In most of RF receivers (RX) the analog chain aims to improve the overall Signal to Interference plus Noise Ratio (SINR) by filtering out interferences in the Signal Of Interest (SOI) vicinity. However, cognitive radio applications require a high dynamic range RX, pushing a high constraint on the analog section. In particular, nonlinearities may damage the overall SINR and Bit Error Rate (BER) due to blocking and intermodulation processes in presence of strong interference. RX linearization state of the art methods that rely on the feedforward concept have shown the RX linearization improvement on the BER. Nonetheless, benefits of other linearization methods such as the feedback loop concept, are unstudied. In this article, our aim is to decrease the power of unwanted harmonics thanks to a feedback loop. We first explain the variable gain enhancement concept. Then, we present an algorithm that dynamically adapts the RX parameters (gains and ADC margins) to make it work in a linear regime. We analyze the effect of such a method in terms of BER. Important BER improvement is noticed in blocking and saturating situations due to co-channel interferers.