Code Division Multiplexing (CDM) is found less suitable than other multiplexing methods, e.g.time division multiplexing or frequency division multiplexing, for high resolution interferometric sensor networks since optical field cross-correlation products yield noise that cannot be separated from the sensor signals. Calculations based on a fundamental approach have been used to simulate the output intensity from networks of interferometric sensors. We have applied the formalism both to a simple optical network, an unbalanced Mach Zehnder interferometer, and to a more complicated nested network of interferometers. The first case was used to show generally how the laser noise convert to intensity noise and to provide insight into how the sensor signal to noise ratio in more complicated networks is related to the laser and network parameters. The more complicated nested network was simulated since it has been intended for use in large scale all optical underwater hydrophone networks. CDM was applied to the network and the output intensity characteristics were investigated. We have performed the numerical calculations with various combination of laser and network parameters. None of the investigated parameter settings were found to give appreciable results. We attribute the negative result to the existence of cross correlation terms in the demultiplexed signals. It was found that the only way to reduce this impact is to use a low coherence laser source. But, even if the use of a low coherence laser source would reduce the cross correlation to a sufficiently low level it would also decrease the tolerable path imbalance of the sensing interferometers. Therefore, it is unlikely that practical systems with acceptable signal to noise and crosstalk conditions can be realized with CDM as the method to distinguish the network sensing elements.