The reason radio (microwave, actually) is so attractive is that there is a big dip in the natural background noise in the microwave (very short radio waves) region of the spectrum, so energy transmitted at those frequencies would tend to be detectable over much greater distances than using any other frequencies. This has been the preferred listening strategy since the 60s.

Another reason for using microwaves is that even though the galaxy is relatively quiet at those freqs, there are a few "spikes" in that region of the spectrum which are the result of natural emitters (excited molecules in the thin gas between the stars). This means any species interested in astronomy would probably spend a lot of time listening in that band during their routine galactic mapping (just as we do).

For example, at the most noise-free part of the spectrum are two "spikes", one for ionized hydrogen gas (H+), and another for the hydroxl radical (OH-). You may of heard of this, SETI weenies call it "the waterhole", and any frequency between these two would be a logical place for someone to transmit a beacon signal. Another option would be to transmit (or listen) at a harmonic of some very noisy frequency, like the monatomic hydrogen "spike" at 21cm. The galaxy's hydrogen emits a lot of noise at that frequency, but at exactly twice, or half, are quiet spots that would be a logical place to attempt communication.

The major limiting factor is that if there is somebody out there transmitting at any of these freqs, and if we guess the right one, the volume of space we could cover with our equipment
would depend on the bandwidth, the integration time and the lobe we chose: the narrower the bandwidth, the more likely we are to detect a weak signal out of the noise, but also the more likely we are to miss it altogether since both receiver and transmitter are moving through space, Doppler shifts could move the signal out of our band in a way which is impossible to predict ahead of time. So if we make the bandwidth too wide, noise hides the signal, too narrow and we miss it all together.

Integration time refers to how long we listen at one spot on the sky. If we listen a long time, the noise tends to cancel out while the signal piles up (like a time exposure with a camera). However, if we listen too long at one spot, we don't have time to observe too many spots.

The "lobe" parameter is the angular size on the sky of the footprint of our receiver. The transmitter is likely to be a point source, so if we listen with an omnidirectional antenna, we get the whole sky, signal+noise. With a tiny reception lobe (very big parabolic antenna at high frequencies) we get the same signal, but much less noise, so we can see deeper. Unfortunately, we have to look at more spots in the sky before we are likey to pick something up.

SETI people spend most of their time arguing the tradeoffs between these competing parameters. SETI is within our technical capabilities, but just barely, and it depends primarily on what the other guy is doing. We spend most of our time trying to guess what WE would do if we were in THEIR shoes, and trying to second-guess alien strategy is, at best, a risky endeavor. The galaxy is big enough and old enough that it could be positively crawling with radio traffic and we would still be very lucky to intercept some of it: it's a needle in a haystack kind of problem.

It is my opinion (not all SETI types agree with me on this) that microwave broadcasts with radio telescopes are the way to go, but I could be wrong. Other systems (like beamed signals or lasers) would work much better but ONLY IF the ETs already knew we were here. I don't think that's very likely. Although I suppose it is possible that we might pick up accidental leakage from some alien industrial activity (the I Love Lucy theory) my own feeling is that aliens would not waste energy by bombarding space with signal that wasn't meant to heard. All their routine communications would be on tight beams, and all their technology would be employed to minimize leakage. The galaxy is big and noisy and ETI are spread thin in both space and time; it seems to me that the only signal we're likely to hear is a deliberate SETI beacon, and it will be from so far away that we won't have a chance of hearing it unless they deliberately designed it to be heard.