E-SRResNet Super-Resolution Network Architecture

Paper context

Paper title: CKM Beyond Channel Gain: Spatial Correlation Map Construction with Deep Learning Abstract: Channel knowledge map (CKM) is a promising technique to achieve environment-aware wireless communication and sensing. Constructing the complete CKM based on channel knowledge observations at sparse locations is a fundamental problem for CKM-enabled wireless networks. However, most existing works on CKM construction only consider the special type of CKM, i.e., the channel gain map (CGM), which only records the channel gain value for each location. In this paper, we consider the channel spatial correlation map (SCM) construction, which signifies the location-specific spatial correlation matrix for multi-antenna systems. Unlike CGM construction, constructing SCM poses significant challenges due to its extremely high-dimensional structure. To address this issue, we first decompose the high-dimensional SCM into lower-dimensional path gain map (PGM) and path angle map (PAM). Then we propose a deep learning model termed E-SRResNet for constructing high-quality SCM from sparse samples, which incorporates multi-head attention (MHA) mechanisms and multi-scale feature fusion (MSFF) to accurately model both local and global spatial relationships of channel parameters and complex nonlinear mappings. Furthermore, we preprocess the dataset to provide priors including line-of-sight (LoS) map, binary building map and base station (BS) map for the model to reconstruct SCM more accurately. Simulations conducted on the CKMImageNet dataset demonstrate that the proposed E-SRResNet achieves signifi Passages referencing this figure: ct PGM and PAM for primary and secondary paths, and generate the line-of-sight (LoS) map, the binary building map, and the base station (BS) map to provide priors for the model. Numerical results demonstrate superior performance of our model over the baseline methods, validating its effectiveness in high-accurate sparse SCM completion. II System model II-A Channel Spatial Correlation Matrix Model Figure 1: Illustration of the scenario setup. Consider massive MIMO scenario, where the BS is equipped with N ≫ 1 N\gg 1 antennas, as shown in Fig. 1 . Let θ ℓ \theta_{\ell} and α ℓ \alpha_{\ell} denote the angle of arrival (AoA) and complex gain of the ℓ \ell -th path, respectively. The channel vector 𝐡 \mathbf{h} can be expressed as 𝐡 = ∑ ℓ = 0 L − 1 α ℓ ​ 𝐚 ​ ( θ ℓ ) = ∑ ℓ = 0 L − 1 | α ℓ | ​ 𝐚 p to provide priors for the model. Numerical results demonstrate superior performance of our model over the baseline methods, validating its effectiveness in high-accurate sparse SCM completion. II System model II-A Channel Spatial Correlation Matrix Model Figure 1: Illustration of the scenario setup. Consider massive MIMO scenario, where the BS is equipped with N ≫ 1 N\gg 1 antennas, as shown in Fig. 1 . Let θ ℓ \theta_{\ell} and α ℓ \alpha_{\ell} denote the angle of arrival (AoA) and complex gain of the ℓ \ell -th path, respectively. The channel vector 𝐡 \mathbf{h} can be expressed as 𝐡 = ∑ ℓ = 0 L − 1 α ℓ ​ 𝐚 ​ ( θ ℓ ) = ∑ ℓ = 0 L − 1 | α ℓ | ​ 𝐚 ​ ( θ ℓ ) ​ e j ​ ϕ ℓ , \mathbf{h}=\sum_{\ell=0}^{L-1}\alpha_{\ell}\mathbf{a}(\theta_{\ell})=\sum_{\ell=0}^{L-1}\left|\alpha_{\ell}\right|\mat