Distributed interference management under low network visibility : theory and applications

Thumbnail Image
Fadel, Maha
Issue Date
Electronic thesis
Electrical engineering
Research Projects
Organizational Units
Journal Issue
Alternative Title
Reliable communication in wireless networks currently relies on the perfect estimation of the varying state of the wireless channel at all communicating nodes. While the next generation of wireless networks promises ultra-reliable massive connectivity, such a requirement is expected to result in prohibitive communication overhead and delay. This thesis proposes novel transmission schemes that alleviate the limitations imposed by requiring the network state to be instantaneously and globally available. Especially, we present novel non-orthogonal multiaccess schemes adapted to fading multi-user channels with partial channel state information at the transmitters (CSIT).
literature with lower encoding and decoding complexity. Moreover, the proposed policy is shown to achieve the sum-rate capacity of the two-user multiple access channel with full CSIT asymptotically. Broadcast approaches are proposed for the two-user Gaussian interference channel without and with partial CSIT. The average achievable rate regions of the proposed approaches are characterized and compared to upper bounds on the average capacity region. Under the assumption of no CSIT, the proposed approach is shown to achieve average sum-rates within a constant gap from the sum-rate capacity in the asymptote of the high signal-to-noise ratio. Finally, a novel outer bound on the average capacity region of the Gaussian interference channel without CSIT is proposed.
Moving to the multi-user setting, a broadcast approach is proposed for two-user multiple access channel with local CSIT and the corresponding average achievable rate region is characterized. The proposed approach is shown to outperform existing schemes in the
For the single-user energy harvesting channel with no CSIT, we characterize the optimal power allocation policies that maximize the average achievable rate or minimize the outage probability under the assumption of non-causally known harvested energy arrivals at the transmitter. Further, online power allocation policies are proposed for the case of causally known harvested energy arrivals. In addition to the energy harvesting channel model, an algorithm is proposed that optimally allocates available resources over time under sequential resource access for any convex objective function.
access channel, and the interference channel models. These channel models represent abstract forms of wireless networks enabling a variety of applications in next-generation wireless networks.
We characterize the performance limits of the proposed schemes from an information-theoretic perspective by focusing on the slowly fading Gaussian channel model with partially available CSIT. We focus on the single-user channel with energy harvesting, the multiple
Deviating from the objective of solely maximizing the transmission rates towards enhancing the latency and fairness among users, we consider a novel non-orthogonal multiaccess approach based on multi-layer superposition coding for the general N-user multiple access channel with no CSIT. For the case of deterministic data arrivals, lower and upper bounds on the average queuing latency at each transmitter are characterized. Additionally, for the case of Poisson data arrivals, a closed-form expression is derived for the average queuing latency at each transmitter. Using the characterized bounds, the proposed approach is shown to outperform the conventional outage approaches in terms of minimizing the average achievable latency. Finally, a measure of fairness among users is considered where it is shown that the proposed approach outperforms the outage approach in achieving user fairness when facing CSIT uncertainties.
August 2020
School of Engineering
Full Citation
Rensselaer Polytechnic Institute, Troy, NY
PubMed ID