1: \begin{abstract}

2: While millimeter-wave (mmWave) communications can enjoy abundant bandwidth resource, their high susceptibility to blockage poses serious challenges to low-latency services. In this paper, a novel intelligent reflecting surface (IRS)-assisted mmWave scheme is proposed to overcome the impact of blockage. The scheme minimizes the user power of a multi-user mmWave system by jointly optimizing the transmit powers of the devices, the multi-user detector at the base station, and the passive beamforming at the IRS, subject to delay requirements. An alternating optimization framework is developed to decompose the joint optimization problem into three subproblems iteratively optimized till convergence. In particular, closed-form expressions are devised for the update of the powers and multi-user detector. The IRS configuration is formulated as a sum-of-inverse minimization (SIMin) fractional programming problem and solved by exploiting the alternating direction method of multipliers (ADMM). The configuration is also interpreted as a latency residual maximization problem, and solved efficiently by designing a new complex circle manifold optimization (CCMO) method. Numerical results corroborate the effectiveness of our scheme in terms of power saving, as compared with a semidefinite relaxation-based alternative.

3: \end{abstract}