A Rough Surface-Aided Non-Line-of-Sight Joint Sensing and Communication System for THz Band

Non-line-of-sight (NLoS) joint sensing and communication (JSAC) are essential in the terahertz (THz) band due to the susceptibility of high frequencies to blockages and severe pathloss, which greatly restrict the range of wireless links. To mitigate pathloss, large antenna arrays with ultra-narrow beams are commonly employed; however, their limited angular coverage poses challenges in NLoS scenarios. This work introduces a joint approach leveraging massive multiple-input multiple-output (mMIMO) systems and rough surface (RS) scattering to enable NLoS JSAC in the THz band. A geometry-based stochastic model (GBSM) is employed to characterize specular and diffuse scattering from the RS which is modeled using Kirchhoff approximation (KA), whereas Beckmann-Kirchhoff (BK) scattering theory is applied to calculate power per ray. A novel algorithm is proposed to identify the angles of target-associated diffusely scattered rays using high-resolution direction-of-arrival (DoA) estimation techniques. Target parameters such as range and velocity are estimated by analyzing the additional delay imparted by the target on the diffusely scattered rays from the RS. Simulation results illustrate the impact of surface roughness, antenna array size, and specular point misalignment on target detection, while communication performance is assessed by analyzing the effects of surface roughness and antenna array size on channel capacity, highlighting significant performance gains for intermediate/medium RS.