Broadband Power Line Communication (BPLC) is a popular technology that utilizes the existing power line
networks for the transmission of information. The power line channel is affected by multipath propagation. The BPLC
system performance is degraded due to multipath propagation and the noise in the channel. One of the promising
approaches to mitigate the problem is to implement a proper modulation technique. Adaptive OFDM (AOFDM) method
can potentially aid to achieve high data rates in Broadband Power Line Communication system. To realize this potential,
the transmitter needs accurate Channel State Information (CSI) for the upcoming transmission frame. This paper proposes
an Adaptive OFDM scheme with an optimal approach of Exponential Effective SNR Mapping (EESM) on prediction of
the BER over all subcarriers. A channel prediction scheme using Neural Network is also presented for OFDM system to
select a suitable Modulation and Coding Scheme (MCS) for the BPLC channel realization. The proposed method of
Adaptive OFDM enhances the system throughput about 8% and reduces the feedback overhead compared to the
conventional schemes in BPLC system. The results show that the proposed adaptive algorithm with neural network channel
prediction scheme can achieve a considerable improvement in the processing time over the conventional method.