Symbol Detection of Ambient Backscatter Communications Under IQ Imbalance

Ambient backscatter communications (AmBC) have been mainly studied under the assumption of perfect inphase/quadrature-phase (IQ) balance, which is hard to achieve in practical AmBC networks. In this paper, we study the symbol detection in AmBC under IQ imbalances. In particular, considering a differential encoding scheme at the AmBC transmitter and an energy difference based detector at the AmBC receiver, we derive the bit error rate (BER) and the near-optimal detection threshold in closed forms. Since the near-optimal detection threshold requires a priori knowledge of the channel state information, the IQ imbalance parameters, the noise power and the transmit power of the ambient source, which is not readily known by the AmBC receiver, we propose a simple but practical method to estimate the near-optimal detection threshold by exploiting the received signal samples at the AmBC receiver. Simulation results validate the accuracy of the derived BER and detection threshold, and reveal that the studied detector is more robust to the amplitude imbalance than to the phase imbalance.