TACM-MR: Topographically-Augmented Channel Model Multi-Receiver Dataset

Electromagnetic spectrum congestion requires new methods to recognize and understand other transmitters for effective sharing. Existing Automatic Modulation Recognition (AMR) research utilizes datasets for single channel (SISO) or collocated systems (MIMO) with simplistic propagation models, creating the need for a novel dataset with distributed multi-receiver scenarios and more realistic channels.We propose the Topographically Augmented Channel Model 2025.1 (TACM2025.1), a spatially-separated but contemporaneous SISO multi-receiver dataset using propagation channels derived from simulated topographies. Multi-receiver classification significantly improves performance, with eight receivers being \\textbf{2x} more accurate at 0dB \\ac{ASNR} and \\textbf{2.4x} -10dB \\ac{ASNR} than a single receiver using the MCLDNN classifier. Additional tests confirm improvements are due to receiver aggregation rather than increased data volume. We show that ML classification models aggregate information from multiple lower quality receivers to provide improved accuracy and range over a single receiver classification with minor impact on inference time.