Adaptive demodulation method and apparatus using an artificial neural network to improve data recovery in high speed channels

Abstract

A neural network demodulator is used within a receiver to provide Inter Symbol Interference (ISI) channel equalization and to correct for I / Q / phase imbalance. The neural network is trained with a single integrated training step to simultaneously handle the channel impairments of ISI equalization and I / Q phase imbalance as opposed to prior art methods of separately addressing each channel impairment in sequence.

Description

RELATED APPLICATIONS[0001]This application is a continuation-in-part of U.S. patent application Ser. No. 14 / 312,072, filed Jun. 23, 2014, hereby incorporated herein by reference, and U.S. Provisional Patent Application No. 61 / 837,742, filed Jun. 21, 2013.FIELD OF THE INVENTION[0002]The present invention relates to a neural network based integrated demodulator that mitigates channel impairments, ISI and I / Q channel leakages, and minimizes the impact on the overall performance of the system. In this process, the inventive demodulator improves the performance for data recovery when operating at high data rates over the transmission channel.BACKGROUND OF THE INVENTION[0003]High-Speed communications systems typically use a wide band channel where the transmission is achieved using Radio Frequency (RF) carriers. RF transmission as it exists today either uses closely spaced narrow band multiple carriers or a small number of carriers containing baseband modulated signals.[0004]An example of...

AI Technical Summary

Benefits of technology

This patent describes a special demodulator that helps to improve the quality of information sent over a communication channel. It can handle I / Q imbalance and Inter Symbol Interference at the same time, and it does not need to be trained separately for each issue. This makes it much more efficient and effective for use in communication systems.

Problems solved by technology

In both of the above scenarios, the performance of high-speed digital transmission suffers high degradation due to the effects of channel impairments.

Specifically, the channel impairments, which include Inter-Symbol-Interference and the leakage of I / Q modulated signals, which are sent over each of the RF carriers, significantly degrades the Bit Error Rate (BER) performance.

In a small number of carrier based systems, both wireline and wireless channels experience leakage of I / Q due to imperfect phase imbalance.

As the transmit systems carry large data rates, the sensitivity to these channel impairments become significant.

Also, there could be bottlenecks in the design process to achieve optimal system performance, when these two impairments are handled sequentially, one after another.

Although a LMSE equalizer works well in a minimum phase channel, its performance is very limited in a channel with spectral nulls.

In such cases, the convergence of an LMS equalizer is not guaranteed and ISI effects cannot be minimized.

While the DFE outperforms the LMS, it is more complex than the LMS equalizer.

Furthermore the DFE suffers from an error propagation problem and therefore is only used at very high SNR scenario.

The MMSE, LMSE and DFE equalizers can only minimize the effect of ISI on the performance, but cannot handle the I / Q phase alignment problem.

In wireless channels, the channel equalization is more complex when handling ISI due to rapid changes in channel behavior because of mobility and channel fading.

Images