AG Kommunikationstheorie


Performance of Relaying: Shannon Capacity Regime vs. Finite Blocklength Regime


Motivated by the increasing demand for higher throughput, broader coverage and higher-level quality of service (QoS), a major effort is being made to study the relaying technology in wireless networks. However, most existing research on investigating the relaying performance over wireless fading channels typically rely on simplifying assumptions, such as perfect channel state information (CSI) and infinite blocklength, which may be inaccurate in realistic wireless systems. In this dissertation, we contribute to developing more extensive and systematical investigation models of relaying. On one hand, both the physical-layer and the higherlayer performance models of relaying are developed while multiple performance metrics are taken into account, e.g., reliability, capacity and energy efficiency. On the other hand, various scenarios with different assumptions regarding CSI and blocklength (e.g., perfect or imperfect CSI and finite or infinite blocklengths) are considered. In general, the following fundamental questions are addressed based on our investigation models: Does relaying become more efficient or pay off less under these scenarios with more realistic assumptions (e.g., imperfect CSI or/and finite blocklengths)? What are performance differences if we do not consider or only consider one of these realistic assumptions? Is it essential to consider these realistic assumptions? What can we suggest for designing relaying systems especially for realistic applications? We achieve the above ultimate aim and find answers for these fundamental questions by two steps. In the first step, we investigate above performance metrics of relaying networks under the Shannon capacity regime (where blocklengths are infinite) and under the finite blocklength regime (where blocklengths are finite), respectively. In particular, both the perfect CSI scenario and the imperfect CSI scenario are considered. In other words, by this step relaying performance models are studied under different assumptions regarding CSI and blocklength, respectively. In the second step, we simulatively compare these relaying performance models under different scenarios/ assumptions. By these two steps, we finally observe some surprising performance advantages of relaying under the finite blocklength regime (in comparison to under the Shannon capacity regime) and conclude a set of important guidelines for the design of efficient relaying systems.

zurück zur Terminübersicht