2015/03/20 by Fanny Parzysz, Parzysz, Fanny, Mai Vu +3
Computer Science · Engineering · #Advanced MIMO Systems Optimization #Cooperative Communication and Network Coding #FOS: Computer and information sciences #Full-Duplex Wireless Communications #Information Theory (cs.IT)
paper · pdf · doi:10.48550/arxiv.1503.06242
openalex publication_date 2015/03/20 · openalex created_date 2022/08/07 · openalex updated_date 2026/07/28
By relying on a wireless backhaul link, relay stations enhance the\nperformance of cellular networks at low infrastructure cost, but at the same\ntime, they can aggravate the interference issue. In this paper, we analyze for\nseveral relay coding schemes the maximum energy gain provided by a relay,\ntaking into account the additional relay-generated interference to neighboring\ncells. First, we define spatial areas for relaying efficiency in log-normal\nshadowing environments and propose three easily-computable and tractable\nmodels. These models allow the prediction of 1) the probability of\nenergy-efficient relaying, 2) the spatial distribution of energy consumption\nwithin a cell and 3) the average interference generated by relays. Second, we\ndefine a new performance metric that jointly captures both aspects of energy\nand interference, and characterize the optimal number and location of relays.\nThese results are obtainable with significantly lower complexity and execution\ntime when applying the proposed models as compared to system simulations.We\nhighlight that energy-efficient relay deployment does not necessarily lead to\ninterference reduction and conversely, an interference-aware deployment is\nsuboptimal in the energy consumption. We then propose a map showing the optimal\nutilization of relay coding schemes across a cell. This map combines two-hop\nrelaying and energy-optimized partial decode-forward as a function of their\nrespective circuitry consumption. Such a combination not only alleviates the\ninterference issue, but also leads to a reduction in the number of relays\nrequired for the same performance.\n