Visible light communication(VLC)is a rapidly developing technology.It has many advantages such as no electromagnetic radiation,no electromagnetic interference,no spectrum certification,high security.It has gradually become the hot research direction in the field of communications.The indoor VLC system using Multi-Input Multi-Output(MIMO)technology can significantly improve the signal coverage,obtain a higher transmission rate,solve the problem of the alignment between transmitters and receivers,reduce the impact of persons walking and shadow.The resource adaptive allocation technology based on Orthogonal Frequency Division Multiplexing(OFDM)and MIMO can improve resource utilization,better adapt to changing channel environment,meet different user needs,eventually achieve the rational allocation of resources and optimize system performance.However,the current research on MIMO VCL still has some shortcomings.On the one hand,there are few researches on adaptive allocation of MIMO-OFDM VLC resources.The rate proportional allocation scheme proposed by Shen is computationally complex and does not consider the allocation of bits.Furthermore,the problem of resource allocation of the VLC system with high-speed users has not been solved yet.On the other hand,at present,the higher BER performance and lower computational complexity of VLC MIMO signal detection methods often cannot be combined.In view of the above issues,this article will conduct the following aspects of research.Firstly,the basic theory and channel modeling of indoor VLC are summarized.Several commonly used modeling methods are introduced,including Ray Tracing Model,Exponential Decay Model and Independent Reflecting Element Interaction Characterization Model.Through simulation,the system performance indexes of indoor visible light communication system are analyzed,such as impulse response,receive optical power,receive SNR and so on.Then,the adaptive resource allocation technology of indoor MIMO-OFDM VLC system is studied.The principle and steps of three kinds of adaptive resource allocation schemes are analyzed,including maximum rate allocation algorithm,proportional rate allocation algorithm,and subcarrier ordering allocation algorithm.The advantages and the disadvantages of the three algorithms are analyzed.The proportional rate allocation algorithm(PRA)proposed by Shen is improved.On the basis of the original algorithm,the subcarriers and power distribution are simplified,the bit loading process is added,and the fairness between users during bit loading is ensured by two steps of preloading and post-adjustment.Simulation shows that the BER of the improved algorithm has increased but is small,less than 1 order of magnitude,but the system capacity is increased by 15%.Meanwhile,the subcarriers,power allocation and bit loading are basically consistent with the user’s rate requirements.For the case of the users with high speed requirements,propose to improve the allocation scheme.The main idea is to allocate as many subcarriers and power as possible to high-speed users during allocation and perform bit loading by using two processes of preloading and post-adjustment,so that high-speed users can complete high-speed services as soon as possible.The improved scheme can basically meet the requirements of high-speed users,and take into account the rate needs of other users.The simulation shows that this scheme can allocate as much resources as possible to the high-speed users under the premise of ensuring the rate requirements of other users and having little impact on the system bit error rate(less than one order of magnitude).Resources can be allocated on demand of users,basically.Finally,the MIMO layered space-time technology and signal detection technology for indoor visible light communication are studied.The diversity technology and the mergence technology are classified and summarized,and the performance of different merging methods are obtained through the simulation comparison.The MIMO layered space-time codes,coding and decoding principles,and commonly used MIMO signal detection methods are introduced categorically.Simulation compares the BER of different OSIC methods.The joint detection scheme combining multiple detection methods is proposed,that is,ZF-ML-OSIC,MMSE-ML-OSIC and ZF-MMSE-ML-OSIC which are respectively based on the SINR,SNR and the 2nd order norm of the characteristic matrix.The simulation analysis shows,when the SNR is 25dB,the BER of the ZF-ML-OSIC detection is 10-4,which is 2 orders of magnitude lower than that of the ZF-OSIC detection.The BER of MMSE-ML-OSIC detection reaches 10"5,which is 2 orders of magnitude lower than that of the MMSE-OSIC detection.The BER of ZF-MMSE-ML-OSIC detection is 10-5,which is nearly 3 orders of magnitude lower than that of ZF-OSIC and 2 orders of magnitude lower than that of MMSE-OSIC. |