Signature codes are widely used for user identification (UI) and channel estimation (CE) in wireless networks due to their high spectral efficiency. To design an effective signature matrix exploiting prior information, we propose an end-to-end machine-learning-aided signature code (ML-SC) system. The ML-SC system comprises an encoder based on binarized neural networks (BNNs) and a decoder based on a modified trainable iterative soft-thresholding algorithm (TISTA). The BNNs efficiently handle the trainable discrete signature matrix, while the modified TISTA enables support for a large number of users with enhanced performance under Rayleigh fading channels. Our simulation results demonstrate that the proposed ML-SC system maintains scalability across different matrix dimensions. With this scalability advantage, the ML-SC achieves consistent performance improvements. The signature matrix generated by the ML-SC system, which we refer to as the ML-signature matrix, yields superior decoding performance compared to randomly generated and deterministic binary matrices, demonstrating an effective SNR gain of approximately 2.5–5 dB compared to conventional approaches. We also verify that the proposed ML-signature matrix demonstrates strong compatibility with various prevalent decoding methods. Furthermore, we confirm significant enhancement of the restricted isometry constant (RIC) for the ML-signature matrix, which provides theoretical support for the observed performance improvements. Copyright © 2026 The Institute of Electronics, Information and Communication Engineers.

High-density LED arrays enable high-speed transmission in image sensor-based visible light communication (VLC) systems. However, when optical spots become blurred and spatially overlapped due to focal shift, resolution limitations, or interference, severe inter-symbol interference (ISI) occurs, significantly degrading decoding performance. Existing methods mitigate ISI by reducing LED transmission signaling density.This paper proposes a robust decoding framework that maintains full LED signaling density. We introduce a pilot-aided geo-metric recognition with a PSF-constrained Hough transform and circle center alignment refinement. By leveraging prior structural knowledge from pilot frames, the system effectively separates overlapping LED signals under severe optical distortion.Experimental results on a real-world VLC testbed confirm that the proposed method achieves superior decoding accuracy and throughput compared to conventional Hough-based and low-density baseline methods. The results highlight its potential for high-efficiency VLC applications in interference-prone environments. © 2025 IEICE.