Multiple channel crosstalk removal using limited connectivity neural networks

Abstract

Limited connectivity neural network architectures are investigated for the removal of crosstalk in systems using mutually overlapping sub-channels for the communication of multiple signals, either analogue or digital. The crosstalk error is modelled such that a fixed proportion of the signals in adjacent channels is added to the main signal. Different types of neural networks, trained using gradient descent algorithms, are tested as to their suitability for reducing the errors caused by a combination of crosstalk and additional gaussian noise. In particular we propose a single layer limited connectivity neural network since it promises to be the most easily implemented in hardware. A variable gain neuron structure is described which can be used for both analogue and digital data.

Spring drought prediction based on winter NAO and global SST in Portugal

“…hidden neurons experiencing difficulty learning, since they are attempting to model at least two functions at once), and the best results were obtained with the training of separate networks for each output (SPI6 April , SPI6 May and SPI6 June ) at each rain gauge. Problems with 'cross-talk' have been reported in other studies, including those of MacGregor and Gerstein (1991) and Craven et al (1996).…”

“…Problems with ‘cross‐talk’ have been reported in other studies, including those of MacGregor and Gerstein () and Craven et al . ().…”

Optimization integrated generative adversarial network for occluded text recognition with language modeling

“…The technique of image analysis actively involves the text recognition 10 in the graphical documents. With the broader position of view, the text recognition from image plays a major research topic in computer vision tasks of various appliances, namely automatic documents digitization, augmented reality, real time multi language translation, and supporting to the blind people 11–13 …”