| In recent years,roads are becoming crowded with the continuous increase of vehicles.It’s urgent to manage traffic intelligently.The intelligent transportation system(ITS)is a safe,convenient,and efficient traffic management system.The dedicated short-range communication based on radio frequency(RF)is a common technical for the traditional ITS.However,with the increase of vehicle density and the complexity of the traffic environment,it is prone to problems such as insufficient frequency band and a serious Doppler effect.Visible light communication(VLC)uses visible light with a wavelength of 380-780 nm,and its rich spectrum resources can solve the problem of bandwidth insufficient of RF communications.So,it can be used as a supplement to RF communication.With the development of semiconductor technology,LED gradually replaces the traditional light source and becomes an important choice of vehicle external lighting equipment,which provides hardware conditions for the vehicular visible light communication(VVLC).However,the application of the VVLC in the actual scenario faces many challenges.Based on the principle from simplicity to complexity,this paper focuses on the VLC of the vehicles with the same direction and lane,and parallel vehicles with the same direction in the adjacent lane.The main challenges are as follows.(1)For the front and rear vehicles with the same direction and lane,when the LED array-based taillight of the front vehicle is used as the transmitter,the image sensor receiver installed on the rear vehicle is affected by the blooming effect,easily causing bit errors.(2)For parallel vehicles with the same direction in the adjacent lanes,there is almost no line-of-sight(LOS)path between the transmitter and receiver,easily causing VLC interruption.The above problems are studied in this paper,and the main research work is as follows.1.The threshold selection algorithm based on probability distribution is proposed to solve the problem that the bit errors are easily caused by the blooming effect in the receiver for the vehicles with the same direction and lane.Firstly,the Gaussian function is used to approximate the blooming effect.Then,for LEDs with equal interval distribution,the output values of pixels and probability distribution curves corresponding to the status of on and off under the influence of the blooming effect are analyzed.Finally,the threshold is selected by using the method of probability statistics.When the probability distribution curves intersect,the traversal method is used to calculate the output values of pixels with the lowest bit error ratio(BER)in the intersecting area which is taken as the threshold.Otherwise,the average value of the minimum output values of pixels in the status of on and the maximum output values of pixels in the status of off is taken as the threshold.The simulation results show that the BER performance of the threshold selection algorithm based on probability distribution is better than the existing expected threshold method and the average threshold method.2.To solve the problem that there is almost no LOS path between the transmitter and receiver for parallel vehicles with the same direction in the adjacent lanes which is easily causing VLC interruption,the VVLC system aided by the mirror array-based intelligent reflecting surface(IRS)is designed,and the nearest neighbor iterative search(NNIS)algorithm for IRS nodes selection is proposed.Firstly,a single IRS aided VVLC system is designed,and the influences of the number of mirrors and transmission distance on the signal-to-noise ratio(SNR)are analyzed.Then,when a single IRS node cannot meet the requirements,the multiple IRSs aided VVLC system is designed,and the influences of the interval and number of IRS nodes for the SNR are analyzed.Finally,based on the principle of maximum resource utilization,the nearest neighbor iterative search method is proposed to select IRS nodes.The simulation results show that the IRS nodes selected by NNIS algorithm improves the resource utilization while meeting the communication requirements.3.To improve the energy efficiency(EE)performance of the mirror array-based IRS aided VLC system for parallel vehicles,the binary search conditional iteration(BSCI)algorithm is proposed to find the optimal number of mirrors under EE maximization.On this basis,the gradient iterative search algorithm is proposed to find the optimal number of mirrors and input signal power under EE maximization when input signal power changes.Firstly,the EE optimization problem when the number of mirrors changes is proposed.The EE is transformed into a unimodal function of the number of mirrors under certain constraints and the BSCI algorithm is proposed to find the optimal number of mirrors under the maximum EE.Then,the EE optimization problem is analyzed when the input signal power changes,and the Dinkelbach method is used to find the optimal input signal power.Finally,the EE optimization problem is proposed when the number of mirrors and the input signal power change,and the EE is maximized through the joint optimization of the two parameters,and the gradient iterative search algorithm is proposed.According to the initial input signal power value firstly,the BSCI algorithm is used to calculate the optimal number of mirrors under EE maximization and then the fixed number of mirrors is used to find the optimal input signal power using the Dinkelbach method,and the alternate optimization is carried out.By analyzing the number of iterations of the algorithm,the effectiveness of the proposed optimization algorithm is verified.4.To reduce the system power consumption at a given target achievable rate,the approximate iterative optimization algorithm is proposed to find the optimal number of mirrors under the lowest power consumption.When the target achievable rate is constant,it is proved that the system power consumption is a unimodal function of the number of mirrors.Then,according to the expression of total power consumption,the optimization problem of total power consumption when the number of mirrors changes is proposed,and the iterative range of the number of mirrors is calculated.Finally,the approximate transformation method is used to equivalent the power consumption to a continuous function of the number of mirrors,and the approximate iterative optimization algorithm is proposed to find the optimal number of mirrors under the lowest power consumption.Simulation results show that the proposed algorithm requires the least number of iterations compared with the bubble sort method and BSCI algorithm. |
PDF Full Text Request