Energy-Spectral-Efficient Randomly Deployed Cellular Networks:Modeling,Analysis And Optimization

With the explosive proliferation of mobile devices and services,the user density of largescale cellular networks continues to increase,which results in further escalating traffic generation.Increasing the wireless access points density becomes the key way to hold this.In this paper,“Randomly Deployed” means the large number of wireless access points and the large number of users are densely distributed.With the wireless access points deployed denser and denser,the interference power increases in a large-scale and becomes more complicated.Meanwhile,massive wireless access points deployed by self-behaviour users will be switched/off randomly,which result in the network topology varied in the time domain.Moreover,the mobile-traffic over the entire network will vary fast and randomly in the time and space domain.Therefore,it is urgent to investigate large-scale users' behaviors aware BS deployment strategies,which can adjust the dynamic large-scale users' behaviors.This paper adopts the stochastic geometry to model the randomness of cellular network,model three types of cellular network.We further use convex optimization to obtain and analyze the optimum solutions.Compared with traditional network model,the results from stochastic geometry are more general,hence could provide the insight of improving the energy-spectral efficiency in 5G.Moreover the concept proposed in this paper is comprehensive since it also include the randomly distribution of users,which is more practical.This paper put the energy-spectral efficiency as the efficiency metric and tends to answer the question that how to achieve the optimal energy-spectral efficiency with Qo S ensured under the three specific network architecture while considering the varying traffic characteristics in the following three parts.Specifically:1)The energy-spectral efficiency of large-scale cellular networks for characterizing its dependence on the base station(BS)density as well as for quantifying the impact of tele-traffic on the achievable energy-spectral efficiency is modeled under the specific quality of service requirements.This allows to match the BS deployment to the network's tele-traffic while conserving precious energy.More specially,a practical tele-traffic-aware BS deployment problem is formulated for optimizing the network's energy-spectral efficiency while satisfying the users' maximum tolerable outage probability.This is achieved by analyzing the optimal BS-density under special tele-traffic conditions.Furthermore,the energy saving potential of our optimal BS deployment strategy under diverse practical parameters is studied and provide insights into the attainable energy savings in dense random cellular networks.The simulation results conform the accuracy of our analysis and verify the impact of the parameters considered on the network's energy-spectral efficiency.The results also demonstrate that the proposed tele-traffic-aware optimal BS deployment strategy significantly outperforms the existing approaches in terms of its energy efficiency.2)The energy-spectral efficiency(ESE)benefiting from the joint optimization of coordinated multi-point(Co MP)transmission and base station(BS)deployment is evaluated in the context of dense large-scale cellular network.First,a closed-form network ESE expression is derived for a large-scale Co MP-enhanced network,which allows us to quantify the influence of key network parameters on the achievable network ESE,including the BS density and the cooperation activation probability,characterized by a Co MP activation factor as well as the users' behaviors,such as their geographical mobile-traffic intensity and average user rate.With the aid of this tractable ESE expression and for a given BS density,a cellularscenario-aware Co MP activation optimization problem is formulated while considering the users' outage probability as constraints to maximize the network's ESE.Then it jointly optimize the Co MP activation factor and the BS density to maximize the network ESE,again under the constraint of the users' outage probability.The simulation results confirm the accuracy of analysis and verify the impact of several key parameters on the network ESE.Finally,the ESE improvement of the proposed strategies is evaluated under diverse scenarios,which provides valuable insight into the joint Co MP and BS deployment optimization in dense large-scale cellular networks.3)An analytical framework for characterizing the achievable energy-spectral-efficiency(ESE)of HCNs is developed,which explicitly quantifies the relationship between the network's ESE and the randomly time-varying LUBs as well as other network deployment parameters.Specifically,the quantitative impact of the geographical mobile-traffic intensity,the load migration factor,the users' required service rate and the per-tier BS densities on the achievable ESE of network is modeled,while considering the area-spectral-efficiency requirements.Importantly,a closed-form ESE expression is derived,which enables us to explicitly analyze the properties of the network's ESE.Furthermore,the optimal LUB-aware BS deployment strategy is proposed for maximizing the ESE under specific outage constraints.Using numerical simulations,the accuracy of the analytical ESE expression is verified and the impact of several relevant system parameters on the achievable ESE is quantified.Furthermore,the achievable ESE performance of the network under diverse time-varying LUB scenarios is evaluated.The work therefore provides valuable insights for designing future ultra-dense HCNs.