Research On Indoor Visible Light Communication System Capacity And Physical Layer Security Performance

Visible Light Communications(VLC)technology is one of the supplementary technologies for the next generation of wireless communications.It has the advantages of wide spectrum range,high transmission rate,and no electromagnetic pollution.However,there are still many problems to be solved in the basic theoretical research on VLC,especially in terms of channel capacity and physical layer security.Because of the differences between VLC and Radio Frequency Wireless Communications(RFWC),classical conclusions such as the Shannon formula cannot be directly applied to VLC networks.Therefore,this paper mainly studies the channel capacity of the VLC pointto-point channel,the privacy capacity of the eavesdropping channel,and the design of the beamforming scheme in the eavesdropping channel.The specific research contents are as follows:First,a point-to-point dimmable VLC system is considered.In such a system,the main distortion is caused by the additive white Gaussian noise and the input signal-related noise.The input-related noise changes depend on the change of the current signal.Under the constraints of non-negative input signal,peak power and average power,the expressions of the upper and lower bounds of the channel capacity are derived.In the derivation of the lower bound,the variation problem is used to solve the optimization problem and a better input distribution is obtained.Then,using the characteristic that the entropy of the output signal is always greater than that of the input signal,a closed expression for the lower bound of capacity is derived.The derivation of the upper bound is based on ideas such as dual expressions of relative entropy and channel capacity.In addition,the numerical results are used to verify the accuracy of the derived upper and lower channel capacity.Then,based on the research of the point-to-point dimmable VLC system,the secrecy capacity of VLC eavesdropping channel in the physical layer security is studied.Based on the input-dependent Gaussian noise situation,under the premise of signal amplitude constraints,the channel secrecy capacity under two conditions with and without peak power constraints are studied respectively,and closed expressions for the upper and lower bounds of the secrecy capacity are obtained.Because the entropy of the output signal is always greater than that of the input signal,a better input distribution can be derived by using the variational method and the entropy power inequality.Then the lower bound can be derived on this distribution.The upper bound is derived from the dual expression of relative entropy.Finally,the accuracy of the upper and lower bounds of the secrecy capacity results was verified by numerical results.Finally,a safety beamforming scheme for indoor VLC networks is designed.The network consists of a transmitter,a legitimate receiver,an eavesdropper and several friendly jammers.In a VLC network,both the main channel and the eavesdropping channel have a uniformly distributed signal input and are subject to the interference of additive white Gaussian noise.Considering the imperfect Channel State Information(CSI)by using an ellipse approximation,the secrecy rate of the indoor VLC network system is analyzed under the constraints of the amplitude of the data signal and the eavesdropping signal.Based on the closed-form expression of the secrecy rate,an optimization problem of beamforming is proposed to maximize the secrecy rate under the given constraints.Since the optimization problem is non-convex,it is transformed into a solvable quasi-convex linear search problem.Numerical results verify the secrecy performance.