Abstract—Emergent telecommunication systems use spectral efficient multilevel modulation formats together with multicarrier schemes such as Orthogonal Frequency division multiplexing (OFDM). OFDM transmission is an efficient way to deal with multipath. However one of its major drawbacks is its sensitivity to nonlinear distortions due to its greatly variable envelope and high peak to average power ratio (PAPR). Nonlinear distortions are mainly introduced by high power amplifiers (HPA), cause in band distortions. Misalignments in IQ modulators and demodulators causes imbalance, which results in loss of orthogonality and create intercarrier interference and spectral regrowth in an OFDM system. Nonlinear distortions also cause intermodulation effect. This degrades Bit error rate (BER) performance of the system. This paper introduces the self organization map(SOM) and parameterless self organization map(PLSOM) based and Hiperlan/2 receiver structure used for the compensation of nonlinear distortions. Simulation results presented in this paper clearly indicate the improved performance of the proposed system. Simulations are run for AWGN channel model and for two HPA models namely Travelling wave tube amplifier (TWTA) and Solid state power amplifier (SSPA), with and without SOM and PLSOM neural network blocks.
Index Terms—OFDM; HPA; TWTA; SSPA; BER; AWGN; SOM; PLSOM
S. Rathkanthiwar is with Yeshwantrao Chavan College of Engineering,
Nagpur, India, as an Associate Professor in Department of Electronics
Engineering (e-mail: email@example.com).
C. Dethe is working as Principal, in Priyadarshini Institute of Enginnering and Technology, Nagpur, India. (e-mail: firstname.lastname@example.org).
K. Kulat is with Vishvesaraya National Institute of Technology, and is working as Professor and Head of Electronics Engineering Department. (e-mail: email@example.com)
Cite: Shubhangi Rathkanthiwar, Chandrashekhar Dethe, and Kishore Kulat, "A Neural Approach for Compensation of Effects of PAPR Causing BER Degradations in Hiperlan2," International Journal of Modeling and Optimization vol. 2, no. 3, pp. 365-370, 2012.