No abstract

Abstract-A general framework for the design of low complexity timing error detectors (TEDs) for orthogonal space-time block code (OSTBC) receivers is proposed. Specifically, we derive sufficient conditions for a difference-of-threshold-crossings timing error estimate to be robust to channel fading. General expressions for the S-curve, estimation error variance and the signal-to-noise ratio are also obtained. As the designed detectors inherently depend on the properties of the OSTBC under consideration, we derive and evaluate the properties of TEDs for a number of known codes. Simulations are used to assess the system performance with the proposed timing detectors incorporated into the receiver timing loop operating in tracking mode. While the theoretical derivations assume a receiver with perfect channel state information and symbol decisions, simulation results include performance for pilot-symbol-based channel estimation and data symbol detection errors. For the case of frequency-flat Rayleigh fading and QPSK modulation, symbol-error-rate results show timing synchronization loss of less than 0.3 dB for practical timing offsets. In addition it is shown that the receiver is able to track timing drift with a normalized bandwidth of up to 0.001.

The application of selection diversity in conjunction with simple channel coding is considered for a multiuser, slowly fading, Spread‐Spectrum Multiple Access (SSMA), digital radio system. For the most part, the index of performance for our study is the average bit error probability; we also give some consideration to multipath outage as a performance measure. All subscribers are assumed to communicate to a central station; that is, a star network architecture is assumed. Average power control is also assumed. The average mentioned in this context includes averaging over the channel fading statistics. The modulation is direct‐sequence, spread‐spectrum, binary phase‐shift keying. We assume perfect timing and carrier recovery in our coherent receiver, and a slowly varying, Rayleigh fading, discrete multipath model is used. Previous analyses have found that SSMA can tolerate few simultaneous users for fading radio channels. We find that the combination of spread‐spectrum modulation with low‐complexity diversity and/or channel coding can restore fading‐channel user levels to an acceptable figure. In addition, selection diversity plus channel coding is more effective than either method by itself. Finally, it turns out that SSMA is less sensitive to a change in the value of delay spread of a fading channel than, say, time‐division multiple access. The method of moments is used to accurately assess the system error probability. Using this technique, we also assess the accuracy of assuming that the multiuser interference has a Gaussian distribution, which allows it to be analyzed by a simple method. Using this assumption, we compare selection diversity plus channel coding with the maximal‐ratio‐combining technique for diversity reception. Except for a high order of diversity, the former is more efficient and is always less complex than the latter.

No abstract

No abstract