High Data Rate Links Based On Turbo-OFDM For UAV System

It is verified clearly by the wars happened these years that UAV (unmanned aerialvehicle) are playing more and more important role in modern war. UAV is featured bylong endurance, large cruise range, heavy overload capability as well as limitlessattitude alteration and casualty avoidance when flying in hostile territory, compared tomanned aerial vehicle. Compared to satellites, it has advantages of low altitude,controllable movement, convenient arrangement and low cost. Therefore, new types ofUAV are under development in many countries as preponderant battle platforms.As the operation radius and transmission rate for UAVs increase, more efficientand more reliability communication techniques are exigently demanded, among which,channel coding and modulation/demodulation are 2 key research fields for the high datarate transmission in UAV system (UAVS).The work in this dessey is supported by the project of National Defense 973. Themajor work focuses on Turbo-OFDM techniques in high data rate links for UAVS, andcan be summarized as follows.1. Based on the reference model, we modify the LOS (Line-Of-Sight) component,transimission gain, arrival angle and initial angle, and propose a novel simulation modelfor Ricean fading channels. This model is static in wide sense and obtains correctstatistical properties consistent to the theoretic one. Generalizing from this model, anefficient simulation model for multichannel uncorrelated Ricean fading channels can beconcluded, which has correct statistical properties for each sub-channel and areindependent among arbitrary sub-channels. Further, we deduce a novel and efficientsimulation algorithm for cross-correlated Ricean fading channels by taking the solutionof coloring matrix as an optimization problem. From the simualtion results, it isconcluded that this algorithm is universal and correct.2. For given channel estimation error variance, a new metric for Turbo decoding onRicean fading channels is proposed, and a novel decoding algorithm is concluded. theproposed decoding algorithm exhibits 0.1-0.6 dB code gain over the conventional one.Furthermore, the performance with modified metric approaches to that on the AWGNchannel under the condition of significant Ricean fading factor and small channel estimation error. Lastly, these conclusions can also be generalized to fading channels.3. Combining the advantages of Turbo codes and OFDM, we propose a newdecoding metric for Turbo coded OFDM system. From the derivation, it is concludedthat the mean and variance of signal-to-noise ratio (SNR) are important to theperformance. The simulation results show that the proposed algorithm can bring outabout 2.9 dB SNR gain over the scheme defined in IEEE 802.11a. Also the performancewith the proposed algorithm is very good on the condition of significant mean and smallvariance of SNR.4. With the performance and decoding delay considered, 3 adpative Turbo-OFDMschemes are proposed: (1) the scheme based on channel classification, which adaptivelyselects the optimal iterations for the decoder with classification of current channel state.(2) the scheme based on differential logarithmic cross entropy, which stops the decodingprocess by the different of differential logarithmic cross entropy for adjacent 2 iterations.(3) the scheme based on early stopped decoding rules via double CRC-4 checks, whichstops the decoding process with the 2 correct CRCs accessional to the information bits,and overcomes the disvantages of undectected rate of those based on CRC-4 and CRC-8.The simulation results show that these schemes can obtain low average bit error rate andhigh decoding speed. Finally, the performances are analysed and studied.5. Fully considering the characteristics of transformed image data stream, Turbocodes and UAVS channels, we propose an image transmission system based onTurbo-OFDM and wavelet transformation. This system utilizes unequal error protectionand dynamical subchannel allocation techniques to transmit different image datastreams. Simulation results prove that this system can effectively protect the importantstream and improve the image transmision performance over the noisy channel.6. By referring to IEEE 802.11a protocols, we analyze the transmissioncharacteristics such as transmission rate, communication distance and frequency bandsand conclude that the maximum transmission rate can be up to 54 Mbps when thedistance is 100 km. And also, the anti-jamming capability is discussed.The theoretic research and simulation results show that Turbo-OFDM technique isan effective and efficient way to realize the high data-rate transmission for UAVS withits superior performance.