A method for accurately recording heart rate (HR), respiration rhythm (RR), and body movement (BM) during sleep using a network-based system is proposed in this article. Its application to the long-term monitoring of HR, RR, and BM during sleep was examined. HR, RR, and BM were detected by pressure variations corresponding to changes in the heartbeat and respiratory motion, which were measured by a sensor unit placed beneath a pillow during sleep under completely unconstrained conditions. The pressure signals were digitized and transmitted to a remote database server using transmission control protocol (TCP)/Internet protocol (IP) via a netbox. In the server, the data were processed to obtain HR, RR, and BM. The overall performance of the system was examined using data collected over a 13-month period from a female subject. Besides the long-term profiles of HR, RR, and the periods during which the HR and RR were undetectable owing to BMs during sleep, the average frequency of BM in a day varied from 4.4 to 22.4/h, and the mean frequency over 332 nights was 8.3/h with a standard deviation of 2.2/h. Periodic biorhythms can also be assessed using the profiles of the average HR and certain frequency-domain parameters of HR variability. The rhythmic property of these profiles was confirmed to coincide with the subject's menstrual cycle. The results of this 13-month trial operation show that this system can be installed in the home environment; used to monitor HR, RR, and BM during sleep; and analyze various physiological rhythms in humans over prolonged periods.