in a man-in-the-middle attack. An adversary may capture a ledger wallet by installing a malicious program on a target node and changing the address of the destination of blockchain transactions.	The information on all of the public constraints and messages’ exchange in sessions, as well as how to communicate with other roles of model, has been revealed to the adversary. Its objective is to identify all known vulnerabilities such as interception and replay the traffic, decrypt the secret keys, reveal the data of protected sessions, and threaten the legitimacy of entities. By implementing the framework, these back ends check of an attack can be detected. There are certain models that report safe, but others describe how the harm may be done and whose security objective cannot be fulfilled if the model is not secure enough. As proposed by Theorem 1, the proposed framework is protected against passive and active attacks, such as replay and man-in-the-middle attacks.Let us consider at any time a PUF gets its hands on one of the CRPs, an adversary can attempt to repeat an earlier challenge. As a result, it is preferred that a CRP be never reused. After each CRP, PUFs may be reconfigured, making the system resistant to man-in-the-middle attacks.
P6	Attacks such as Denial ofService(DOS) or Distributed Denial ofService(DDOS) can not affect the network security guarded with the proposed framework.	Attacks such as Distributed Denial of Service (DdoS) or Denial of Service (DOS) are meant to take advantage of weak spots in a system. This is performed by flooding a program with more requests than it can process, exceeding its network card’s capacity. Blockchain servers are overloaded with queries, which cause them to lose connectivity to other apps.	Before being sent to the global blockchain, all transactions in a local blockchain are authorized by the RSU or controller nodes. The global blockchain transactions are invalid on the local blockchain because the local blockchain and the global blockchain use different encryption algorithms. In the region cluster, a vehicle can only connect with another vehicle if the RSU has created a shared key between them. To avoid sending transactions to other vehicles in the cluster, RSUs search for matches in their key-list before sending any. Depending on the RSU’s capacity, the maximum number of transactions it may accept is limited. As soon as the limit is reached, vehicles are prevented from transmitting transactions to the destination vehicle.
P7	The proposed framework can defend well against the 51% attack.	An attacker must gain control of all the mining power on a specific blockchain in order to commit a 51% attack. With a mining advantage of more than 50% and the ability to mine faster than everyone else, the adversaries are doing quite well.	Based on the consensus method, the attack may be handled during the validation of neighbors or by other layer 2 nodes entirely via mutual authentication.
P8	Public BC modification attack is not possible in the proposed framework.	False blocks are advertised as the longest ledger by the attacker. As a result, every node acknowledges the ledger maintained by the attacker as the authentic ledger.	Layer 2 nodes, such as RSU and controllers, are limited in the number of blocks they can create in a given period of time. This restricts the number of malicious blocks that a layer 2 node may add, preventing the attacker from producing a ledger that is longer than the genuine ledger.
P9	Physical and cloning attacks are absolutely not possible with the proposed framework.	The investigation of information systems in order to uncover the concealed features of devices and systems by making use of the attributes of their implementation is known as cryptanalysis. One sort of cryptanalysis is known as physical attacks.	A vehicle or an IoV device might be cloned by an adversary to seem legitimate. An adversary may clone a device if it is physically compromised and the secrets from the seized device are extracted. The employment of PUFs, on the other hand, makes it exceedingly difficult for an opponent to conduct such attacks. In order to reliably assess PUF delays, a cloning attack on PUFs would require the deployment of intrusive procedures, which are not economically practical. To protect IoT devices with PUFs against physical and cloning threats, PUFs have been proven in [40,41] to be effective.
P10	The proposed framework is immune to side-channel attacks.	A security vulnerability known as a “side-channel attack” aims to harvest information from a chip or a system via an open channel. Various physical characteristics may be measured or analyzed to do this. Side-channel attacks are made possible by the ease with which an attacker may get access to IoT devices. Timing, power monitoring, electromagnetic attacks, and differential fault analysis are prominent examples of attacks in this category.	Statistical measurement of the time needed by a CPU to complete cryptographic operations is often used in timing attacks to discover the secret key. PUFs, on the other hand, employ a challenge response model instead of secret keys, making it more difficult to correctly measure the timing delays of a circuit in an IC. In addition, PUFs are deemed isochronous, which makes them immune to timing attacks. Attacks that rely on power consumption monitoring during calculations are known as power monitoring attacks. A data analysis method has been used by [42] to demonstrate a power side-channel attack against PUFs. They demonstrated that the number of zeros and ones stored in the latches of an arbiter PUF may be determined by utilizing power consumption information. PUFs, on the other hand, may be made safe against these attacks if the amount of zeroes and ones in the latches remains consistent. Electromagnetic attacks are a lot more difficult to carry out than power-monitoring attacks. As with power analysis attacks, the PUF may be protected against electromagnetic attacks by decreasing current fluctuations. Differential fault analysis is performed by exposing security hardware to aberrant environmental circumstances in order to introduce defects within it. Physical data corruption in cryptographic systems is often exploited by these approaches. Because of their sensitivity to temperature and voltage fluctuations, certain PUF types (such as delay-based PUFs) may be exploited by an adversary, although the physical data contained inside these PUFs cannot be exploited.