Research On Optimal Spectral And Energy Efficiency Of Visible Light Communication System Based On OFDM

With the development of intelligent devices and 5G wireless communication popularity,spectral resource is becoming less but the consumption of energy is still growing.Under the quality of user's service,how to improve the system's energy efficiency and spectral efficiency has become an urgent problem.Visible light communication(VLC),as a technology that both consider lighting and communication,has gradually become a research hotspot in the field of communication for its rich spectrum resources,large bandwidth,low susceptibility to electromagnetic interference and high security.Orthogonal frequency division multiplexing(OFDM)can effectively combat multipath interference,so it can guarantee the stability of VLC systems and increase the information transmission rate.At the same time,in order to promote green and low-carbon development,the next-generation wireless communication technology must also consider the issues of spectrum resource utilization and energy efficiency to achieve balanced development.This paper studies the spectral and energy efficiency maximization problem of OFDM-based VLC systems under the condition of optimal power allocation.The specific works are as follows:(1)For asymmetrically clipped optical OFDM(ACO-OFDM)VLC system,the system model is established and a closed-form expression of its reachable rate is derived.Under the constraints of transmitted optical power and electrical power,the problem of maximizing spectral efficiency needs to be solved.Based on the Gaussian input distribution,this problem is a convex problem,which can be solved by Lagrange function to get the optimal power distribution.For the finite-alphabet input distribution,the relationship between the mutual information and the minimum mean square error(MMSE)is established.According to the Karush Kuhn Tucker(KKT)condition,we get the optimal power allocation scheme,which is solved by bisection method.When it comes to the lowerbound of mutual information,we develop a low-complexity suboptimal power allocation scheme,which can be solved by the interior point method.Simulation results show that when the input is Gaussian distribution,the system will allocate more power to the subcarrier with better channel gain.Compared with Gaussian distribution input,finite-alphabet input can achieve higher spectral efficiency for OFDM VLC systems.(2)For ACO-OFDM energy efficiency maximization problem,the constraints of power and rate are taken into consideration.Under three different input conditions,the problem is quasi-concave,and it can be converted to a convex problem by Dinkelbach-type iterative algorithm.In each iteration,the interior point algorithm is used to obtain the best power allocation.Numerical simulations show that the power distribution of finite-alphabet input is better than the Gaussian input distribution.As the rate constraint increases,energy efficiency tends to remain unchanged and then decline.In addition,related expressions of spectral efficiency and energy efficiency are given to verify the trade-off relationship.(3)For DC-biased optical OFDM(DCO-OFDM)VLC system,it analysis the system model,the spectral and energy efficiency maximization problem are established under the constraints of power,rate and cut off.For finite-alphabet input spectral efficiency maximization problem,the Lagrange function is given and solved by the "mercury-waterfilling method".In the case of the lower bound of mutual information,the maximization problem can be obtained by the interior point method.The energy efficiency maximization problem can be transformed into a convex problem by Dinkelbach algorithm to solve.The numerical simulation results show that with the increase of power constraints,the spectral efficiency and energy efficiency both increase first and then remain unchanged.Compared with the lowerbound of mutual information,finite-alphabet input can achieve higher spectral and energy efficiency for OFDM VLC systems,the spectral efficiency of DCO-OFDM is better than that in ACO-OFDM.This thesis contains 31 figures,7 tables and 85 references in total.