Abstract: In order to improve the robustness and energy efficiency of non-orthogonal multiple access (NOMA) enhanced device to device (D2D) group links, a robust resource allocation algorithm with energy efficiency optimization is proposed considering the imperfect channel station information (CSI). Firstly, the maximum and minimum robust energy efficiency model is established under the constraints of guaranteed subchannel allocation, minimum rate of cellular users and D2D group, and maximum transmission power of D2D group. Secondly, the channel uncertainty is modeled as a bounded channel estimation error by considering the worst case method, and the original multivariable coupling problem is transformed into a convex optimization problem by using Taylor series expansion, convex relaxation and variable transformation methods. Finally, it use Lagrangian duality theory to solve it. Simulation results show that the proposed algorithm controls the transmission rate above the minimum rate threshold, has good robustness, and improves energy efficiency by 8.3% compared to other algorithms.