Distance bounding protocols have been suggested to protect an RFID communication system against relay attack. In such protocols, a verifier establishes an upper bound for the physical distance between a user and itself as well as authenticating the user. Recently, mutual distance bounding protocols in which both communication parties play not only as a user but also as a verifier have been proposed to be deployed for secure positioning protocols. In this paper, the performance of the existing mutual distance bounding protocols is analysed over a noisy communication channel. The analytic results show that these protocols provide a much higher rejection probability of a valid user due to channel errors than distance bounding protocols. Furthermore, a novel method is proposed to convert every distance bounding protocol into a mutual distance bounding protocol. Within this method, for each one of the two communication parties, the produced mutual distance bounding protocol provides the performance and security level exactly equal to those of its own associated distance bounding protocol at the same noisy environment, without transmitting any extra messages at the end of the protocol.