Location estimation of UWB-based wireless capsule endoscopy using TDoA in various gastrointestinal simulation models

Abstract

Abstract

Wireless capsule endoscopy (WCE) is a revolutionary field that aids in treating gastrointestinal disorders. For the development of a futuristic endoscopic capsule, identifying the location of abnormality is challenging yet a crucial step in determining the treatment procedure. Though the present-day wireless capsule endoscopes certified for endoscopic procedures predominantly work on the Medical Implant Communication System (MICS) band, the applications based on the Ultra-Wide Band (UWB) are gaining popularity for their immense possibilities. While received signal strength, time of arrival, phase of arrival, and angle of arrival are the basic parameters applied in research for location estimation of WCE, this paper uses a time difference of arrival (TDoA) approach using the Chan algorithm. To test the effectiveness of the algorithm, a series of UWB propagation experiments are performed utilizing human voxel models to find out the variance in distance error using an advanced electromagnetic simulation environment, which is then applied as the error to the distance estimate of the Chan algorithm, and the performance is analysed using different cases. Positioning receivers in three rows reduces the estimation error by 44%, and positioning the reference receiver in the middle row reduces it by 33%. The algorithm performance is observed for different variances in the distance estimation using different numbers of receivers, and the results are compared to the Cramér-Rao lower bound (CRLB). The calculated error in thickness from the different sections of abdominal tissues of the individual voxel models is applied to the distance estimates from the corresponding receiver sections prior to the WCE location estimation. The RMSE in WCE location estimation is found for individual voxel models, and the error is observed to reduce approximately from 4 mm to 1 mm, with variations in individual models, when the number of receivers are increased from 9 to 33.