Algorithm 2 Joint Optimization Algorithm for Maximizing Energy Efficiency

1:Initialize variables ${\tau }_I^0$, $Q_U^0$, $P_I^0$, and $D^{\mathrm{loc}\left(0\right)}$. Set the minimum error value $ϵ>0$. Let the number of iterations $r=0$, update rate $\xi =1$, and update factor $\epsilon =0.1$. 2:Given $Q_U^r$ conditions, use the continuous convex optimization approximation algorithm to solve problem (P2), obtain optimal solutions ${\tau }_I^{*}$, $P_I^{*}$, $D^{\mathrm{loc}\left(*\right)}$, and let ${\tau }_I^{r+1}=\xi ({\tau }_I^{*}-{\tau }_I^r)+{\tau }_I^r$, $P_I^{r+1}=\xi (P_I^{*}-P_I^r)+P_I^r$, $D^{\mathrm{loc}\left(r+1\right)}=\xi (D_I^{*}-D_I^r)+D_I^r$. 3:Given ${\tau }_I^{r+1}$, $P_I^{r+1}$, and $D^{\mathrm{loc}\left(r+1\right)}$, Algorithm 1 is used to solve problem (P3), obtain the optimized relay UAV trajectory $Q_U^{*}$, and let $Q_U^{r+1}=\xi (Q_I^{*}-Q_I^r)+Q_I^r$. 4:Update the number of iterations $r=r+1$, $\xi =\xi /\left(1+\left(r-1\right)\times \epsilon \right)$. 5:Calculate the increment of the target value $\Delta$. If $\Delta <ϵ$, then the algorithm converges and ends. Otherwise, continue with steps 2–4. 6:Output the optimized solution ${\tau }_I^{*}$, $Q_U^{*}$, $P_I^{*}$, $D^{\mathrm{loc}\left(*\right)}$, and the maximum secret energy efficiency.