Intrusion detection
Hi demokesola:
Both radar (electromagnetic waves) and ultrasonics (sound waves) are very commonly used for intrusion detectors, and both can do the job quite well. The body is a very satisfactory reflector for radar waves because it is full of water and conductive chemicals.
Radar intrusion detectors tend to be somewhat more expensive due to the technology involved. They normally operate on the doppler effect by which a continuous wave is transmitted and reflected waves are detected in the same cavity that forms the transmitter. If nothing in the field of view of the transmitter/receiver is moving, the jumble of reflected signals are all at the same frequency as the transmitted signal (no doppler shift) and, when mixed with the transmitted signal inside the transmitter cavity, there is no difference frequency and therefore, no audio frequency output.
If something in the field of view is moving (with a component of motion toward or away from the transmitter), the return signal is doppler shifted, there is a difference frequency created in the mixer in the cavity, and an audio frequency output results. This frequency of this AF output is directly proportional to the velocity of the moving object long the transmitted beam. For the commonly used frequency for intrusion detectors (about 24.5 GHz, I believe), I think the AF is about 30 Hz per mile per hour target velocity. (I use units of mph because I built one of these for vehicle speed detection many years ago.)
While it is possible to build something like this "at home" using a tin can as the radar cavity/antenna (google Ramsey electronic kits), it is much easier and wiser (from a safety and compliance perspective) to use ready-made radar modules consisting of a die-cast cavity with attached mini-horn antenna, complete with transmitter gunn diode and receiver/mixer diode. You just apply well regulated DC power and are handed the AF output on a platter. All your circuit work is at audio frequencies and, therefore, non-critical and simple.
For intrusion detection you don't normally care about MEASURING the velocity of the target. You just want to know if something is moving in the field of view. For that you just need a simple AF amplifier, rectifier, and threshold detector to activate a relay. This is just what is inside commercial radar intrusion detectors costing several hundred dollars (US). Just a few transistors or an op-amp, rectifier, and comparator required.
At the time I was experimenting with them, the complete transmitter/receiver modules were in the US$100 neighborhood, and several people were making them. At the time, I used modules by Amperex (the tube maker), but I think they are defunct. ALPHA had a pretty complete line of modules at one time. Don't know if they still exist, either.
Interestingly, 25 KHz and 40 KHz ultrasonic wavelengths are of the same order of magnitude as radar intrusion detector module wavelengths, and you can use a similar approach, as some makers of ultrasonic intrusion detectors do. You can also use a true radar-ranging approach, in which a very short burst of ultrasonic sound is transmitted and the time-of-flight for the return signal is detected. This is used for the architectural distance measuring devices you can buy in the hardware store. This approach is a little more complex to implement and is probably overkill for an intrusion detection scheme. At one time, National Semiconductor produced a chip to implemeent this ranging function. Look in their "special function ICs" section.
You can buy a matched pair of ultrasonic transducers from many surplus stores and catalog electronics outlets for a dollar or two. For these, you have to generate the ultrasonic frequency to drive the transmitter.
Googling, "ultrasonic intrusion detection," should keep you out of the bars for a while, with about 23,000 hits. "Radar intrusion detection" gives you about 81,000 hits.
Have fun.
awright