کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
10344966 | 698123 | 2013 | 10 صفحه PDF | دانلود رایگان |
عنوان انگلیسی مقاله ISI
Performance comparison of accelerometer calibration algorithms based on 3D-ellipsoid fitting methods
ترجمه فارسی عنوان
مقایسه عملکرد الگوریتم های کالیبراسیون شتاب سنج بر اساس روش های سه بعدی بیضی شکل
دانلود مقاله + سفارش ترجمه
دانلود مقاله ISI انگلیسی
رایگان برای ایرانیان
کلمات کلیدی
کالیبراسیون، غیر تکراری، شتاب سنج، اتصالات الی فساد،
موضوعات مرتبط
مهندسی و علوم پایه
مهندسی کامپیوتر
علوم کامپیوتر (عمومی)
چکیده انگلیسی
Calibration of accelerometers can be reduced to 3D-ellipsoid fitting problems. Changing extrinsic factors like temperature, pressure or humidity, as well as intrinsic factors like the battery status, demand to calibrate the measurements permanently. Thus, there is a need for fast calibration algorithms, e.g. for online analyses. The primary aim of this paper is to propose a non-iterative calibration algorithm for accelerometers with the focus on minimal execution time and low memory consumption. The secondary aim is to benchmark existing calibration algorithms based on 3D-ellipsoid fitting methods. We compared the algorithms regarding the calibration quality and the execution time as well as the number of quasi-static measurements needed for a stable calibration. As evaluation criterion for the calibration, both the norm of calibrated real-life measurements during inactivity and simulation data was used. The algorithms showed a high calibration quality, but the execution time differed significantly. The calibration method proposed in this paper showed the shortest execution time and a very good performance regarding the number of measurements needed to produce stable results. Furthermore, this algorithm was successfully implemented on a sensor node and calibrates the measured data on-the-fly while continuously storing the measured data to a microSD-card.
ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Computer Methods and Programs in Biomedicine - Volume 111, Issue 1, July 2013, Pages 62-71
Journal: Computer Methods and Programs in Biomedicine - Volume 111, Issue 1, July 2013, Pages 62-71
نویسندگان
Matthias Gietzelt, Klaus-Hendrik Wolf, Michael Marschollek, Reinhold Haux,