A Sleep Management System Based on Automatic Gain Control and Snore-Driven Intervention
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Snoring and obstructive sleep apnea are common in elderly and home settings, affecting cardiopulmonary health and disturbing bed partners. Traditional polysomnography (PSG) is costly, inconvenient, and has low compliance, while many existing non-contact monitoring systems remain limited to passive recording. This paper proposes a sleep management system with automatic gain control (AGC) and snore-driven intervention for home applications. The system integrates an under-mattress piezoelectric ceramic sensor, weight sensor, and dual acoustic sensors near the headboard to build a non-contact multimodal vital-sign monitoring scheme. An energy-feedback AGC is introduced on the pressure channel to mitigate signal amplitude differences caused by mattress types, postures, and body weights. Combined with snore-side localization based on Generalized Cross-Correlation with Phase Transform (GCC- PHAT) and Inter-channel Level Difference (ILD) and two silent electric linear actuators with Hall sensor feedback, the system performs gradual head-of-bed elevation on the snoring side, forming a closed-loop of "monitor-analyze-intervene." In multi-posture experiments with 10 healthy adults, the proposed system achieved heart rate (HR) and respiratory rate (RR) mean absolute errors (MAEs) of approximately 0.86-2.74bpm and 0.82-1.69 bpm in three major sleep postures. In a cross-mattress experiment on six mattresses of different thicknesses and materials with one subject, AGC reduced the average HR MAE from 3.08 bpm to 1.72 bpm and the RR MAE from 2.14 to 1.06 bpm, improving robustness across mattresses. Snoreside localization accuracy reached 89.7% (left) and 86.1% (right). A single-case snore-intervention experiment showed that, after closed-loop head-of-bed elevation to about 12 °, the subject's snore count and intensity decreased markedly, and snoring almost disappeared within about 10 minutes, demonstrating the engineering feasibility of closed-loop snore intervention. Overall, this work provides an integrated solution for non-contact sleep monitoring and active intervention in home and elder-care scenarios.
Fields of Study & Subjects
Non-Invasive Vital Sign Monitoring
Obstructive Sleep Apnea Research
Context-Aware Activity Recognition Systems
Publication details
- Journal / source
- International Conference on Consumer Electronics
- Pages
- 1–6
- Publication date
- 3 February 2026
- Language
- EN