This paper provides a low-complexity scheme for peak-to-average power ratio (PAPR) reduction without side information for frequency switched transmit diversity orthogonal frequency division multiplexing (FSTD-OFDM). The proposed scheme generates multiple signal candidates (SCs) by dividing parallel data symbols into subparts and mapping the subparts onto the antennas under a criterion reducing the correlation between the SCs to select one SC with minimum PAPR. The proposed scheme is shown to provide reasonably good PAPR performance at a lower complexity than the conventional schemes. In addition, the proposed scheme can detect the transmitted SC via energy detection without side information.

This paper introduces a reconfigurable mixed frame structure and a resource allocation algorithm for OFDMA systems supporting both the band-AMC and diversity modes in the frame. To reduce the total feedback without significant performance loss, we assign a user to the diversity subchannel when the band-AMC mode has no gain over the diversity mode based on the frequency selectivity of the user. After user selection, we apply joint power and rate allocation to exploit the asymmetry in required transmit power of the band-AMC mode and the diversity mode. The results reveal that the mixed frame with simple user classification provides a performance comparable to the frame with band-AMC mode only at a much reduced feedback. In particular, it is observed that the mixed frame with frequency division is preferable to that with time division by increasing the throughput of the users at the cell boundary in a large cell.

Since the received signal levels in mobile communications are unknown and the location is varying, acquisition schemes for pseudonoise (PN) sequences with fixed thresholds cannot provide satisfactory performance. This fixed-threshold scheme may cause too many false alarms or result in a low detection probability for a selected threshold value. We present an adaptive acquisition scheme for PN sequences which estimate the background power level, multiply it with a threshold coefficient to keep the false alarms constant, and use it as a threshold.

A new channel estimator, termed as the sinc interpolator, based on the pilot symbol assisted modulation channel sounding is proposed for Rayleigh fading channels. This method uses sinc function to estimate channel state information. A Wiener filtering estimator known to have optimum performance needs a priori information of autocorrelation function of the channel gain, Doppler frequency and signal-to-noise ratio to optimize filter coefficients. But the proposed method does not need a priori information while its performance is similar to that of the Wiener filtering estimator.

This paper reports experimental results of a pilot symbol aided 16QAM system for land mobile communications at 1.9 GHz. To combat the multipath fading phenomena in Rayleigh fading channels, the pilot symbol assisted modulation (PSAM) channel sounding technique, trellis coded modulation (TCM), and space diversity are employed. The TCM coder, with 8 states, has the length of the shortest error event path of 2 and the product of the branch distances along that path of d/sub pmin//sup 2/=0.8. We have also designed a two branch MRC space diversity receiver in which each branch has a channel estimation capability. As a result of field experimental measurements, we have obtained the BER=6/spl times/10/sup -4/ at E/sub b//N/sub o/=20 dB in a 16QAM system with diversity and TCM.

This paper analyzes the bit error rate (BER) for quadrature amplitude modulation (QAM) with L-fold maximal ratio combining (MRC) space diversity in Rayleigh fading channel. The formula is obtained by averaging the bit error probability of M-ary QAM in an additive white Gaussian noise channel over a chi-square distribution with 2L degrees of freedom. This formula overcomes the limitations of the earlier work, which has been limited only to deriving a BER for QAM with two branch MRC space diversity. The BER performance of quadrature phase shift keying with MRC diversity is obtained from the derived formula by setting M=4.