Abstract:
This thesis investigates the deployment and performance of indoor/outdoor DS-CDMA systems assuming both conventional receivers and multiuser detection receivers at the base stations.
In the first part of this thesis, measured propagation data and a DS-CDMA system performance testbed (based on a Monte Carlo analysis) are used to quantify both the downlink and uplink outage probabilities for a variety of deployment scenarios: (a) indoor-only systems; (b) interfering indoor/outdoor systems; and (c) interfering indoor systems located adjacent to each other. The results reveal that optimal indoor deployment strategies are heavily dependent on both the location and strength of interference emanating from outdoors. Indoor base station deployment strategies that are optimal in the absence of outdoor interference are often suboptimal if outdoor interference is present.
In the second part of this thesis, two simple multiuser detection techniques are chosen for implementation in the DS-CDMA system performance testbed: successive interference cancellation (SIC) and parallel interference cancellation (PIC). The performance estimation results show that orders-of-magnitude improvements in uplink outage probability are possible with both SIC and PIC. The extent of these improvements is dependent on the deployment strategy used, the fractional residual interference cancellation error, and the uplink power control algorithm (for the case of SIC). If perfect interference cancellation is assumed, it is evident from the results of this thesis that the cancellation of signals in parallel (PIC) provides no additional benefit to system performance than ranking and cancelling the signals sequentially (SIC).