TY - GEN
T1 - Semantic-Aware Co-Design of Spatio-Temporal Focusing and Adaptive Modulation for Wearable Brain-Body Neuro-Interfaces
AU - Zhang, Yitong
AU - Yan, Tianyi
AU - Wu, Chenxuan
AU - Pan, Hongzhen
AU - He, Zhaolan
AU - Xu, Hongfei
AU - Dou, Yongyi
N1 - Publisher Copyright:
© 2026 IEEE.
PY - 2026
Y1 - 2026
N2 - This paper presents a communication-theoretic, semantic-aware co-design for a wearable brain-body neuro-interface that addresses the dynamic somatic dysregulation in Major Depressive Disorder. At the physical layer, we derive an analytical spatio-temporal focusing model for temporal interference (TI) stimulation using a multilayer spherical head formulation. Closed-form envelope metrics enable efficient optimization of focusing gain and mainlobe-to-sidelobe ratio (MSR) under scalp-current safety constraints, yielding a library of high-precision stimulation actions. At the MAC layer, wearable biosignals are abstracted into semantic channel state information, and adaptive modulation is cast as a Markov decision process with energy/safety-aware rewards; the optimal policy is obtained by value iteration. Analytical and simulation results show that the proposed co-design sustains high MSR at deep targets with phase-locked steering, improves long-term therapeutic reward versus static or myopic baselines, and shortens the time to a desired "Calm"state, with multiplicative benefits over isolated layer designs.
AB - This paper presents a communication-theoretic, semantic-aware co-design for a wearable brain-body neuro-interface that addresses the dynamic somatic dysregulation in Major Depressive Disorder. At the physical layer, we derive an analytical spatio-temporal focusing model for temporal interference (TI) stimulation using a multilayer spherical head formulation. Closed-form envelope metrics enable efficient optimization of focusing gain and mainlobe-to-sidelobe ratio (MSR) under scalp-current safety constraints, yielding a library of high-precision stimulation actions. At the MAC layer, wearable biosignals are abstracted into semantic channel state information, and adaptive modulation is cast as a Markov decision process with energy/safety-aware rewards; the optimal policy is obtained by value iteration. Analytical and simulation results show that the proposed co-design sustains high MSR at deep targets with phase-locked steering, improves long-term therapeutic reward versus static or myopic baselines, and shortens the time to a desired "Calm"state, with multiplicative benefits over isolated layer designs.
KW - Markov decision process (MDP)
KW - Spatio-temporal focusing
KW - Temporal interference (TI) stimulation
KW - Wearable brain-body neuro-interface
UR - https://www.scopus.com/pages/publications/105045342019
U2 - 10.1109/ICC59461.2026.11587835
DO - 10.1109/ICC59461.2026.11587835
M3 - Conference contribution
AN - SCOPUS:105045342019
T3 - IEEE International Conference on Communications
BT - ICC 2026 - IEEE International Conference on Communications, Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2026 IEEE International Conference on Communications, ICC 2026
Y2 - 24 May 2026 through 28 May 2026
ER -