Spectral imaging acquires more analyzable and distinguishable information than RGB imaging and has become an emerging technique powering Internet of Things (IoT). Thus, generating spectral images from existing RGB cameras is instrumental to high-level vision tasks. In this article, an IoT-oriented spectral reconstruction model with parallel fusion of convolutional neural network (CNN) and transformer (PFCT) is proposed to efficiently recover hyperspectral images (HSIs) from RGB counterparts, facilitating low-cost and nonhardware-specific spectral image acquisition. Recent works mainly utilize CNNs to extract local features by stacking more layers, ignoring the latent correlations of global features. In our PFCT network, we take advantage of lightweight CNNs to efficiently perceive local details, and exploit transformer blocks to fully capture the global context. Based on the architecture, a parallel fusion module is further designed to deeply interact and fuse the features obtained by CNN and Transformer in both directions. The final output spectral image is generated with same spatial sizes of the original image based on the deeply fused features. Consequently, the proposed PFCT network achieved high performance on four benchmark data sets compared to several state-of-the-art networks with a relatively small number of parameters.

U-UV codes were recently proposed as a competent short-to-medium length coding scheme. With well designed component codes, U-UV codes can outperform similar rate cyclic redundancy check (CRC)-polar codes with the successive cancellation (SC) and the SC list (SCL) decoding. In order to improve the coded transmission spectral efficiency, this paper proposes the bit-interleaved coded modulation (BICM) scheme with U-UV codes as the channel codes. A bit interleaver structure is proposed to facilitate the component code rate allocation based on the polarized subchannel capacities. Under the BICM paradigm, the component code rates can be allocated by first estimating the modulation subchannel capacities, then adjusting based on the finite length rates and the equal error probability rule. Theoretical performance bounds and their approximations on decoding error rates are further analyzed. It provides the theoretical benchmarks for our simulations and guides the optimized design of the coded modulation scheme. Finally, simulation results of the U-UV coded BICM scheme are provided to demonstrate its error-correction competency. It can outperform the relevant bit-interleaved polar coded modulation (BIPCM) scheme.