Adaptive filtering using LMS for digital TX IM2 cancellation in WCDMA receiver

In this paper, a digital technique for canceling second-order inter-modulation (IM2) of transmitter (TX) leakage in WCDMA direct-conversion receivers is described. Digital TX signal is used as a reference to generate a digital IM2 signal and then is passed through the auxiliary digital filter chain as in receiver chain. Adaptive least mean square (LMS) algorithm combined with a delay estimator provides a very robust technique to act as a matched filter to estimate the channel which causes leakage from TX to RX. The delay between the receive path and the reference path is estimated by peak correlation detection. This IM2 cancellation works well even though the actual IM2 components are far below the desired signal and thermal noise. The effect of the DC offset is removed using a high pass filter in both paths. The receiver is implemented and the algorithm is applied using fixed point model in ADS.