To meet the stringent requirements of latency, reliability and energy efficiency (EE), we introduce rate splitting multiple access (RSMA) into ultra-reliable and low-latency communication (URLLC) IIoT networks, where RSMA has a good flexibility in interference management. By jointly optimizing beamforming design and common rate allocation, we investigate the worst-case EE maximization problem with imperfect channel state information (CSI). Although the problem is nonconvex, we exploit the monotonicity of the problem to develop an optimal solution, where a monotonic optimization framework based on polyblock outer approximation (PA) and boundary searching is proposed to find the optimal points on the Pareto boundary. Simulation results show the convergence of the proposed optimal algorithm, and RSMA can achieve higher EE than space division multiple access (SDMA) and non-orthogonal multiple access (NOMA) in URLLC IIoT scenarios.

In this paper, we investigate the cross-layer design in hybrid reconfigurable intelligent surface (RIS)-assisted cell-free systems. To maximize long-term energy efficiency while ensuring network stability, we formulate a queue-aware stochastic problem that jointly optimizes the beamforming matrices at the remote antenna units (RAUs) and hybrid RIS, along with the operation modes of RIS elements. By utilizing the Lyapunov theory, the formulated stochastic optimization problem is decomposed into a sequence of slot-level problems that maximize energy efficiency while minimizing the Lyapunov drift. To tackle the formulated non-convex slot-level problem, we propose an alternating optimization algorithm to transform the slot-level problem into transmission and element mode switching subproblems. In particular, for efficient RIS element selection, we first derive the number upper bound of the elements that can be activated given the RIS power budget, and then the element selection optimization is transformed into multiple tractable smaller problems that can be solved in parallel by utilizing a greedy algorithm. Simulation results demonstrate that the proposed algorithm outperforms the baseline schemes in terms of energy efficiency and can effectively improve system stability as well as user fairness.

The Industrial Internet of Things (IIoT) is envisioned as the future paradigm for the next-generation industrial system. To provide low-latency and high-reliable communication for a large number of battery-limited industrial equipments, this paper proposes a cell-free network-assisted full-duplex (NAFD)-enabled short packet communications (SPC) scheme in IIoT networks, where access points (APs) operating in the uplink (UL) mode or downlink (DL) mode simultaneously serve UL sensors and DL actuators on the same frequency resource. Considering the UL-to-DL error propagation as well as the dynamic between latency, reliability and energy efficiency (EE), we take the effective EE as the performance metric and investigate the effective EE maximization problem by jointly optimizing UL power, DL beamforming matrices, data rates and duplex mode selection. The formulated problem is a mixed-integer fractional programming and decomposed into three subproblems. We develop the optimal and suboptimal algorithms for joint power control and beamforming design. A semi-closed-form solution is derived for the optimal data rates, and an efficient duplex selection algorithm is also proposed to find the computationally optimal duplex mode. Simulation results demonstrate the convergence of proposed algorithms, and the flexible cell-free NAFD scheme can achieve better EE performance than benchmark schemes such co-time co-frequency full duplex (CCFD), frequency division duplex (FDD) and greedy algorithm (GA) schemes.