Comment on "Quantum key agreement protocol"

The first two party Quantum Key Agreement (QKA) protocol, based on quantum teleportation, was proposed by Zhou et al. (Electronics Letters 40.18 (2004): 1149-1150). In this protocol, to obtain the key bit string, one of the parties use a device to obtain inner product of two quantum states, one being unknown, and the other one performs Bell measurement. However, in this article, we show that it is not possible to obtain a device that would output the inner product of two qubits even when only one of the qubit is unknown. This is so because existence of such device would imply perfectly distinguishing among four different states in a two-dimensional vector space. This is not permissible in quantum mechanics. Furthermore, we argue that existence of such a device would also imply violation of the "No Signalling Theorem" as well. Finally, we also comment that this protocol is not a valid key agreement protocol at all.