Joint Source-Channel Coding for Multi-Channel Vibrotactile Sensors

Abstract

Wireless transmission of vibrotactile information holds great promise for immersive Extended Reality (XR) applications, particularly in enabling fine-grained remote manipulation through multi-channel vibrotactile signals. Existing vibrotactile coding solutions rely on conventional source and channel coding techniques to transmit high-quality vibrotactile data over band-limited and error-prone wireless channels. However, these methods often suffer from severe quality degradation due to bit errors and the unrecoverable source coding distortion. This study presents a novel vibrotactile communication scheme based on joint source-channel coding (JSCC) principles. The key advancement of this approach is the replacement of traditional source and channel coding with a joint source-channel vibrotactile coder and transceiver. The proposed coder compresses vibrotactile data using a two-dimensional discrete cosine transform (2D-DCT) without binarization, and the transmitter maps the DCT coefficients directly to transmission symbols. Optimized power allocation is used to avoid the catastrophic quality degradation and to maintain high vibrotactile quality under various wireless channel conditions. Experimental results using up to 30 vibrotactile sensors and 13 gestures show that the proposed scheme consistently outperforms state-of-the-art vibrotactile coding schemes, delivering high-quality vibrotactile signals over a range of wireless channels, from low to high channel signal-to-noise ratios (SNRs). In addition, the proposed scheme achieved approximately 80 % traffic reduction compared to the existing schemes under the same vibrotactile signal quality.

Keywords

• Multi-channel vibrotactile signal
• joint source-channel coding (JSCC)