The evaluation of bit error probabilities for anti-jam communication systems is generally more difficult than for conventional communication systems. This is partly because of the fact that intentional jamming signals can be more varied than the additive Gaussian noise model which is typically assumed in conventional communication systems. In addition, most anti-jam communication systems use coding techniques where coding gains at 10 ^{? 5} bit error rates are typically 30 dB to 60 dB compared to 4 dB to 5 dB with the additive Gaussian noise channel. (This will be shown in greater detail in Part 2 , Chapters 1 and 2 .) Even with conventional communication systems, the evaluation of exact coded bit error probabilities is usually difficult to do and, thus, easy-to-evaluate upper bounds are used [1 ], [2 ], [3 ].

The purpose of this chapter is to present a general expression for upper bounds on coded bit error probabilities which apply to all coded communication systems that use enough interleaving and deinterleaving so that the channel can be modeled as memoryless. We shall show that these error bounds are tight over ranges of interest by comparing them with exact bit error probabilities when they can be found. They are also applicable to arbitrary decoding decision metrics and a wide class of detectors. This generalization is necessary for anti-jam communication systems where there are no optimum detectors and maximum-likelihood decoding decision rules are not known at the...