Throughput Analysis of a Decentralized RLAN Based on Asynchronous DS/CDMA Using Random Periodic Spreading Sequences

Packet throughput figures are obtained for a decentralized radio local area network (RLAN) which is based on asynchronous direct-sequence code-division multiple-access (DS/CDMA). Packets arrive at the receiver nodes with different power levels. Techniques are developed to derive the probability of packet success for a system employing random periodic spreading sequences. It can be shown, that this system performs far better than a network using random spreading sequences. Based on the packet error probability, throughput figures of slotted DS/CDMA-ALOHA are presented for various scenarios. The effect of applying forward error correction (FEC) is investigated. For finite user environments, additional sources of errors have to be considered. These have a major impact and reduce the overall system performance. Finally, the throughput figures of a system applying the binary exponential backoff algorithm to avoid unstable behavior is investigated. The performance figures of the various systems described in this paper show that DS/CDMA forms a valuable choice for future RLANs.