Distributed Fiber Optic Sensing - The Sensing Mechanism

7 May 2014

Alright, here it is, the day has finally come; the mechanism is light scattering in the form of Raman, Brillouin, or Rayleigh scattering! I guess ... we're done? We'll ... maybe not so much. Scattering seems to be a topic I've found a lot of engineers are timid about. It might have to do with the stigma associated with high frequency theory being difficult. It really isn't hard; things interact with one another and in the process sometimes interesting effects can happen. For example, take the CLASSIC moving train example. A train approaches you, not directly of course but in your general direction, and you hear the horn begin to blow. As soon as the train moves past you the horn pitch changes, that is, the perceived frequency emitted by the horn changes. This, of course, is the Doppler effect and if you're unfamiliar with it I suggest reading up on it since it has numerous applications. In fact, it is a large part of Distributed Sensing based on Brillouin Scattering. Since my background is Brillouin Scattering, I will discuss the details of it in a fiber optic cable.

As a laser signal travels down a fiber optic cable it interacts with the medium that it travels on: the glass. The laser is usually a very low wattage, but because the core of the fiber is so small, the power density is huge! Suppose a 1 mW laser is directed down a fiber with a 10 µm core; the resulting power density is 3.2 MEGA watts per meter squared. Because the power density is so high, the lightwave from the laser can significantly modify the glass shape as it moves, and this creates a periodic disturbance of the index of refraction of the fiber (look up Electrostriction for the theory buffs). The net effect of this disturbance is an acoustic wave that moves in the fiber. The velocity of the acoustic wave is given by:where V_A is the acoustic wave velocity,

k is the bulk modulus of the fiber,

and ρ is the fiber density.

Have you made the connection yet? Think about what happens to the density of the fiber if the temperature changes, or if the fiber is strained. The density changes, and if the density changes the velocity of the acoustic wave changes. To make a sensor out of this, we'd need to measure the velocity or frequency of the acoustic wave (frequency because frequency and velocity are related). This frequency is referred to as the Brillouin frequency and is defined as:

where f_B is the Brillouin frequency,

n is the refractive index of the fiber,

V_A is the acoustic wave velocity,

and λ_o is the wavelength of the incident light.

So, to summarize all of these statements:

The frequency of the acoustic wave, or the Brillouin frequency, is dependent on the temperature and strain of the fiber. If you can measure this frequency, one can determine the temperature and/or strain of a fiber optic cable.

Enter the Doppler shift. The lightwave meets this acoustic wave and is backscattered (back towards the sending end) shifted by the acoustic wave frequency. We just need a clever way to detect these frequency changes, which of course, will be the topic of the next blog post.

Unrelated Interesting Fact: Ever wonder why movement stops at absolute zero? Temperature doesn't really "exist". It's just a parameter that measures the average speed of the molecules in a sample volume. So, if the average speed (not velocity) of the molecules is zero, none of them are moving and the temperature is absolute zero.