Intelligent Reflecting Surface Enabled Sensing: Cramér-Rao Bound Optimization

This paper investigates intelligent reflecting surface (IRS) enabled non-line-of-sight (NLoS) wireless sensing, in which an IRS is dedicatedly deployed to assist access point (AP) sense a target at its NLoS region. It assumed that the AP equipped with multiple antennas and uniform linear array. We consider two types of models, namely extended targets, for aims estimate target's direction-of-arrival (DoA) response matrix respect IRS, respectively, based on echo signals from AP-IRS-target-IRS-AP link. Under this setup, we jointly design transmit beamforming reflective minimize Cram\'er-Rao bound (CRB) estimation error. Towards end, first obtain CRB expressions models closed form. shown case, estimating DoA depends both beamformers; while only beamformers. Next, optimize joint CRB, via alternating optimization, semi-definite relaxation, successive convex approximation. For optimal solution Finally, numerical results show cases, proposed designs minimization achieve improved sensing performance terms mean squared error, as compared other traditional schemes.