AI Field of vision

How should I go about implementing my AI-s field of vision? Raycasts? Something else?

EDIT

Here is a better explaination. how can I find out what it can see. When it looks to one side, than it cannot look to the other.

-------N-----------
--------SSSSN------
----N-|-S--->--N---
---------SS-N------

S is what it can see, N is what it cannot see, | is wall, > is enemy facing left, - is just void.

Raycasts are kind of slow and dependent on raycast resolution. You might want to look at using a few box triggers to detect what is in the field of vision and then do raycasts for checking occlusion.