Data-aided channel estimation for wideband CDMA

The effect of channel estimation error on the performance of a wideband code division multiple access link using a pilot channel multiplexed with the data channel is analyzed. The gain achievable by improving the channel estimation algorithm is studied. Analysis and simulation results are presented for a data-aided channel estimation scheme that effectively extends the duration of the pilot signal. Using the maximum likelihood detection principle, it is shown that making tentative minimum mean-squared error decisions on the data bits using estimates obtained from the pilot channel can extend the number of samples used in the channel estimation, thus improving the channel estimate quality. Simulations show that a gain of up to 1.5 dB in block-error rate performance is achievable using the proposed channel estimation scheme for both Convolutional and Turbo codes. It is shown that the performance gain depends on the information bit rate, the channel characteristics, including the Doppler frequency, and the multipath fading channel power delay profile. Simulation results also show that for weak paths, data-aided channel estimation benefits significantly from diversity combining.