Jukka Henriksson

Learn More
A neural convolutional decoder which exploits the channel information is introduced. The method uses a recurrent neural network, tailored to the used convolutional code and the channel model. No supervision besides possible channel estimation is required. Also, no distinct equalizer is needed. As an example, we show the structure of the neural decoder for(More)
This paper describes different techniques for mitigation of high power and long duration interference in terrestrial digital video broadcasting (DVB-T/H) systems. The interference is assumed to distort the DVB-T/H signal severely so that on average one fourth of the DVB-T/H symbol is lost. We investigate the use of long enough time interleaver in(More)
IN DYNAMIC DISCRETE-SIGNAL DETECTION Teuvo Kohonen , Kimmo Raivio , Olli Simula , Olli Ventä , and Jukka Henriksson+ ) Helsinki University of Technology Laboratory of Computer and Information Science TKK-F, Rakentajanaukio 2 C, SF-02150 Espoo, Finland +) Nokia Research Center, Transmission Systems P.O. Box 156, SF-02101 Espoo, Finland ABSTRACT Linear(More)
Pan-European project B21C (Broadcasting for the 21st Century) aims to develop technology for DVB-T2 (Digital Video Broadcasting, Terrestrial), which is a spectrum efficient broadcasting system for future. The evaluated technologies include MIMO, which has been widely studied in perspective of B3G (Beyond 3rd Generation) systems, but never for any UHF band(More)
Novel receiver structures combining traditional transversal equalizers and neural networks have been introduced for adaptive discrete-signal detection to improve the equalizer performance especially in compensating nonlinear distortions. In addition to noise and nonlinearities, various interfering signals may be present. In this paper, the behavior of the(More)
Recently, novel equalizer structures combining traditional transversal equalizers and neural networks have been introduced for adaptive discrete-signal detection. It has been shown that the equalizer performance can be improved using neural networks, especially in compensating nonlinear distortions. In addition to noise and nonlinear distortions various(More)
This paper discusses the performance of neural receiver structures in fading (frozen) multipath channels. In addition to noise, non-Gaussian interference is present in system. The modulation under study has been 16QAM. Especially, nonlinear channel models have been investigated. The performance of two receiver structures based on the Self-Organizing Map(More)
Real communication channels with multipath propagation, interference and possible nonlinearities pose a difficult problem to the detecting receiver. This paper deals with neural approaches to solve those difficulties. Two types of neural networks, self-organizing map and radial basis functions have been studied. The results show that, while there are no(More)