Capacity Analysis And Constellation Design Of Visible Light Communication Systems

Visible light communication(VLC)using light-emitting diodes(LEDs)has become a promising candidate in the field of wireless communication in recent years due to its appealing advantages such as reusability of LED lighting infrastructures,freedom from hazardous elec-tromagnetic radiation,high transmission rate and costly frequency licensing Since the visible light signal is non-negative and upper-bounded,the capacity-reaching probability distribution function of the signal is significantly different from that of a conventional RF wireless com-munication system.This dissertation focuses on the capacity analysis and constellation design problem of visible light communication systems.The main work is as follows:1.The capacity of the SISO-VLC system with no average optical power constraint is an-alyzed.The main jobs include:1)The capacity problem of SISO-VLC system is written as a functional optimization problem.It is proved that the objective function of the functional op-timization problem is concave and the feasible region of the functional optimization problem is convex in the normed linear space.Thus,the capacity optimization problem is a concave functional optimization problem with one and only one solution;2)Two functional equation-s that are satisfied by the solution of the capacity optimization problem are deduced;3)It is further proved that the capacity-reaching PDF of the input signal must be combined of a finite number of discrete points;4)Two algebraic equations that are satisfied by the number,position,and probability of the optimal finite constellation points are deduced;5)Under the assumption that the constellation number N is given,the general conditions satisfied by the position and probability of the N discrete points are obtained.The exact position and probability of the N constellations are given when N=2 and N=3;6)Under the assumption that the input signal is an N equiprobable discrete point,the lower bound of the capacity is given.2.The capacity of the SISO-VLC system with average optical power constraint is ana-lyzed.The main jobs include:1)The SISO-VLC capacity problem with the added average optical power constraint can be transformed into a new unconstrained functional optimization problem by Lagrangian multiplier method;2)Since the objective function of the transformed functional optimization problem is concave and the feasible region is a convex set of the normed linear space,the transformed functional optimization problem is concave with one and only one solution;3)The functional equation satisfied by the PDF solution of the transformed concave functional optimization problem is given.It is further proved that the capacity-reaching PDF of the input signal must be combined of a finite number of discrete points;4)The algebraic equations that are satisfied by the number,position,and probability of the optimal finite con-stellation points are deduced;5)It is proved that when A/?is fixed,the optimal constellation points corresponding to the dimming coefficient?and the ones of 1-?distribute symmetrical-ly;6)Under the assumption that the constellation number N is given,the general conditions satisfied by the position and probability of the N discrete points are obtained.The exact position and probability of the N constellations are given when N=2;7)Under the assumption that the input signal is an N equiprobable discrete point,the conditions satisfied by the position of the N points are given.3.The capacity of multiple input multiple output visible light communication(MIMO-VLC)systems has been studied.The main jobs include:1)The capacity problem of M×N order MIMO-VLC system is equivalent to the joint capacity problem of min(M,N)decou-pling sub-channels obtained by SVD decomposition of the M×N order channel matrix;When N?M,the original MIMO capacity problem is completely equivalent to the joint capacity problem of the decoupled subchannels;in the case of N<M,the solution to the original MI-MO capacity problem is not unique,but the value of the capacity is equivalent to the decoupling sub-channel joint capacity problem;2)It is proved that the min(M,N)decoupling sub-channel joint capacity problem is concave thus has one and only one solution;3)The functional equation satisfied by the PDF solution of the min(M,N)decoupling sub-channel joint capacity prob-lem is given.4)It is proved that the capacity-reaching PDF of the joint capacity problem of min(M,N)decoupling sub-channels must be combined of a finite number of discrete points;5)Under the assumption that the constellation number K is given,an upper bound of joint capacity of the decoupled sub-channel is deduced.4.The joint constellation location and labeling optimization problem is studied in VLC-CSK system.The main jobs include:1)The symbol error rate and bit error rate of 4-CSK,8-CSK,16-CSK and 2~n-CSK VLCsystem are deduced based on the given constellations;2)The expression of the pragmatic mutual information for the equivalent parallel binary chan-nel model of VLC-CSK system is deduced.3)Taking the pragmatic mutual information as the objective function,the joint constellation location and labeling optimization problem is in-vestigated.Simulated annealing algorithm and interior point iterative algorithm for the joint optimization problem are given.