Federated Learning Over Wireless Networks Challenges and Solutions

Limited energy resources remain a challenge for wireless devices in federated learning (FL) networks. To tackle this issue, a joint resource allocation problem is formulated to minimize the total energy consumption for wireless-powered FL networks with non-orthogonal multiple access. A two-dimensional search algorithm (TDSA) is first proposed to obtain the optimal solution for this challenging problem. To lower the computational complexity, the bisection algorithm (BA) is further employed by exploiting the inherent structure of the formulated problem. Numerical results verify the equal performance between the BA and TDSA, and the superiority of the proposed scheme over the benchmark schemes in energy minimization.