| Focus on the problem of short-distance and large-capacity data transmission between underwater vehicles and buoys and submerged beacons,the imaging MIMO-ACO-OFDM communication technology has been studied based on the optical imaging MIMO technology and the transmission characteristics of visible light and the characteristics of LED devices underwater.In this paper,underwater optical LOS links are analyzed based on Monte Carlo method,the spatial correlation of imaging MIMO is further modeled and analyzed,and the effect of spatial correlation on the performances of channel capacity andBER are analyzed.And then,the techniques to reduce spatial correlation and methods to suppress clipping distortion and nonlinear distortion in LED modulation are proposed.In the MISO link from multi-node to AUV,a code domain non-orthogonal transmission technology based on joint detection of compressed sensing is proposed.The main work and contributions completed on the paper include:?1?Aiming at underwater LED visible light channel and large-capacity transmission problems,this paper proposes an imaging multiple input multiple output?MIMO?system for underwater optical communication,so that the LED array space division multiplexing performance is better than non-imaging MIMO systems.Although the receiving end uses an imaging lens to separate signals,there is still spatial propagation correlation in the imaging MIMO system.Aiming at the problem of spatial correlation,we modeled it based on LOS channel model.When the object distance is fixed,the spatial correlation becomes weaker as the LED distance increases,and when the LED distance is fixed,as the object distance increases within a certain range,the spatial correlation increases and tends to be stable.Considering that the underwater short-range optical communication is a flat channel,the water injection method and the equal power method are used to derive the channel capacity and bit error rate performance under different spatial correlation conditions.The simulation results show that with the decrease in spatial correlation and the increase in the number of LED light source antennas,the channel capacity of the imaging MIMO system becomes larger,and the BER gradually decreases.When the number of channel conditions drops to about 1,the BER of the imaging system reaches the degree of 10-6 when SNR is20 dB.Compared with the non-imaging MIMO system,SNR obtained a gain of 12 dB at the same BER.Further,when SNR of channel is low,the channel capacity obtained by the water injection power allocation method is slightly higher than the channel capacity obtained by equal power allocation.When the SNR is high,the channel capacity obtained by the water injection method is the same as equal power allocation.?2?Aiming at the problem of the spatial correlation of the array of underwater imaging optical MIMO systems,an imaging MIMO-ACO-OFDM system reduction correlation algorithm based on SVD power distribution,maximization of minimum European distance of the received signal set,and lattice reduction detection algorithm are proposed.By power allocation,the difference in sub-channel gain is reduced,thereby reducing the spatial correlation.Under the same BER,the precoding algorithm based on SVD power allocation is better than that without power allocation method,the SNR is increased by 7dB gain.On the other hand,in order to maximize the minimum Euclidean distance of the received signal set,the precoding matrix is optimized and solved under the constraints of non-negative optical signals and total power.By using the optimal power allocation precoding matrix to perform precoding on the signal and ML detection,the simulation results show that,compared with the SVD based power allocation precoding algorithm,the precoding algorithm based on the minimum Euclidean distance maximization of the received signal set can obtain about 5 dB gain on the same channel,which further improves the BER performance of underwater imaging MIMO optical communication system.?3?By comparing to the traditional algorithms,such as ZF,MMSE,OSIC and ML detection algorithms,although the BER performance is improved by precoding based on the minimum European distance maximization of the received signal set,the algorithm complexity is higher.Therefore,this paper takes into account the complexity andBER performance and proposes a low complexity LLL-ZF detection algorithm based on lattice reduction,which improves the orthogonality of the original channel matrix by reducing the lattice basis of the optical MIMO channel matrix,and achieves 3dB gain compared to the precoding algorithm based on the minimum Euclidean distance maximization of the received signal set,but the complexity is greatly reduced combined with the ZF detection algorithm.?4?To address the ISI problem caused by the delay extension of underwater optical channel,this paper study the principle of the new underwater visible optical modulation technique DCO-OFDM/OQAM system and the time-frequency filter,and the M-OSLM peak-to-average suppression algorithm in the DCO-OFDM/OQAM system is proposed to solve the distortion problem caused by clipping when the system complex signal is turned into real signal.The simulation results show that by using the M-OSLM algorithm to suppress clipping distortion,the SNR of the DCO-OFDM/OQAM system is improved by3dB compared with the DCO-CP-OFDM system at the same BER on the underwater multipath delay spread channel.Foucs on the problem of DC bais optimizing,the modelling of the transmit power of visible DCO-OFDM/OQAM system is optimally allocated between the OFDM/OQAM signal and DC offset in order to match the linear modulation interval of the LED,so that maximize the effective SNR at the receiving end maximum.Simulation results show that under the optimal power distribution ratio,the effective SNR reaches the maximum and the BER is the minimum by solving optimization model.An underwater green visible light DCO-OFDM system based on FPGA was also built,and the bandwidth expansion performance of the system under certain synchronization performance was analyzed,which improved the spectral efficiency and laid the foundation for the next step to build an imaging MIMO system.?5?Aiming at the problem of high-capacity transmission of data collected by sensors in underwater AUVs,a non-orthogonal chip mapping multiple access communication technology based on code domain is proposed to improve the average throughput of the system in the MISO uplink scenario.Based on the characteristics of large number of nodes in detection,few active nodes within a certain period of time and sparse signal,a non-orthogonal multiple access transmission system with chip mapping code division multiple access is designed with compressed sensing theory.Compressed sensing technology is used to jointly detect the active nodes and data in the uplink non-orthogonal multiple access transmission system.Simulation results show that the system can not only reliably detect active nodes and data using compressed sensing theory,but also achieve overload performance.The rate of overload reaches 300%when the active nodes are sparse.And the spectrum utilization and average throughput in underwater multi-node communication system are improved significantly. |
PDF Full Text Request