How to filter noise on loop of wire?

I have an antenna that is just a loop of wire operating at HF (8MHz) and nearby computer induces noise on it. How to filter this noise on the antenna itself. I have no access to equipment past antenna terminals.
Wire loop is ~2.5 m 16 gauge wire, measures 3 ? 4 uH inductance and no resistance to speak of. When connected to terminals it is resonant and resistive part of impedance varies in the range 200 ? 1000 Ohm.
I tried clipping (unknown) ferrite beads on the wire but far did not have much luck with it. In fact it seems it makes it worse.

A sharp Band Pass Filter may work..

Thank you BigBoss,
Do you know how to make it?
I did bandpass filters for 50 Ohm, 300 Ohm balanced but here ... it is just a wire.
?

Is this antenna for receiving purpose ?
In fact inserting a sharp filter will not solve the problem %100 because if there is a noise coming from the computer in the environment, this noise will also have some components at 8MHz.
But it's better than nothing.Also, 8MHz sharp filter will decrease the sensitivity of the receiver, it shouldn't be ignored.
Anyway, a ceramic based filter will no doubt have a huge insertion loss.You can design a L-C filter but it will also have a remarkable loss.
So, the best option is to fade away the computer.. -

I wish it would be that simple. : )

Suitable bandpass L-C filter would do the job likely but to calculate one requires to start with pre-determined impedance. An here impedance is Z = 0 ohm +j200.

Filter can be designed for arbitrary complex termination of both ports. However, to achieve a useful filter characteristic with finite Q, at least one port must expose a real impedance component. This can be usually expected for the receiver side.

Sounds interesting FvM. Can you point to calculator or instruction explaining how to do it? Calculator preferably. I tried calculating Chebyshef filter long time ago, gave up. : ). Qucs software works reliably.

Nuhertz Free is the free version (feature limited) of Nuhertz filter synthesis software.
http://www.nuhertz.com/index.php
http://www.nuhertz.com/register/reg_free.html

But the same fundamental question as BigBoss: computers create a wide spectrum of signals in the MHz range. Is the computer noise that causes trouble outside your system's operating frequency range, so that an additional band pass filter can remove it? What bandwidth does your system use? What effect from that noise do you observe in your system?

You mentioned that ferrite beads didn't help much. Did you place them on all the computer cables, as close as possible to the computer?

a Shielded Loop Antenna

replace your wire with coaxial cable. Can use a balun for the output to your black box. Between the balun and the shield...should cut the computer noise down a alot

I tried shielded loop and surprisingly I did not see noise improvement. Also coaxial cable loop measured higher inductance than wire loop. Is it caused by structure or it is caused by dielectric coefficient of cable?
Still thinking about band-pass filter. How one make band-pass filter to and inductor?

The shield is only effective with a respective ground terminal at the receiver. And it only (partly) blocks capacitively injected interferences, not magnetic fields.

If the receiver has already reasonable band selection, an additional filter might be useless. At worst case, the interfering signal is exactly at the same frequency as the wanted signal.

You might have ground loop noise. Try isolating both terminals of the antenna with a series capacitor

Thank you everybody who replied. I am seemingly solving this problem using 'other means'.
But I still do not know how to design band-pass filter for inductor. Or capacitor - pure reactance