Research On Cognitive-intensive Cell Interference Coordination Technology For 5G

In the next 10 years,wireless communications will face explosive growth in data traffic.The increase in capacity demand drives the continuous evolution of the network structure,and the low-cost,miniaturized,low-power,low-power Small Cell has gradually become the protagonist in the communication network.Higher spectrum,higher bandwidth,and higher dense networking have become a major trend in 5G communication technology.The dense deployment of small base stations,irregular deployment,and proximity to terminals make serious interference problems in dense cell networks,and this has become an urgent problem in 5G networks.Cognitive Radio(CR)is a technology that automatically detects the surrounding radio environment for spectrum sensing and distribution.It can intelligently adjust the parameters of the system to adapt to changes in the environment.Integrating the CR function into the base station has been considered an effective solution for interference coordination.In this paper,the CR function is assigned to the dense cell system to form a CR communication system for interference coordination.The specific work is as follows.Aiming at the problem of co-layer interference between cognitive dense cells,the problem model is established with the goal of maximizing the micro-cell system capacity under the constraint of certain interference threshold.The problem is decomposed into channel allocation and power control based on Lagrangian double decomposition method.The two sub-problems are solved step by step.Based on the greedy algorithm for channel allocation and the optimal power control based on Lagrangian theorem,a joint channel allocation and power control algorithm is proposed.The simulation results show that SBS(Small Base Station)with cognitive ability can achieve nearly 60% capacity gain compared with ordinary base stations.At the same time,when the number of available channels is less than 6,the proposed joint resource allocation scheme has nearly one-time gain compared to the capacity achieved by the existing channel coloring scheme.When the joint channel allocation and power control algorithm performs 7 iterations,the average SBS capacity reaches 2.20 Mbps convergence value.Compared with the fixed power allocation scheme,the capacity gain is close to 5%.Finally,the joint resource allocation algorithm proposed in this paper can be effectively coordinated.Interval interference between dense cells.Aiming at the cross-layer interference problem of two-layer heterogeneous cognitive dense cell network,a cognitive power allocation(CPA-Q)algorithm based on the enhanced learning Q-Learning algorithm is proposed to maximize the micro-base station user QoS constraints.Zone capacity,set two optimization targets CPA-1 and CPA-2.The simulation results show that CAP-2 has a gain of nearly 0.1 Mbps compared with CPA-1 on the average SBS capacity,which is nearly 0.2 Mbps higher than the Q-Learning algorithm proposed in the existing literature.At the same time,a Cooperative Q-Learning(CQL)model is proposed to improve the unfair and global optimal problems between base stations caused by single-agent mode.Simulations show that when SBS deployment density is 50%,CPA-2 has nearly 35% system capacity gain in CQL mode compared to single agent mode.At the same time,the fairness between SBS in CQL mode is always maintained at around 1,which further improves the fairness between cells.