Multi-agent learning in networks supported by RIS and multi-UAVs
2024
In this chapter, we illustrate a method for maximising the energy efficiency (EE) of a RIS-assisted UAVs-based network by jointly optimising the UAVs' power allocation and the RIS's phase-shift matrix through a DRL approach. Moreover, the parallel learning approach is also proposed as an effective possible solution for reducing the information transmission requirement of the centralised approach. Numerical results show a significant improvement in our proposed schemes compared with the conventional approaches in terms of EE, flexibility, and full suitability for real-time applications. * This chapter has been published partly in [1].