Wireless ad-hoc networks have seen much attention in recent years, and are still a topic of much interest, especially as a result of an increasing need for ubiquitous connectivity. Ad-hoc networks have applications in tactical military systems as well as in commercial systems because of the numerous benefits they offer. There are still a number of open research questions regarding these networks, and this dissertation work addresses the question of which duplexing, multiplexing, and multiple access (D/M/MA) techniques are preferable in ad hoc networks. These techniques have substantial impact on network performance, yet surprisingly this particular topic has seen little attention. Moreover, most of the existing literature is focused on higher layers such as medium access control (MAC) and network layers, while our work concentrates on the physical (PHY) layer, which plays an important and fundamental role in network design.
In this dissertation, we investigate D/M/MA techniques in ad-hoc networks and more generally, how to allocate time/frequency resources to achieve better network performance. We provide a comparison of time, frequency, and time-frequency schemes in terms of a number of features, including duty cycles, required data rate per transmission, required transmit power per link, throughput, relative range and capacity. To keep the analyses tractable, topologies we study are full mesh networks, relay networks, and several special cases of ad-hoc networks. The main results show that for a peak power constraint, in terms of data rate, range, or capacity, continuous single-carrier waveforms are superior to bursted multi-carrier waveforms, and these schemes are attained with appropriate application of “hybrid” time-frequency division. Latency and throughput simulation results are provided for mesh networks, and analytical signal-to-noise-plus-interference ratio, multi-hop gain and simulation results for relay networks are also presented, to support our theoretically-based claims and aid in the network design.