| This dissertation presents methods to approach throughput maximum and latency minimum in dense wireless Ad Hoc networks. For throughput maximization, we propose an idle probability based broadcasting method, iPro, which employs an adaptive probabilistic mechanism to improve performance of data broadcasting over dense Wireless Ad Hoc Networks. In one-hop broadcasting scenarios, the modeling and simulation results of the proposed iPro are shown to be consistent and both significantly outperform the standard IEEE 802.11 under saturated conditions. Moreover, without estimating the number of competing nodes and changing the contention window size, the proposed iPro can still approach the theoretical throughput bound. We further apply iPro to multi-hop broadcasting scenarios, and the experiment results show that, with the same elapsed time after the broadcasting from the sources, the proposed iPro has significantly higher Packet Delivery Ratios (PDR) than traditional methods.;For latency minimization, we analyze latency performance based on p-Persistence CSMA, and show that when the number of competing nodes is much larger than the number of messages a node requests to receive from neighbors and the packet, length is a lot longer than the slot time, the optimal transmission probability for latency minimized p-Persistent CSMA (LMPP) in an event-driven scenario equals the optimal transmission probability for throughput maximization in a saturated wireless local area network. Instead of creating a completely new protocol, we further propose Latency Minimized Probabilistic CSMA/CA (LaMP), which is a multiple access scheme with constant contention window size built atop standard IEEE 802.11 CSMA/CA, to approach the performance of LMPP. However, when the number of competing nodes is unknown, a quick and reliable estimate of the number of competing nodes is highly desired to apply the proposed method and achieve minimize latency. Therefore, we propose an Exponentially Decreased Matching (EDeM) strategy to estimate the number of competing nodes and approach the minimum latency. Simulation results show that the proposed method consistently outperforms existing methods in latency performance. |
