Unmanned aerial vehicle (UAV) is considered as an important component of 6G Internet of Things (IoT) because of its deployment flexibility. In this paper, we consider a aerial cooperative short packet communication system model in urban environments. The number of transmitted bits is maximized by jointly optimizing the UAV’s altitude, transmit power and blocklength allocation with constraints on system reliability, latency and fairness. The successive convex approximation (SCA) method is adopted to solve the difficulty that the optimization problem is non-convex. Specifically, we firstly decompose the optimization problem into three subproblems to be solved separately, and then obtain the solution of the original problem by the proposed optimization algorithm. Finally, the performance of the proposed algorithm is verified by simulation results. In addition, we investigate the impact of different communication environments and the number of IoT devices on the system performance.