How to shift a laser wavelength? Recently I have read something about the laser surface velocimeter, and after I have to rebuild from scratch (university project), I got a question: In the description they wrote that one of the laser beams is shifted with around 40 MHz compared to the other one. With this shift it is possible to see if the device is working and in which direction the surface is moving.
Second question: How difficult is it to create such an device, and what should I consider?
For explanation as requested:
This project is done for the university racing team, the device should be build into the race car for calibration. Time: Always if I have nothing to do, Equipment: I can use the local university equipment here, and budget: I hope I do not need more than the stuff which is available here (except for finalisation). Is this feasible for a one-man-project (master student, already built some interferometers)?  
Ah, and btw., the distance of the moving object is between 5 to 10 cm, but the speed goes up to 30 m/s, so is this a big problem (I thought I just have to use a faster photodetector (~30 MHz clocking rate according to the formulas on wiki))?
 A: There are two typical ways of generating two laser beams with a precise frequency offset from each other.  
The first requires you to have a specialized device known an an acousto-optic modulator (AOM).  These devices create a standing pressure wave in a specialized crystal.  This pressure wave acts as a diffraction grating in which the different diffraction orders are frequency shifted by $\pm nf$ where $n$ is the diffraction order and $f$ is the modulation frequency.  
The second way which is easier to work with, and cheaper if you have two lasers but no AOM, is to phase lock the two lasers to each other with an offset.  To do so you combine the two lasers on a fast photodetector and tune them until their frequencies are close enough together to see the beatnote.  Electronically mixing that beatnote with a reference oscillator provides an error signal for use in a feedback loop to lock one laser to the other.  There is a description of the technique here.
A: Can you link to the paper that you have read, please? In general, reflection from a moving object will cause a Doppler shift proportional to the object's velocity. Another way of looking at it is, that the movement will cause a time dependent phase change of the incident wave, which is most easily measured with an interferometer. 
So "all" you need to measure velocity with a laser is an interferometer, which measures the beat frequency of the reflected light against that of the direct light coming from the laser. Having said that, the devil is in the detail. 
A beat frequency of 40MHz means that the object is moving at 40 million times the laser wavelength (actually, there should be a factor of two in there, if I remember correctly), so that's approx. 40e6/s*0.5e6m/2=10m/s. That's a fairly high speed for a moving interferometer arm and, I would venture to guess, that it's not an easy task to build a reliable light path and detector for that kind of scenario, if the object were moving at an arbitrary velocity up to 10m/s and if it could move far away (as in more than a few meters). At the very least you may want to use a corner reflector on your moving surface, if you can.   
Can you tell us what that university project is for, and what kind of access to experimental hardware you have? What's the budget and how much time have you been given? 
A: This is usually achieved by using electro-optic and acousto-optic modulators (EOMs and AOMs resp.). These use the (nonlinear) dielectric and acoustic responses of an appropriate crystal to modulate its refraction index using an external signal. If used appropriately, they can be used to shift the frequency $\omega$ of an optical beam by an amount $\Delta$ by applying a sinusoidal microwave signal at frequency $\Delta$.
I am not completely sure of the exact range of frequencies they cover and it obviously depends on the application and the types of crystal that your device will require. EOMs can go up into the GHz range; AOMs are typically slower, but I think they can still meet your needs.
After that, the choice of which type of modulator, and which specific device, is all engineering, i.e. the hard bit. Unfortunately this is not really the site for that kind of expertise, but hopefully this will point you in the right direction.

This goes, of course, only as far as shifting a laser by a desired (and possibly variable) amount. To detect it, you need a spectrometer that's sensitive enough. If the shift is very small, though (and ~MHz shifts are small), you may have trouble getting a spectrometer that's sensitive stable enough for your needs, particularly if the environment is noisy.
If that is the case, one thing you can do is turn the concept of an EOM on its head and use it at the tail end of an interferometer to detect beat frequencies. The principle here is that you split your initial beam into two: a reference beam which you keep locally, and a probe beam which you send off towards your measured object. You then take the reflected probe beam and you mix it with your reference beam, and the acquired Doppler shift in the former will cause it to beat with the latter at the 40 MHz of the shift. Using exactly the same sort of nonlinear crystal used in EOMs and AOMs to get the beat signal out as a measurable output voltage, which can be analyzed using appropriate electronics. This technique is used, for example, for using and callibrating frequency combs for precision spectroscopy.
