Search

Ricean fading is common in dense urban cellular networks and, as a mobile moves through that environment, the K-factor of the Ricean fading will change. This paper presents a statistical model for dense urban vehicular nonstationary Ricean fading, where the K-factor gradually changes due to movement through changing surroundings. This model is empirical and is based on K-factor fluctuations that are observed in dense urban cellular radio channel measurements. The K-factor is modeled using a random process with a distribution that is fit to the measured K-factor values. An autoregressive (AR) model is also used to ensure that the autocorrelation of the simulated K-factor process matches the empirical data. The nonstationary Ricean fading envelope that is generated using this model is verified by comparing it with the fading envelope that is observed in the measurements. Index Terms--Auto-regressive model, K-factor, radio channel measurement, Ricean fading.

This paper uses logfile data from commercial code division multiple access networks to quantify the base station power amplifier (PA) size reduction achieved with sector transmit power pooling. Sector transmit power pooling in a base station involves sharing the transmit power load of several sectors across the same bank of amplifiers. This transmit power trunking results in a reduction in required PA size. The logfile analysis used to estimate this reduction is augmented with simulated short-term power variations. A new constrained Poisson call arrival model is used to ensure that the simulated power variations accurately reflect actual network operating conditions. A Gamma random variable approximation is also used to improve simulation efficiency. The results of this analysis indicate that sector power pooling can significantly reduce required PA size for realistic network operating conditions. Index Terms--Code division multiple access, Gamma distributions, land mobile radio cellular systems, Poisson distributions, power amplifiers, sector transmit power pooling.

This paper proposes an approach to improve turbo code performance on the space time transmit diversity (STTD) code-division multiple-access (CDMA) forward link through better estimation of the channel state information (CSI) required by the mobile turbo decoder. An improved CSI estimation scheme that accounts for the unique nature of CDMA STTD intracell interference is presented. It is compared to a standard CSI estimation scheme that assumes all STTD interference and noise can be represented as additive white Gaussian noise at the mobile receiver input. An STTD CDMA forward link simulation shows that link performance is significantly improved when the mobile turbo decoder uses the more accurate CSI provided by improved estimation. However, the degree of improvement depends on the turbo decoding algorithm, mobile velocity, and whether the base station is using STTD or single-antenna transmission. Index Terms--Code-division multiple-access, space time block codes, transmit diversity, turbo coding.

This paper investigates how to improve channel decoder trellis path metric calculations when convolutional or turbo codes are combined with the chip-level linear minimum mean square error (LMMSE) code division multiple access (CDMA) mobile receiver. These metric calculations require channel state information (CSI) that consists of desired signal amplitude and interference plus noise variance. Several techniques suitable for use with the chip-level LMMSE receiver are presented for estimating CSI. Chip-level simulations are used to evaluate CDMA forward link performance when these techniques are used by the mobile. The results indicate that the channel state estimation scheme that provides the best forward link performance depends on the type of channel code and channel decoder algorithm used in the system. Index Terms--Channel state estimation, code division multiple access (CDMA), LMMSE equalization, Rake receiver, turbo codes.