An introduction to channel equalization

During its propagation path from transmitter to receiver the radio signal is distorted by the channel and affected by noise. HF-communication (1-30 MHz) is characterized by ionosphere propagation that leads to multipath and fading. The distortion of the signal complicates signal analysis e.g. modulation classification by an unintended receiver, such as an Electronic Warfare (EW) system. The channel distortion also increases the symbol error rate at the intended receiver. The commonly used modems standardized by e.g. STANAG 4285 include an equalizer that compensates for the distortion caused by the channel. The equalization often includes a training sequence known to the receiver. An unintended receiver, e.g. an EW system, doesn´t have access to the training sequence. In this case blind equalization techniques could be used. The purpose of this report is to give an introduction to channel equalization. A simple model of HF-propagation, including multipath and Doppler spread, is implemented in MATLAB and used in the simulation runs to evaluate different equalization techniques. Algorithms for linear equalization and equalization using the Viterbi-algorithm are implemented in MATLAB. These techniques includes a training sequence, known to the receiver. Simulations are run to evaluate these techniques on a 8PSK-signal defined by the standardization agreement STANAG 4285.