New Simple and Efficient Bounds on the Probability of Error in the Presence of Intersymbo1 Interference and Gaussian Noise

The problem of determining or of bounding the probability of error in the presence of intersymbol interference and Gaussian noise is an important one in data communications. In this concise paper we describe new upper and lower bounds for the probability of error when the peak distortion is finite (in practice, when the eye is open) and the data are independent. Over a range of S/N of interest in data communications; these bounds compare favorably with previously published bounds of similar complexity, especially when the eye opening is small. They can be applied to multilevel AM and CPSK transmission.