Adaptive Optimization of Control Parameters for Feed-Forward Software Defined Equalization

In this paper we briefly describe the design, implementation, and evaluation of a novel adaptive optimization approach for the feed-forward software defined equalization (FFSDE) method using the least mean squared (LMS) algorithm. In our design, we adaptively change the filter length (N) and step size ($$\mu$$μ) to achieve the optimal bit error rate value. We used a vector signal generator RF PXI-5670 and a vector signal analyzer (VSA) RF PXI-5660 to test the validity of our approach. We implemented our method for the M-ary quadrature amplitude modulation (M-QAM) scheme in the VSA (which served as a receiver). The experimental results showed that we achieved high convergence speed and accuracy for rapidly changing transmitter channel characteristics. The automatic optimal setting feature of the LMS Algorithm parameters N and $$\mu$$μ, enabled us to solve the hardware configuration problem for the FFSDE method. Determination of the LMS Algorithm training sequence size for the particular M-QAM allowed us to eliminate redundant data of the training sequence and increase the throughput.