1: \begin{abstract}

2:     	%wireless federated learning, pca

3:   	A wireless federated learning system is investigated by allowing a server and workers to exchange uncoded information via orthogonal wireless channels.

4:   	Since the workers frequently upload local gradients to the server via bandwidth-limited channels, the uplink transmission from the workers to the server becomes a communication bottleneck.

5:   	Therefore, a one-shot distributed principle component analysis (PCA) is leveraged to reduce the dimension of uploaded gradients such that the communication bottleneck is relieved. 

6:   	A PCA-based wireless federated learning (PCA-WFL) algorithm and its accelerated version (i.e., PCA-AWFL) are proposed based on the low-dimensional gradients and the Nesterov's momentum. 

7:   	For the non-convex loss functions, a finite-time analysis is performed to quantify the impacts of system hyper-parameters on the convergence of the PCA-WFL and PCA-AWFL algorithms.

8:   	The PCA-AWFL algorithm is theoretically certified to converge faster than the PCA-WFL algorithm. 

9:   	Besides, the convergence rates of PCA-WFL and PCA-AWFL algorithms quantitatively reveal the linear speedup with respect to the number of workers over the vanilla gradient descent algorithm.

10:     Numerical results are used to demonstrate the improved convergence rates of the proposed PCA-WFL and PCA-AWFL algorithms over the benchmarks. 

11:     \end{abstract}