1: \begin{abstract}

2: In this paper, the adoption of an intelligent reflecting surface (IRS) for multiple user pairs in two-hop networks is investigated.

3: Different from the existing studies on IRS that merely focused on tuning the reflection coefficients of all elements, we consider the \textit{true} reflection resource  management (RRM) which builds on the premise of the introduced modular IRS structure,  which is composed of multiple independent and controllable modules.

4: In particular, the \textit{true} RRM can be realized via the best triggered module subset identification.

5: To provide fairness among users, our goal, therefore, is to maximize the minimum signal-to-interference-plus-noise ratio (SINR) at destination terminals (DTs) via joint triggered module subset identification, transmit power allocation, and the corresponding passive beamforming, subject to per source terminal (ST) power budgets and module size constraint.

6: Whereas this problem is NP-hard due to the module size constraint,  to deal with it, we transform it into the group sparse constraint by introducing the mixed $\ell_{1,F}\text{-norm},$ which yields a suitable semidefinite relaxation.

7: In order to address this approximated problem, a two-block alternating direction method of multipliers (ADMM) algorithm is proposed based on its separable structure.

8: Numerical simulations are used to validate the analysis and assess the performance of the proposed algorithm as a function of the system parameters.

9: In addition,  energy efficiency (EE) performance comparisons demonstrate the necessity and meaningfulness of the introduced modular IRS structure.

10: Specifically, for a given network setting, there is an optimal value of the number of triggered modules, when the EE is considered.

11: 

12: \end{abstract}