With the access of massive terminals of the Internet of Things (IoT), the low-power wide-area networks (LPWAN) applications represented by Long Range Radio (LoRa) will grow extensively in the future. The specific Long Range Wide Area Network (LoRaWAN) protocol within the LoRa network considers both low power consumption and long-range communication. It can optimize data transmission to achieve low communication latency, ensuring a responsive system and a favorable user experience. However, due to the limited resources in LoRa networks, if certain terminals have heavy traffic loads, it may result in unfair impacts on other terminals, leading to increased data transmission latency and disrupted operations for other terminals. Therefore, effectively optimizing resource allocation in LoRa networks has become a key issue in enhancing LoRa transmission performance. In this paper, a Mixed Integer Linear Programming (MILP) model is proposed to minimize network energy consumption under the maximization of user fairness as the optimization goal, which considers the constraints in the system to achieve adaptive resource allocation for spreading factor and transmission power. In addition, an efficient algorithm is proposed to solve this optimization problem by combining the Gurobi mathematical solver and heuristic genetic algorithm. The numerical results show that the proposed algorithm can significantly reduce the number of packet collisions, effectively minimize network energy consumption, as well as offering favorable fairness among terminals.

In the intelligent reflecting surface (IRS)-assisted MIMO systems, optimizing the passive beamforming of the IRS to maximize spectral efficiency is crucial. However, due to the unit-modulus constraint of the IRS, the design of an optimal passive beamforming solution becomes a challenging task. The feature input of existing schemes often neglects to exploit channel state information (CSI), and all input data are treated equally in the network, which cannot effectively pay attention to the key information and features in the input. Also, these schemes usually have high complexity and computational cost. To address these issues, an effective three-channel data input structure is utilized, and an attention mechanism-assisted unsupervised learning scheme is proposed on this basis, which can better exploit CSI. It can also better exploit CSI by increasing the weight of key information in the input data to enhance the expression and generalization ability of the network. The simulation results show that compared with the existing schemes, the proposed scheme can effectively improve the spectrum efficiency, reduce the computational complexity, and converge quickly.