repairing a shielded cable?

I am rewiring a car with an updated ignition system/ECU and there are a couple of shielded cables (low voltage items like a knock sensor, etc). One cable is basically two 22ga wires wrapped with a stranded wire (almost RCA patch cable-esque); the other looks almost like a very small coax cable. The ends are very proprietary and not easily (if at all) serviceable. My only real option is to splice two good halves together. I have heard of the "carputer" guys using aluminum tape to help shield the wire, but no real idea. Does anyone know of a way to do this properly? Thanks.

Yep - usually "any shield" will work. Rarely is "continued impedance" a issue for car sensors. (Unlike antennas & video etc.) Sometimes shields are not needed. IMO they were included on older systems because they thought they needed it. But they may have been added to some systems because they did (need it). Even in cases where shields are needed, it is often possible to be exposed (unshielded) for a small length provided the shields are joined. (A floating shield is often useless.) And where impulse noise is injected, small caps can often replace shields. (IE - caps or filters remove/shunt high-frequencies injected onto low frequency timing and O2 sensor signals.) One problem is ascertaining WHY the shield is there. Is it okay without it EXCEPT with wet ignition, or using mobile phones, or in proximity to transmitters, etc?

The only two signals I'm most concerned with is the O2 and the knock sensor. It is my understanding that they shielded these because the signal/voltage is so low. I have both harnesses for the knock sensor and O2 that will make it within inches of the splice at the ECU, so my guess (hope?) is that being inside the car and inches away from the ECU will suffice.

Both should be fine - just join the shields. O2 sensors are uA devices (never test their resistance with a DMM!), and knock sensors are microphones. Yes, both signals are high impedance, though being "analog level" in nature, noise is unlikely to effect them. They are/were probably more concerned with destruction of the O2 sensor (uA injection) else induced knock-sensor kHz noise being confused with knocking. High switching currents could do the former, and spark could effect both.

Thanks very much oldspark! I was hoping that was the case :)