Using mobile phones to capture mode shapes: a feasibility study

The issue of using the sensing capacity of mobile phones for structural monitoring has received increased attention the past few years. This paper presents a study carried out to examine the feasibility of using a number of mobile phones to capture the mode shapes of a footbridge. The ability of an individual mobile phone to accurately measure acceleration /frequency has been demonstrated by the appearance of a number of different mobile applications (apps) on both android and ios platforms. However, to calculate mode shapes the individual sensors need to be synchronised. For conventional sensing system this is easily achieved as data from all sensors is being recorded by a central data logger. For phones this is not the case, i.e. each phone is logging the data directly to its own memory. Where a unique identifier (e.g. clapper board type event) can be implemented simultaneously at the start of all the phone signals this will allow the signals to be lined up in post processing. However, there is no guarantee that the signals won't drift from one another subsequently. Other issues include, the variation in sensitivity /noise floor of the accelerometers in different phones, and the difficulty some phones have in maintaining a constant scanning frequency. All of the above issues are examined in this paper and it is shown that the drift between individual phones is actually quite small. Moreover it is shown that for simple structures with sufficient levels of vibration, a series of phones can be used to determine an approximate mode shape. However, for lower levels of vibration the MEMS accelerometers in the phones struggle to capture the structure's vibration, i.e. the level of vibration is below the noise floor of the sensor and mode shapes cannot be detected reliably.