Interference Analysis And Management In Heterogeneous Cellular Networks

Heterogeneous cellular networks (HCNs), which are composed of traditional macrocells and small cells, can improve coverage and enhance quality of service (QoS) with lower power consumption. Interference management is one of the dominant challenges in the study of HCNs. Orthogonal frequency division multiple access (OFDMA) is well-known for its flexibility in spectrum allocation and robustness to multipath fading and it has been widely deployed in HCN to mitigate co-tier and cross-tier interference. However, in the uplink of OFDMA HCNs, since macrocell user (MU) signals are scheduled to be synchronized at the macrocell base station (MBS), their signals generally arrive at the small cell base stations (SBSs) with different delays and can hardly be synchronized in time causing inter-carrier interference (ICI) and inter-block interference (IBI) in OFDM demodulation at the SBS. In this paper, the detrimental effect of ICI and IBI on the uplink of small cell is analyzed. With or without cooperation between MUs and SUs, two novel schemes are proposed to mitigate or remove ICI and IBI in the system. Besides, an optimal cognitive radio enabled uplink channel access scheme is proposed to maximize the average throughput in OFDMA two-tier HCNs, and a novel distributed power control scheme is proposed to improve successful transmission ratio of femtocells as well. The main contributions of this paper can be listed as follows.(1) The effect of ICI and IBI on the small cell uplink caused by MU signals under single path' channel is analyzed. First, the cumulative distribution function (CDF) of MU arrival time is derived. Evaluation shows that the derived CDF is accurate and agrees with simulation results well. Based on the CDF, a closed-form expression of probability mass functions (PMFs) of relative delay of MU signals and probability of ICI and IBI occurance for two synchronization schemes are developed. Based on the PMF of relative delay, ICI and IBI power averaged over MU locations for the macrocell using fractional power control is derived. This new theoretical result helps analyze the signal to interference plus noise ratio (SINR) of the femtocell, femtocell capacity and symbol error rate (SER). It is shown that the performance analysis is quite agreeable with computer simulations. The theoretitical analysis provides a solid evaluation of the effects of ICI and IBI on system performance for different small cell locations.(2) The statistics of ICI and IBI caused by MU signals under multipath fading channels are analyzed. Firstly, the ICI and IBI of OFDMA HCNs with fractional power control in multipath fading channels are formulated, and the closed-form expressions of their covariance matrices at the SBS with random MU locations are derived. The average power of total interference can be obtained from the correlation analysis. Besides, the closed-form expressions of average small cell uplink performance, including capacity and SER, are derived. The analysis is agreeable with simulation results very well. It is shown that the capacity and SER is seriously degraded and bounded in high SNR regime in the presence of IBI and ICI.(3) A precoding method for small cells for mitigating the effects of ICI and IBI on uplink capacity is proposed. In the method, the received signals over small cell subcarriers at the SBS are decorrelated jointly by small cell user (SU) precoders and an equalizer at the SBS. Closed-form design procedures for precoding of different subcarrier sizes to maximize uplink capacity under different information feedback are developed. Simulation results demonstrate that the proposed method for precoding has a large capacity gain over that of no small cell precoding. In detail, for using all the subcarriers for precoding, the proposed scheme has the largest capacity. Moreover, the method can perform very well even MU locations are the only information fed back from the macrocell.(4) A novel method based on interference alignment (IA) is proposed to remove ICI and IBI in the small cell completely, which ensures data streams on all subcarriers are transmitted interference freely in two-tier HCNs. In the new scheme, by MU precoding over parallal subcarriers, ICI and IBI are constrained into subspace with dimensions no more than the number of subcarriers adopted by macrocell and are zero-forced at the SBS. Extensive simulations show that the proposed method can remove the upper bound of the capacity of the SUs and improve the system degree of freedom (DoF) from the number of subcarriers adopted by macrocell to the number of total system subcarriers. In the scheme, the MU and SU precoders should be designed jointly to deal with ICI and IBI. Hence, it can achieve better performance than that without the cooperation between MUs and SUs.(5) A optimal cognitive radio enabled small cell uplink channel access scheme is proposed to avoid co-channel interference and maximize the average throughput of OFDMA two-tier HCNs. When fractional power control and adaptive modulation transmission are adopted in the system, the closed form of average throughput of two-tier cognitive HCNs is derived with imperfect spectrum sensing results. Based on the derivation, the optimal channel access probability of small cells is achieved. Simulation results verify the validity of derived results. Besides, in the derivation, the spatial randomness of small cells, MUs and SUs are all considered, and thus the derived access probability is robust to MU and SU location changes.(6) A novel downlink interference mitigation strategy based on adaptive power control is proposed. The proposed solution is composed of three parts. The first is femtocells clustering, which maximizes the distance between femtocells using the same subchannel resources to mitigate co-tier interference. The second is to assign MUs to clusters by max-min criterion, by which each MU can avoid using the same resource as the nearby femtocells. The third is a novel power control scheme with femtocells downlink transmit power adjusted adaptively based on SINR levels of neighboring users. Simulation results show that the proposed scheme can effectively increase the successful transmission ratio and average outage capacity of femtocells, while guaranteeing QoS of the macrocell.