Security loopholes in the controlled quantum dialogue robust against conspiring users protocol

摘要

Recently, a new controlled quantum dialogue (CQD) protocol which is secure against the conspiring attack from the users without the controller's permission was put forward. However, it is found that this CQD protocol is weak against the different initial state (DIS) attack and the denial-of-service (DoS) attack from the dishonest controller. For the DIS attack, the controller prepares different initial states to eavesdrop on the users' messages. For the DoS attack, the controller deliberately announces the wrong classical information which is used for the users to encode secret messages, and thus, one user gets the wrong message from the other user. To mend the DIS attack, a security check to the controller is added to prevent the dishonest controller from preparing the different initial states. To mend the DoS attack, a message authentication is added to prevent the dishonest controller from publishing the wrong classical information. It shows that the security check to the controller and the message authentication can effectively prevent the DIS attack and the DoS attack from the dishonest controller, respectively.