Double Intelligent Reflecting Surface for Secure Transmission With Inter-Surface Signal Reflection

In this correspondence, we use double intelligent reflecting surface (IRS) assisted design to enhance the secrecy performance of wireless transmission. Different from the existing studies in the literature, the inter-surface signal reflection is considered in this work, which makes the formulated secrecy rate (SR) optimization problem difficult to tackle. In particular, to solve this problem, we propose a product Riemmanian manifold (PRM) based alternating optimization (AO) algorithm to jointly optimize the beamformer at transmitter as well as phase shift coefficients at double IRS, in which the PRM optimization algorithm is applied to optimize the phase shift coefficients at both IRSs simultaneously. Simulation results show that the proposed algorithm greatly enhance the SR compared with the existing benchmark schemes. And compared with conventional semi-definite relaxation based AO algorithm, the proposed PRM based AO algorithm can achieve much lower computational complexity as well as faster speed of convergence.