Abstract: To cope with spectrum resource shortage and channel link interference in integrated sensing and communication (ISAC) systems, we investigate unmanned aerial vehicle (UAV)-integrated sensing and communication based on non-orthogonal multiple access (NOMA). A dual-function UAV sends superimposed signals to multiple communication users according to the downlink non-orthogonal multiple access, and the superimposed signals can also be used for target perception simultaneously. To balance the performance of communication and perception, we propose a beamforming design based on a successive convex approximation that aims to maximize the weighted sum of effective sensing power and total bit rate under the constraint of minimum bit rate requirements. To deal with the fact that the optimal solution obtained using the semi-definite relaxation algorithm does not strictly satisfy the rank-one constraint, we further propose a reconstruction method based on an improved semi-definite relaxation algorithm that can reconstruct the beamforming design results to satisfy the maximum transmit power and the minimum communication rate while maintaining system performance. Simulation results show that the performance of the proposed ISAC is improved by 7.37% compared with the ISAC system of ground base stations based on orthogonal multiple access, and that of the proposed beamforming design is increased by 10.31% based on the conventional semi-definite relaxation algorithm.