Graph-Based Multicell Scheduling in OFDMA-Based Small Cell Networks

Abstract

This paper proposes a novel graph-based multicell scheduling framework to efficiently mitigate downlink intercell interference in OFDMA-based small cell networks. We define a graph-based optimization framework based on interference condition between any two users in the network assuming they are served on similar resources. Furthermore, we prove that the proposed framework obtains a tight lower bound for conventional weighted sum-rate maximization problem in practical scenarios. Thereafter, we decompose the optimization problem into dynamic graph-partitioning-based subproblems across different subchannels and provide an optimal solution using branch-and-cut approach. Subsequently, due to high complexity of the solution, we propose heuristic algorithms that display near optimal performance. At the final stage, we apply cluster-based resource allocation per subchannel to find candidate users with maximum total weighted sum-rate. A case study on networked small cells is also presented with simulation results showing a significant improvement over the state-of-the-art multicell scheduling benchmarks in terms of outage probability as well as average cell throughput

Load-Aware Dynamic Spectrum Access for Small-Cell Networks: A Graphical Game Approach

“…Due to the spatial distribution and lower transmission power of SAPs, the transmission of a small cell only directly affects the neighboring small cells [4], [13], [14], [18]. To characterize the interference relationship among the small cells, the following interference graph is introduced.…”

“…Based on (14) and (15), we have Φ(a * n , a −n ) − Φ(a n , a −n ) = u n (a * n , a −n ) − u n (a n , a −n ),…”

Interference management enablers for 5G radio access networks

“…For the rest, coordinated scheduling is applied, where dynamic frequency partitioning (or muting of resources for some NNs) is performed. The dynamic frequency partitioning that is used in this study is based on [7] . Figure 3 shows an exemplary system model where a number of nomadic nodes are activated in hotspots to enhance capacity and coverage and also offload traffic from the macro.…”

“…In [7] it was proposed a mechanism to reduce the energy consumption in a non-fully loaded network, controlling a cluster of cells with a centralized scheduler that could switch off certain nodes, and at the same time exploit Joint Transmission (JT) and Dynamic Point Selection/Dynamic Point Blanking (DPS/DPB) on the remaining ones in order to address the traffic needs of the active users. The overall energy savings that could be achieved where analyzed considering power models proposed in the EARTH [11] project for macro and micro Base Stations (BS), considering equipment capabilities in 2010.…”

“…The 5GREEN project [12] estimated that an improvement of 8% every year can be achieved in the dynamic part of power models, so that the overall power consumption will scale more significantly with the actual radiated power every year, and also "sleep" mechanism in the nodes will become more and more efficient (see Figure 6). Figure 6 shows the impact of power models suitable for 2020 transmission nodes considering the proposed scheme, comparing them to results that were reported in [7] with 2010 power models. Results shown in Figure 7 are obtained under the simplified Madrid Grid scenario described in [7].…”