Internal Attack. (1) Participants’ Attack In the semi-quantum key agreement protocol (SQKA) proposed in this paper, the quantum cloud server Xxxxx participates as a quantum producer. Xxxxx prepares entangled particles and distributes them to everyone. The quantum center Xxxxxxx participates in entanglement detection and key measurement agreement stage. Compared with Xxx, the internal eavesdropper Xxxxx and Xxxxxxx have more advantage in acquiring information about entangled particles. Assuming that Xxxxxxx is a dishonest quantum center, he can access some quantum resources from the beginning and try to obtain the session key KAB of Xxxxx and Xxx. In the key agreement stage, although Xxxxxxx knows the measurement results of his part of the particles and all possible results in Table 1, he cannot determine the measurement results of Xxxxx and Xxx. For example, the measurement results of Xxxxxxx, Xxxxx and Xxx are 011 respectively. At this time, according to Table 1, the measurement values of Xxxxx and Xxx should be the same. In order to obtain the correct key, Xxxxxxx can only select one randomly from {0,1} as the key, so that for each bit of the agreement key, the probability of Xxxxxxx getting the correct key is only 50%. For the N bits of the agreement key, the probability of Xxxxxxx getting the complete key is 1 𝑁
Internal Attack. In the QKA protocols, a collusive attack is the most powerful internal attack in which two or more dishonest participants collude together to extract sensitive information or generate the final key alone without revealing their malicious behavior. In this subsection, we show that the proposed model is immune to collusive attacks, such that any group of dishonest participants trying to perform a collusive attack (including the two attack strategies mentioned in section 2) will be detected immediately. Indeed, dishonest participants rely mainly on two important processes to successfully achieve the collusive attack; 1) sharing information about the carrier quantum states that will be used to encode the private data and generate the final key, 2) deceiving the honest participants to deduce their private data by sending forged data. Therefore, to prevent the collusive attack, dishonest participants should be prevented from conducting these two processes. In our protocol, a semi-honest server is used, as indicated in Step (1), to generate the initial quantum states () that will be used to encode the private inputs of the participants. The server shares () with all participants after they receive the encoded data. In that case, all participants use the shared information to deduce the final key fairly. Also, the server checks the security of the quantum channel between every two participants and makes sure that the receiver has received genuine quantum states. Using these two processes, the protocol guarantees that the honest participant has received genuine data, and the dishonest participants cannot obtain useful information to generate the final key alone or steal the private inputs of honest participants.
