Aerospace and Electronic Systems Magazine September 2017 - 34

Feature Article:

DOI. No. 10.1109/MAES.2017.160248

Evaluating Performance of MEMS Barometric Sensors
in Differential Altimetry Systems
Dimosthenis E. Bolanakis, Hellenic Air Force Academy, Tatoi, Greece

INTRODUCTION
Micro-electro-mechanical-systems (MEMS) barometric sensors
are extensively employed in positioning systems for the acquisition of altitude data. Confidence in their general approval on position location applications is due to the following reasons:
A. the relatively high accuracy of vertical position measurements,
which is constantly improved in the most recent commercial
sensor devices;
B. the inability of the allied-and dominant in outdoor environment-GPS and GNSS products to operate within buildings.
Indicative examples of barometric altitude readings in consumer,
medical, and aerospace applications [1] are as follows.
Detecting floor [2] and mode of transition (i.e., elevators, escalators or stairs) [3] inside large buildings can be addressed for
indoor navigation;
Identifying human physical activities (e.g. sit-to-stand) [4] and
falls [5] is able to provide telehealth solutions;
Fusing information from barometric altimeters with GPS may improve performance of GPS-based landing systems [6], while
the employment of two barometers in a system can be used for
measuring the airplane position angles (e.g. pitch and roll) [7].
The altitude determination by the use of atmospheric pressure
measurements (also known as barometric altimetry) features two
main problems that severely affect accuracy of measurement. The
first problem is induced by the ambient air pressure changes, as
time elapses, and is particularly detectable in single-device systems. In a single-device setup, the vertical position is decided on
two consecutive measurements of altitude at two distinct positions
(also known as absolute barometric altimetry method). A relatively
small change of air pressure in-between the two measurements
may result in a significant error in vertical position detection.

Author's address: Department of Air Force Science, Hellenic
Air Force Academy, Dekelia Air Base, Acharnea, Tatoi, 13671
Greece. E-mail: (dbolanis@cc.uoi.gr).
Manuscript received November 5, 2016, revised January 5,
2017, and ready for publication January 5, 2017.
Review handled by M. Braasch.
0885/8985/17/$26.00 © 2017 IEEE
34

The second problem results from manufacturing dissimilarities of MEMS sensors and has an effect on dual-device systems.
In a dual-device setup, the vertical position is decided on concurrent altitude measurements by the simultaneous arranging of the
two devices at two distinct positions (also known as differential
barometric altimetry method). Because of these manufacturing
dissimilarities, such as differentiations in the reference voltage
provided by the on-chip voltage regulator (which subsequently
affect the pressure-response curve), two identical MEMS devices
generate a slightly different measurement when sensing the exact
same pressure value. The deviation in sensors' output signal varies
randomly with the range of sensing pressure and is affected by the
air temperature, as well. Thereby, it cannot be defined by a single
calibration [8]. Differential barometric altimetry is more resistant
to the fluctuations of air pressure (because of their simultaneous
influence to both sensors of the system) and hence, preferable
when high accuracy is required [9], [10]. However, this particular attribute of MEMS sensors affects the long-term measurement
stability of the system.
This article presents a prototype testbed which is addressed
to evaluate performance of MEMS barometric pressure sensors
in differential altimetry systems. Performance analysis applies to
up-to-date commercial sensor devices. That is, bme280 [11] and
lps25hb [12] sensor devices, of the leading and prevalent Bosch
Sensortec and STMicroelectronics MEMS suppliers, respectively.
The proposed technique is ideal for the arrangement of low-cost
experimental observation of the system performance, over the
conventional and exceptionally expensive apparatus required for
conducting accurate experiments. Measurement analysis illustrates
that sensors of comparatively equal accuracy may produce significant disagreements in the system performance.

EVALUATING PERFORMANCE OF MEMS SENSORS
TESTBED PROTOTYPE AND MEASUREMENT TECHNIQUE
The experimental testbed applies to wireless sensor networks and
consists of four endpoint measurement (battery-operated) devices
for the concurrent acquisition of atmospheric pressure, as well as
one coordinator for collecting data (through USB) to a host PC.
Data acquisition process is controlled by a LabVIEW-based user
interface. The coordinator device consists of pic-p26j50 microcontroller and mod-mrf24j40 rf boards, delivered by Olimex.
Endpoint devices employ an Olimex mod-zigbee-pir board, which

IEEE A&E SYSTEMS MAGAZINE

SEPTEMBER 2017

Table of Contents for the Digital Edition of Aerospace and Electronic Systems Magazine September 2017