Mitigating Smart Jammers in Multi-User MIMO

Wireless systems must be resilient to jamming attacks. Existing mitigation methods based on multi-antenna processing require knowledge of the jammer's transmit characteristics that may difficult acquire, especially for smart jammers evade by transmitting only at specific instants. We propose a novel method mitigate attacks massive multi-user multiple-input multiple-output (MU-MIMO) uplink which does not jammer active any instant. By formulating an optimization problem unifies estimation and mitigation, channel estimation, data detection, we exploit cannot change its subspace within coherence interval. Theoretical results our formulation show solution is guaranteed recover users' symbols under certain conditions. develop two efficient iterative algorithms approximately solving proposed formulation: MAED, parameter-free algorithm uses forward-backward splitting with box symbol prior, SO-MAED, replaces prior MAED soft-output estimates discrete constellation deep unfolding optimize parameters. use simulations demonstrate effectively wide range without priori about attack type.