This paper studies several algorithms for constructing quasi-optimal GMW subfamilies in terms of minimized bit error rate under co-channel interference. The results show that performance of resultant sub-families is very sensitive to the algorithms applied and sub-family sizes. A new criterion based on combinational (even plus odd) maximum cross-correlation is introduced for code selection and the resultant highest-peak-deleting and most-peak-deleting algorithms are effective to construct the GMW quasi-optimal sub-families.

Some CDMA systems serve a finite number of simultaneously active users. Thus, its performance is spreading code dependent and Gaussian approximation is not appropriate to evaluate its performance. To choose proper spreading codes for a finite CDMA system, both peak cross-correlation and squared-sum of cross-correlations should be considered. An algorithm based on multiple variable Bernoulli process is introduced to evaluate the BER due to co-channel interference. The algorithm is useful for spreading code dependent capacity analysis, regardless of the CDMA system size and operation. The results show that small Kasami and Gold (1968) like codes offer a high capacity in a finite CDMA system.

A bandwidth efficient quadrature-overlapped modulation, MSK raised cosine (MSKRC) modulation, is introduced. Its power spectral density (PSD) is the product of PSDs of MSK and quadrature overlapping raised cosine (QORC) signals. The PSD main lobe of the MSKRC signal is the same as that of the MSK signal but the PSD sidelobes of the MSKRC signal diminish approximately as fast as 1/f/sup 10/, much faster than those of OQPSK, MSK and QORC signals (with 1/f/sup 2/, 1/f/sup 4/ and 1/f/sup 6/, respectively). The intersymbol interference (ISI) performance for the MSKRC modulation system is analyzed. The analysis reveals that the MSKRC signal produces a lower ISI level than other reported modulated signals.>

A new collision-avoiding protocol, the receiver-and-transmitter (R&T) code-sensing protocol, is formulated and its performance upper bounds are numerically analyzed. The results reveal that a high throughput and a stable performance of a distributed code-division multiple-access (CDMA) packet radio network can be achieved by broadcasting both transmitter and receiver codes into the shared CDMA channel to avoid possible packet conflicts. To measure the efficiency of the orthogonal codes used, the effective code utilization (ECU), defined as the ratio between the throughput and code consumption, is introduced and evaluated. The evaluation shows that the R&T code-sensing protocol has an ideal ECU.>