Ultra low power wearable sleep diagnostic systems

File Description SizeFormat 
Rajagopal-Mohan-Kumar-2014-PhD-Thesis.pdfThesis33.76 MBAdobe PDFView/Open
Title: Ultra low power wearable sleep diagnostic systems
Authors: Rajagopal, Mohan Kumar
Item Type: Thesis or dissertation
Abstract: Sleep disorders are studied using sleep study systems called Polysomnography that records several biophysical parameters during sleep. However, these are bulky and are typically located in a medical facility where patient monitoring is costly and quite inefficient. Home-based portable systems solve these problems to an extent but they record only a minimal number of channels due to limited battery life. To surmount this, wearable sleep system are desired which need to be unobtrusive and have long battery life. In this thesis, a novel sleep system architecture is presented that enables the design of an ultra low power sleep diagnostic system. This architecture is capable of extending the recording time to 120 hours in a wearable system which is an order of magnitude improvement over commercial wearable systems that record for about 12 hours. This architecture has in effect reduced the average power consumption of 5-6 mW per channel to less than 500 uW per channel. This has been achieved by eliminating sampled data architecture, reducing the wireless transmission rate and by moving the sleep scoring to the sensors. Further, ultra low power instrumentation amplifiers have been designed to operate in weak inversion region to support this architecture. A 40 dB chopper-stabilised low power instrumentation amplifiers to process EEG were designed and tested to operate from 1.0 V consuming just 3.1 uW for peak mode operation with DC servo loop. A 50 dB non-EEG amplifier continuous-time bandpass amplifier with a consumption of 400 nW was also fabricated and tested. Both the amplifiers achieved a high CMRR and impedance that are critical for wearable systems. Combining these amplifiers with the novel architecture enables the design of an ultra low power sleep recording system. This reduces the size of the battery required and hence enables a truly wearable system.
Content Version: Open Access
Issue Date: Sep-2014
Date Awarded: Feb-2015
URI: http://hdl.handle.net/10044/1/38556
Supervisor: Villegas, Esther Rodriguez
Department: Electrical and Electronic Engineering
Publisher: Imperial College London
Qualification Level: Doctoral
Qualification Name: Doctor of Philosophy (PhD)
Appears in Collections:Electrical and Electronic Engineering PhD theses



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commonsx