You didn't specify whether or not you actually cheated, so let me answer you in both cases.
(a) if you didn't cheat, then you won't know what it is that arouse your professor's suspiciouns. Your first priority is discovering what exactly the alleged evidence is against you. Send your professor an email, asking him to clarify what exactly he's accusing you of and to detail his evidence ahead of the meeting, and then you can meet him to defend yourself against the allegations. Asking for a meeting with you without telling you what exactly you did wrong is an informational asymmetry meant to catch you flat footed. Innocent people get accused and convicted, it happens, and sometimes it's because they mistakenly self-incriminated.
Once you know the evidence behind the accusation, it's time to meticulously prepare your defense. Again, just because you're innocent, doesn't mean you can't get convicted. You might even seem guilty to a reasonable person. Document everything. Write down a thorough rebuttal to each accusation. Give perfectly reasonable and believable explanations to anything about your work that seemed suspicious to your professor.
NOTE: different institutions deal with allegations of cheating in very different ways. In some institutions, the professor is judge jury and executioner. In others (like mine), the professor's only role is in submitting their allegations to a council, which independently determines guilt and sanctions, while giving you the opportunity to defend yourself. If your institution is the second type (I believe most are), then the council are the real people you have to convince. Of course it might still be worthwhile to talk to the professor, maybe you can convince them and they won't submit the case to the council (which will save you a lot of stress and headache, because these things can take a long time).
(b) If you did cheat...the ethical thing to do is to come clean. Note that you will not get any brownie points for doing so. In most cases they won't go easier on you just for being forthcoming. If the ethics of this aren't important to you, you can do exactly what I described in (a), especially if your case is borderline. If you argue it right and cast enough doubt on their suspicions, you might get off.