With the rapid boost of wireless networks, it is challenging to guarantee the secure transmission and massive connectivity in the process of data collection. This letter considers an unmanned aerial vehicle (UAV)-aided non-orthogonal multiple access (NOMA) communication network, where a UAV is deployed to collect the data from sensing devices (SDs) in the NOMA manner subject to the eavesdropping attack, and a group of auxiliary devices (ADs) are deployed to provide the cooperative jamming to the eavesdropper. We aim to maximize the total secrecy throughput by jointly optimizing SDs' transmit power and ADs' scheduling decisions. We first formulate this problem as a mixed-integer non-convex optimization problem, which turns out to be mathematically intractable due to the non-convexity. Then, an alternative optimization algorithm is developed to find a suboptimal solution. Specifically, the successive convex approximation approach and the cross-entropy technique are used to solve the two sub-problems decomposed from the optimization problem, respectively. Numerical results demonstrate that the proposed algorithm can achieve considerable performance compared with the benchmark schemes.