In this paper, we formulate Subcarrier Assignment and Discrete Power Allocation for multi-UAV millimeter-wave cooperative Orthogonal Frequency Division Multiple Access (OFDMA) networks as a joint optimization problem considering the heterogeneous user data rate quality-of-service (QoS) requirements. The formulated joint optimization problem, named Discrete Power and Subcarrier Allocation (DPSA), is a nonconvex and mixed-integer nonlinear programming (MINP) problem, making it NP-hard to solve. We then transform the DPSA into the Subcarrier and Power Resource Efficient Cooperative Potential Game (SRECPG) based on game theory which facilitates distributed execution. We analyze the conditions under which a Nash Equilibrium (NE) exists in the SRECPG and provide rigorous proof of its existence. Furthermore, to enhance computational efficiency, we propose a BR-SSO algorithm based on better response dynamics. BR-SSO dramatically reduces the computational burden compared to the best response dynamics based on local exhaustive search (BRLES) while still ensuring convergence to a NE. Through extensive simulations, we demonstrated the effectiveness of the proposed SRECPG and BR-SSO algorithms. The results show significant improvements in throughput, fairness, and QoS guarantees compared to the baseline schemes. Our approach offers valuable insights into the design of efficient resource allocation schemes for multi-UAV millimeter-wave networks with varying QoS demands.INDEX TERMS Constrained potential game, downlink multi-cell OFDMA, Nash Equilibrium, unmanned aerial vehicle, blockage, millimeter-wave, resource allocation