Peak to Average Power Ratio Reduction and Bit Error Rate Improvement in Wireless Orthogonal Frequency Division Multiplexing Communication Systems

• Main Author: Al-Azzo, Wisam Fahmi
• Format: Thesis
• Language: English; English
• Published: 2009
• Online Access: http://psasir.upm.edu.my/id/eprint/7363/1/FK_2009_52a.pdf; http://psasir.upm.edu.my/id/eprint/7363/
• Institution: Universiti Putra Malaysia

Orthogonal frequency division multiplexing (OFDM) offers high data rate transmission with high spectral efficiency, immunity to multipath fading, and simple implementation using fast Fourier transform (FFT). OFDM is readily implemented by present day processors in many high speed networks. However, one of the major drawbacks of OFDM systems is the high peak-to-average power ratio (PAPR); this can result in poor power efficiency, degradation in bit-error-rate (BER) performance, and spectral spreading. The effective PAPR reduction of OFDM signals by simple processing has been a challenge for the limited power and processing capability of portable OFDM applications. This thesis investigates the problem of high PAPR in OFDM systems and presents many simple implementation PAPR reduction techniques, and one error-resilient technique. The first part of this thesis presents two time-domain PAPR reduction techniques, viz, square-rooting the envelope of the OFDM output signals, and the smoothing technique. The square-rooting process changes the statistical distribution of the OFDM output signals from Rayleigh to Gaussian-like distribution and reduces the differences between the values of peak and average power, which consequently reduces the PAPR significantly. About 6 dB reduction in PAPR is achieved with moderate degradation in BER performance. For the smoothing process, which is derived from the image enhancement technique, the smoothing applied on the OFDM signals mitigates the PAPR due to its averaging effect. Up to 2.5 dB reduction is achieved by smoothing. Two new probabilistic based non-iterative frequency-domain PAPR reduction techniques are introduced in the second part of the thesis. These techniques reduce PAPR by changing the statistical distribution of the OFDM modulated symbols from uniform distribution to Gaussian-like distribution. This task is performed by two different methods in two different PAPR techniques. The first method of PAPR reduction is done by the addition of complex Gaussian random signals, while the second one is done by insertion of dummy Gaussian subcarriers. The two techniques provide PAPR reduction in the order of 5 dB for PSK-OFDM systems with no out-of-band radiation. The adaptive operation of these techniques enhances significantly both the BER performance and reduce the transmission power. The last part of this thesis presents a new modulation-based error resilient technique referred to as multi-dimensional modulation technique (MDM). In this technique concatenation of digital modulators of decreasing modulation orders are employed. The MDM technique improves the BER performance linearly with increased size of modulation order; up to 12 dB improvement in Eb/No ratio is achieved relative to the conventional OFDM systems at high modulation orders, M≥1024. Also, the MDM technique offers both error resilience and PAPR reduction when it is combined with the conventional OFDM systems in time domain. As a conclusion, the proposed techniques described above offer new solutions to the problem of high PAPR in OFDM systems, and for one of them offer improvement of BER performance at the same time. Besides, they can be applied for different systems parameters and applications requirements. Moreover, the PAPR reduction techniques proposed in this thesis are data-independent and can be implemented in one-shot; while the MDM technique uses only digital modulation and dc-offset signal processing, which can be implemented by simple circuits and/or processors.