Yuan Zhonghao | Searcharxiv

arXiv2026

WHTDM: Walsh-Hadamard Transform Division Multiplexing for Doubly-Selective Channels

We propose Walsh-Hadamard Transform Division Multiplexing (WHTDM), a multicarrier waveform that replaces the conventional IFFT/FFT pair in OFDM with a real-valued, unitary Walsh-Hadamard transform (WHT). WHTDM inherits the CP-OFDM transceiver structure while eliminating all complex multiplications from the transform stage, yielding a transmitter with zero real multipliers in the core modulation block. For detection under doubly-selective channels, we adopt a cross-domain memory approximate message passing (CD-MAMP) equalizer that operates on the banded structure of the equivalent WHT-domain channel matrix. Simulation results under the 3GPP TDL-C channel model at 28 GHz demonstrate that WHTDM with CD-MAMP significantly outperforms conventional OFDM 1-tap MMSE at high mobility, achieving over an order of magnitude lower BER at 120 km/h. Among the compared CD-MAMP-equalized new waveforms, WHTDM achieves the best BER performance while maintaining a transmitter complexity 2.5 $\times$ lower than OFDM and completely eliminating complex multipliers from the transform stage, making it well-suited for low-power IoT terminals.

eess.SP↗

arXiv2025

HIT-ROCKET: Hadamard-vector Inner-product Transformer for ROCKET

Time series classification holds broad application value in communications, information countermeasures, finance, and medicine. However, state-of-the-art (SOTA) methods-including HIVE-COTE, Proximity Forest, and TS-CHIEF-exhibit high computational complexity, coupled with lengthy parameter tuning and training cycles. In contrast, lightweight solutions like ROCKET (Random Convolutional Kernel Transform) offer greater efficiency but leave substantial room for improvement in kernel selection and computational overhead. To address these challenges, we propose a feature extraction approach based on Hadamard convolutional transform, utilizing column or row vectors of Hadamard matrices as convolution kernels with extended lengths of varying sizes. This enhancement maintains full compatibility with existing methods (e.g., ROCKET) while leveraging kernel orthogonality to boost computational efficiency, robustness, and adaptability. Comprehensive experiments on multi-domain datasets-focusing on the UCR time series dataset-demonstrate SOTA performance: F1-score improved by at least 5% vs. ROCKET, with 50% shorter training time than miniROCKET (fastest ROCKET variant) under identical hyperparameters, enabling deployment on ultra-low-power embedded devices. All code is available on GitHub.

cs.LG↗