Studies On Adaptive Reception Techonology In Optical Wireless Communication

Optical wireless communication (OWC) springs up again along with the development of optical fiber communication technology and the progress of the semiconductor lasers and optoelectronic detectors. OWC has widely promising applications (especially military fields) since it integrates the advantages of fiber communication and other traditional wireless communication technologies, which include large capacity, low cost, good security performance, convenient deployment and flexibility.However, the transmission media of OWC is atmosphere; atmospheric effects cause OWC channels to have a highly variable transmission, which affect the communication quality.This thesis concerns on the reception technology of OWC to suppress atmospheric turbulence. The main parts are as follows:(1) Based on the widely learning of references, the technical character and applications of OWC are reviewed in the first chapter, and the research status is introduced in brief.(2) The reception problem of all-weather OWC system is over-all researched. A reception signal model involved turbulence noise and other multiplicative noise was proposed. To solve the signal fading induced by atmospheric turbulence, simulation and experiment results are given.(3) Various reception technologies to overcome atmospheric turbulence, especially their merits and faults, are introduced overall. Adaptive received signal processing is chosen to overcome fading in OWC under strong and weak turbulence condition because of their advantages over others. Under the weak turbulence condition, Kalman filter and LMS filter were adopted to estimate and predict the statistics characteristics of the signal and turbulance channel. The corresponding design structure of the two kinds of filters and the optimum design algorithm were provided, and the two kinds of filters were compared in their applicable field. The BER performance of the adaptive filters was analyzed contrast with MLSD. Under the strong turbulence condition, the received signal of OWC over strong turbulence channels is assumed to be a mixture of K-distributed fading and Gaussian distributed thermal noise. Second-order spectral analysis is unable to separately estimate the mixed signal. In order to mitigate the fading induced by turbulence, the decision threshold-updating algorithm based on second and higher order cumulants is proposed and is able to operate in an unknown turbulence environment. The performance of the adaptive processing scheme has been evaluated by means of Monte Carlo simulations. Simulation results show the improvement of the bit error rate (BER) performance.The main innovation of the thesis was described in brief as follows:(1) The received signal model includes atmospheric turbulence noise and other photoelectric noise was proposed for the first time;(2) Based on adaptive filter, the adaptive receiver technology was presented to overcome atmospheric turbulence fading in optical wireless communication system for the first time;(3) We have analyzed the estimation of wireless communication channel parameter and the statistical characteristic of Gaussian noises estimation based on the high-order cumulants for the first time. The optimal detection strategies were set up to realize optimum threshold over strong turbulence.The main conclusion is:(1) To solve the signal fading problem induced by atmospheric turbulence in optical wireless system, a reception signal model involved turbulence noise and other multiplicative noise was proposed. A modified match filter was used to increase the signal-noise ratio (SNR) of 3-20dB.(2) According to the Maximum a prior criteria, the optimal threshold of the OWC system used IM/DD was deduced. The relationship of the optimal threshold with system parameters was discussed, which indicates that the estimation errors can cause significant errors in the calculation of the optimum detection threshold of 10-20's of percent.(3) The Kalman adaptive filter was adopted to estimate the statistical characteristic of signal and noise over weak turbulence channel. In case of constant optical power, the algorithm reaches the stabilizing value when the sample bit number is about 600- 800. In case of changing optical power, the algorithm reaches the stabilizing value when the sample bit number is about 800- 1000. The adaptive system with Kalman filter yields of BER performance reduced by 1-2 orders of magnitude contrast with non-adaptive system.(4) The decision threshold-updating algorithm based on LMS adaptive filter is introduced and investigated. An accurate expression for the steady-state tap-weight covariance matrix is derived for the real LMS algorithm. The effect of the parameters in LMS algorithm on OWC system performance was introduced through theoretical analysis and computer simulation. It is shown that in the OWC system with non-causal filter the turbulence effect has been apparently abated. The BER of the system over weak turbulence based on LMS adaptive filter is below 10-8, which can satisfy the communication requirement. The non-causal filter of 255 orders is the optimal structure.(5) To reduce the computational cost, we propose to make use of ad hoc 1-D slices of the fourth-order cumulant in the nonredundant support region (or principal domain). In the hypothesis of an exponential correlation structure of the fading sources, we estimated the values ofρI,σ02,σ12 andρυ2 by estimating the AR(1) and ARMA(1,1) model parameters. We estimated separately the parameter of the K-distributed fading from sixth-order cumulant fourth-order cumulant. The performance of the adaptive processing scheme has been evaluated by means of Monte Carlo simulations. Simulation results show the improvement of the bit error rate (BER) performance.