基于GAN的导频配置和信道估计联合优化算法

GAN-based joint pilot configuration and channel estimationoptimization method

随着通信设备的爆炸式增长,信道环境变得愈加复杂,传统信道估计方法需要进一步增加导频开销以维持现有信道估计精度。然而,这会导致系统吞吐量的下降。首先,提出了一种基于GAN的信道估计方法以在OFDM通信系统中解决这一问题。然后,采用GAN模型去学习从低维的潜向量到真实信道样本的映射关系。最后,在此基础上进行联合的导频配置优化。

With the explosive growth of communication devices,channel environments become increasingly complicated,which increases the pilot overhead of conventional channel estimation methods to maintain the estimation accuracy.However,the added extra overhead will lead to a decrease of system throughput.To solve this problem in the OFDM system,a channel estimation method based on the generative adversarial network(GAN)is first proposed.Then,the GAN model is used to learn the mapping relationship between the low-dimensional latent vector and the real channel samples.Finally,a joint pilot configuration and channel estimation optimization problem is considered to be solved.