How much space between components in a matching network?

[scotty7]

I'm looking for a rule of thumb for distances between components in a matching network.

I have a project that I've been making good progress on, and I've come to the part where I need to match an antenna.
The antenna I have selected has a fancy matching network in the datasheet. Now I know I should follow their layout exactly, but since I can't follow it due to space constraints, I'm going to ask you friendly bunch and try it myself. I have a VNA to test with, so this isn't a blind approach.

Attached is a picture of the *very preliminary* layout. I'm wondering if I can place RF components this close to each other and get away with it. I don't know enough about matching network layout, so I'm hoping someone can chime in with experience.
From what I'm envisioning, there will be some leakage into the nearby components, but I should be able to compensate for it by adjusting values (right?).

If there's anything you can recommend please let me know! I'm here to learn
(top of PCB goes to antenna, bottom is a u.FL connector)

[rfclown]

In general, unless some of the linterconnect lines are calculated transmission line lengths, putting things closer together is ok (and usually better). At high frequencies (>GHz) even short lengths of interconnect line can be significant. Putting components side by side COULD induce some unwanted coupling. In your layout the only place I see this is in the 4th component from the node labeled ANT. You can easily move the component in series with the line going to the U.FL connector to straight inline with that trace (which I hope is 50 ohms if the antenna connected to the U.FL is a 50 ohm antenna). Then you have room to straighted out the components that are side by side. Hopefully I'm explaining my self; eliminate potential broadside coupling between components which are in series with each other. The ones that are wired in parallel are fine being parallel. Even saying all that, I mention it mainly because you have the room to spread it out. If your space with such the U.FL position was right at the first component, I'd be ok with it (although I'd still swing the lower left component verticle so that it wasn't broad side with the component above it - because it's easy to do and the room is there).

I highly recomend QUCS which is a free program. You can model your circuit and see what the impedance is (with ideal components) at the ANT terminal (with 50 ohms at the U.FL terminal). If it is close to what the datasheet is suggesting for your part, then you know that that the circuit isn't dependent so much on the layout.

[rfclown]

I agree with everything that evb149 said. I was just answering your original question, but now that I look at the arrangement, it looks more like a filter than a matching network. Straightening it out helps with being able to adjust the topology if desired.