A characteristic function-based method is used to derive closed-form bit error probability (BEP) expressions for orthogonal frequency-division multiplexing (OFDM) systems in the presence of channel estimation error over frequency-selective Rayleigh fading channels and frequency-selective Ricean fading channels. Both single channel reception and diversity reception with maximal ratio combining (MRC) are examined. The BEP expressions are shown to be sums of several conditional probability functions which can be calculated by using proper complex Gaussian random variable theory and a characteristic function method. The closed-form BEP expressions can be used to accurately investigate the bit error rate performance degradation caused by channel estimation error under different wireless channel environment models. The performances of two interpolation methods, a sine interpolator with Hamming windowing and a Wiener interpolator, are compared.

Recent published work suggests that correlative coding can improve OFDM performance by several dB. This work, however, does not consider fully a symbol error rate performance measure for the OFDM systems. Exploiting an equivalence between correlative coding and partial-response signaling, a bit error rate performance measure is employed to assess the performance of OFDM systems using correlative coding. It is shown that although partial-response pulse-shaping or correlative coding can reduce intercarrier interference due to carrier frequency offset, they do not necessarily improve the system bit error rate performance because of the introduction of multilevel signaling with associated reduced receiver decision distance. It is observed that bit error rate performance improvement for symbol-by-symbol detection can be achieved only in some cases of large frequency offset. In the case of smaller, practical, values of frequency offset the bit error rate is increased by correlative coding or partial-response pulse-shaping.

It is proved that the intercarrier interference matrix in orthogonal frequency division multiplexing systems is unitary for an arbitrary value of frequency offset and any number of subcarriers. The benefit of a recently published precoding scheme in practical applications is thereby established.

An exact bit error rate (BER) expression for an OFDM system with windowing reception is derived. The effects of several Nyquist windows, including the raised-cosine window, the "better than" raised-cosine (BTRC) window, the second order continuity window (SOCW), the Franks window, and the double-jump window on the performance of the system are examined based on both BER measure and signal-to-interference (SIR) measure. It is found that the SIR comparison is not necessarily consistent with the BER comparison. The BTRC windowing gives the smallest BER among the five windowing functions considered for small to medium values of pulse roll-off factor. However, the Franks windowing or SOCW windowing shows better BER performance when the roll-off factor approaches one

An exact closed-form bit error rate (BER) expression is derived for an orthogonal frequency-division multiplexing system with π/4-shifted differentially encoded quadrature phase shift keying (π/4-DQPSK) in the presence of carrier frequency offset over frequency-selective fast Rayleigh fading channels. Different system configurations, including time domain differential modulation, frequency domain differential modulation, single channel reception, and multi-channel reception with maximal ratio combining diversity, are considered in the exact BER analysis. For a small number of subcarriers, the BER expression can be calculated directly. A Monte Carlo method is designed to evaluate the BER for a large number of subcarriers. The analytical expression can be used to investigate the effect of several channel parameters, including mean delay spread and maximum Doppler spread, on the system BER performance. Particularly, the effect of carrier frequency offset on the system performance, and the differences and opportunities to use frequency domain differential modulation or time domain differential modulation, can be studied in a quantitative way for a more realistic wireless channel environment model. By using the exact closed-form BER expression, an optimum number of subcarriers can be found, and the allowable carrier frequency offset, Doppler spread, and mean delay spread for given operating conditions can be determined.

A characteristic function-based method is used to derive closed-form bit error probability (BEP) expressions for orthogonal frequency-division multiplexing (OFDM) systems in the presence of channel estimation error over frequency-selective Rayleigh fading channels and frequency-selective Ricean fading channels. Both single channel reception and diversity reception with maximal ratio combining (MRC) are examined. The BEP expressions are shown to sums of several conditional probability functions which can be calculated by using proper complex Gaussian random variable theory and a characteristic function method. The closed-form BEP expressions can be used to accurately investigate the bit error rate performance degradation caused by channel estimation error under different wireless channel environment models. The performances of two interpolation methods, a sinc interpolator with Hamming windowing and a Wiener interpolator, are compared.

A precise method for calculating the bit-error probability (BEP) of a discrete cosine transform (DCT)-based orthogonal frequency-division multiplexing (OFDM) system on additive white Gaussian noise (AWGN) channels in the presence of frequency offset is derived. These accurate results are used to examine and compare the BEP performance of a DCT-OFDM system and the conventional discrete Fourier transform (DFT)-based OFDM system in an AWGN environment. Several signaling formats, such as binary phase-shift keying, quaternary phase-shift keying, and 16-ary quadrature amplitude modulation are considered. The performance of a DCT-OFDM with a zero-padding guard-interval scheme is then compared with a zero-padded DFT-OFDM with the employment of minimum mean-square error (MMSE) detection and MMSE decision-feedback detection with ordering scheme over frequency-selective fast Rayleigh fading channels. Analysis and simulation results show that the DCT-OFDM system outperforms the DFT-OFDM system in the presence of frequency offset, and in frequency-selective fast-fading environments

The effects of several Nyquist windows, including the raised-cosine window, the "better than" raised-cosine window, the second order continuity window, and the Franks window on the performance of an OFDM windowing reception system are considered. The intercarrier interference analysis and bit error rate calculation of this system are provided

A precise method for calculating the bit error probability of a discrete cosine transform (DCT)-based orthogonal frequency-division multiplexing (OFDM) system on AWGN channels in the presence of frequency offset is derived. These accurate results are used to examine and compare the bit error probability performances of a DCT-OFDM system and the conventional discrete Fourier transform (DFT)-based OFDM system. Several signaling formats, such as binary phase shift keying (BPSK), quaternary phase shift keying (QPSK), and 16-ary quadrature amplitude modulation (16-QAM) are considered. Analysis and simulation results show that the DCT-OFDM system outperforms the DFT-OFDM system in the presence of carrier frequency offset.

An exact closed-form bit error rate expression is derived for an orthogonal frequency-division multiplexing system, each subcarrier modulated by /spl pi//4-shifted differentially encoded quadrature phase shift keying, in the presence of carrier frequency offset over frequency-selective fast Rayleigh fading channels. Both single channel reception and diversity reception with maximal ratio combining are considered. The closed-form expression allows us to investigate the effect of several channel parameters, including mean delay spread and maximum Doppler shift, on the system bit error rate performance. Particularly, the effect of carrier frequency offset on the system performance can be studied for a more realistic wireless channel environment model. Small carrier frequency offsets have slight influences on the system performance. Compared with the same amount of Doppler shift, carrier frequency offset usually leads to a greater degree of performance degradation. Systems with fewer subcarriers are more sensitive to the channel mean delay spread. By using the exact closed-form bit error rate expression, an optimum number of subcarriers can be found, and the allowable carrier frequency offset, Doppler shift, and mean delay spread for given operating conditions can be determined.

A discrete cosine transform (DCT)-based orthogonal frequency-division multiplexing (OFDM) system is described. The performance of the DCT-OFDM system with a zero-padding guard interval scheme and a minimum mean-square error (MMSE) receiver over time-varying multipath Rayleigh fading channels is investigated. The results show that employing the proposed DCT-OFDM system rather than the conventional DFT-OFDM system can provide better bit error rate performance in practical systems, by as much as 5.5 dB in signal-to-noise ratio (SNR).

An exact method for calculating the bit error rate of pulse-shaped orthogonal frequency-division multiplexing systems in the presence of frequency offset is derived. This method represents a unified way to calculate the bit error rate of different one-dimensional and two-dimensional subcarrier modulation formats, such as, binary phase shift keying (BPSK), quaternary phase shift keying (QPSK), and 16-ary quadrature amplitude modulation (16-QAM). The precise bit error rate expressions are obtained using a characteristic function method. These accurate analytical results are used to examine and compare the bit error rate performance of pulse-shaped orthogonal frequency-division multiplexing systems employing the rectangular pulse, the raised-cosine pulse, and the "better than" raised-cosine pulse when a frequency offset is present. The dependence of the bit error rate on the roll-off factor of the pulse employed for a specific system in the presence of frequency offset is also investigated. Analysis and numerical results show that employing the "better than" raised-cosine pulse rather than the raised-cosine pulse can provide better bit error rate performance in practical systems, by as much as 2-3 dB in SNR.