I have created an iPhone application which fetches data held on a server in an XML file. How do I check that the request for the data is coming from my app and not from some other source such as another iPhone app or a desktop browser since currently you could just trace the iPhone request on your LAN with Wireshark and then use the captured URL to load the data in a desktop browser. I'm thinking I'm going to need to serve the file via PHP or something and use some sort of User Agent validation or a challenge-response sequence. If someone could provide a code sample I'd appreciate it.
Short answer: You can't. But you can indeed make it harder.
Whatever you do, it will be possible to circumvent it - user agent validation is extremely easy to circumvent; challenge-response will require disassembling of your app, but it's still possible.
However, all your nice protections won't help against network sniffing. Unless you also encrypt the transfer someone can simply sniff the plaintext data instead of breaking your "protection".
IMO the main question shouldn't be "How do I protect it" but rather "Why would somebody want to get the raw data? Why shouldn't he get it?"