Study On Interference Management In Heterogeneous Cellular Networks

The explosive growth of data services drives the evolution of cellular mobile network architecture from the previous single tier networks to the future multitier heterogeneous networks(Het Nets).Within the coverage area of conventional macrocells,picocells and femtocells are deployed to provide high-rate access services for outdoor and indoor hot zones,respectively.Due to the scarcity of spectrum resources,multitier Het Nets are expected to use the universal frequency reuse.However,the universal frequency reuse will cause severe co-channel interference.Hence,it is critical to design an effective interference management solution for the successful deployment.The interference in Het Nets is categorized into two types: co-tier interference(e.g.interference among macrocells and interference among femtocells)and cross-tier interference(e.g.interference between macrocells and femtocells).The heterogeneous characteristics are as follows: transmit power(or coverage area),number of antennas,network topology and wireline backhaul.For example,macrocells are deployed by mobile operators in a planning manner and connected via optical fibers;femtocells are deployed by consumers in a random and dense manner and connected with macrocells through the IP-based wireline backhaul.Such difference on the network architecture will lead to different requirements for the interference management.Since this IP-based backhaul has a large delay,the real-time information exchange is impossible.Hence,the conventional joint resource optimization approach cannot be applied to femtocells.Considering these characteristics of Het Nets,this dissertation proposes a series of interference management schemes to mitigate the interference in Het Nets.First,for the co-tier interference issue among macrocells,a distributed interference coordination mechanism for macrocells is presented by exploiting Almost Blank sub-Channel(ABC).Second,for the co-tier interference issue among femtocells,a primary sub-channel self-configuration algorithm based on game theory is proposed.Third,for the interference management issue of device-to-device(D2D)communications,a Qo S-driven resource allocation scheme is developed for joint admission control and resource allocation.At last,we propose a cross-tier interference mitigation scheme by exploiting massive antennas equipped at macro base stations(MBSs).The research works are supported by National Natural Science Foundation of China(No.61231008);State Key Laboratory Foundation(ISN1002005,ISN090305);111 Project(No.B08038);Program for Changjiang Scholars and Innovative Research Team in University(No.IRT0852).The main achievements and results of this dissertation are listed as follows:1.The static soft frequency reuse(SFR)served as the cell-centric inter-cell interference coordination(ICIC)may significantly degrade the cell throughput,especially for the case of load imbalance among neighboring cells.To address this issue,a distributed ICIC via dynamically scheduling ABCs is proposed.According to the realistic interference condition,each cell adaptively decides the number of ABCs scheduled.Considering the coupling relationship of resource allocation among neighboring cells,we design ABC mapping patterns to minimize the number of ABCs scheduled in each cell.Combined with the proposed resource allocation algorithm,simulation results show that the proposed ICIC scheme can efficiently mitigate the inter-cell interference and improve the cell throughput.Different from existing dynamic ICIC schemes,the proposed scheme is distributed and low signaling overhead,which can be easily implemented to the large-scale networks.2.The random and dense deployment of femtocells has brought new challenges for interference management.Inspired by SFR designed for macrocells,we propose a semi-static ICIC scheme based on adaptive frequency reuse(AFR).The proposed AFR scheme involves two algorithms,respectively named game-based primary sub-channel self-configuration(PSC-SC)and intra-cell interference-aware resource allocation two algorithms.Different from SFR,each femtocell access point(FAP)decides the optimal frequency reuse factor according to its local interference graph.Since the convergence of the conventional best-response algorithm is very slow,we design a PSC-SC algorithm with fast convergence.Simulation results show that not only mitigate the interference among femtocells but also has the following advantages: distribution,scalability,low signaling overhead and low complexity.3.We consider the general resource allocation problem to minimize the aggregate power consumption while satisfying the minimum rate requirements of device-to-device(D2D)links and the one-MCS-per-user constraint.In order to decouple it,we divide the original problem into two cascaded sub-problems,i.e.sub-channel allocation and modulation and coding scheme(MCS)assignment.Given the requested number of sub-channels,a game-based sub-channel allocation scheme is proposed to obtain the optimal spatial frequency reuse pattern in term of minimizing the aggregate interference.Based on the obtained reuse pattern and rate requirements,a link adaptation scheme is developed to determine feasible MCSs for D2 D links.Simulation results show that our scheme can achieve better performance than existing schemes except of the greedy-based scheme which requires high computational burden.4.To deal with the cross-tier interference issue,an enhanced ICIC(e ICIC)based on almost blank sub-frame(ABS)is proposed by 3GPP.However,the e ICIC scheme reduces the spectrum utilization.Considering that the massive MIMO equipped at MBSs in the next generation of cellular networks,we can exploit some Do F to reject the cross-tier interference.Since the MBS is hard to obtain the covariance matrix of interfering signals under the conventional time division duplex(TDD)mode,we design a new operating mode,named reverse TDD(RTDD),i.e.while the macro-tier is in the downlink period,the pico-tier is in the uplink period and vice versa.Thus the MBS can easily obtain the covariance matrix of interfering signals based on the channel reciprocity.Then a corresponding precoding scheme is presented.In addition,a joint user association and resource coordination scheme is proposed to mitigate the interference between macrocell users(MUs)and co-channel picocell users(PUs).We evaluate the performance of our proposed scheme with two schemes,respectively named co-channel TDD(Co-TDD)and orthogonal TDD(Or-TDD).Simulation results demonstrate that a significant performance improvement for picocells can be achieved by the proposed scheme.