This paper proposes a novel interference-resistant receiver for wideband long-term evolution (LTE) system which employs orthogonal frequency division multiple access (OFDMA) for the downlink. In case the received spectrum is underutilized, compressed sensing (CS) enables us to operate at a sub-Nyquist sampling rate for lower device cost, higher data transmission rate and faster spectrum sensing speed. To improve the receiver performance under a mutually interfering and noisy environments, a resource block (RB)-detection based iterative interference cancellation algorithm for LTE OFDMA signals is presented.

Active RBs are blindly detected in order to improve performance of the proposed CS-based receiver. Specifically, the RBs of active users are estimated by minimum mean square error (MMSE) and those of interference users are estimated via ℓ₁– norm minimization in a semi-blind manner, both in compressive signal domain. By canceling the interferences directly from compressive measurements iteratively, the proposed algorithm improves the estimates of both the desired and interfering signals. Simulation results indicate that such an interference-resistant receiver yields significant gains in bit error rate at low sampling rates.