This paper proposes minimum mean square error (MMSE)-based amplify-and-forward (AF) relay amplifying matrices under perfectly and imperfectly known channel state information by imposing a constraint on the transmit power of the relays. The main objective of this paper is to determine optimal relay amplifying matrices with and without channel uncertainty under the transmit power constraint at the relays.
Additionally, by using the derived optimal relay amplifying matrices, it is proven that the better BER performance is observed when power is constrained at the relays during data transmission, compared to the no-power constraint case in [6].Finally, simulation results show that the MMSE cost function values with and without channel uncertainty are always less than the number of sources and destinations (M), regardless of the number of relays (N).