| Ultraviolet communication（UVC）has a wide application prospect because of its rich spec-trum resources,non-line-of-sight communication,high-security communication,etc.How-ever,the interference of the atmospheric environment causes serious inter-symbol interfer-ence and increases pulse broadening,which not only limits the data rate but also reduces the system reliability.The introduction of channel coding technology can effectively improve the reliability of the system.Polar codes,as the only channel encoding scheme that has been theoretically proven to reach Shannon capacity,have good error correction ability and can effectively improve the reliability of the system.There are few applications of polar codes in UVC systems.The UVC system based on polar codes is researched in this thesis and the simulation is performed to verify its applicability.Firstly,the influence of the atmospheric environment on ultraviolet communication is stud-ied.The Non-Line-Of-Sight（NLOS）single-scattering model with high model accuracy,suitable for short-range communication,and low computational complexity is selected as the channel model for the UVC system.Due to the complexity of the channel model and the difficulty of extracting the channel transition probability,this thesis simulates the channel transition probability utilizing interval quantization,and then constructs polar codes by the Monte Carlo method.A suitable quantization interval can effectively reduce computational complexity while ensuring simulation accuracy.Secondly,the UVC system based on polar codes is constructed and simulated under dif-ferent parameters.The performance of the UVC system varies when different transceiver alignment methods,communication distances,and communication rates are selected.By using the cyclic redundancy check-aided serial cancellation list decoding algorithm,simu-lations are carried out under different code lengths,code rates,and decoding list sizes.The error rate performance is compared with that of the low density parity check（LDPC）codes under the same conditions.Polar codes show a 0.2d B gain over the LDPC codes at a bit error rate of 10-4.the UVC system based on polar codes is verified that can effectively reduce the bit error rate and improve communication reliability.Thirdly,due to the increase of inter-symbol interference in high-speed transmission,this the-sis selects the orthogonal frequency division multiplexing（OFDM）technique to solve this problem.When the communication rate is limited to 2Mbit/s,the UVC system,based on polar codes with different subcarrier modulation methods,the number of subcarriers,and different channel model parameters are simulated.The results show that the addition of the OFDM to the UVC system can eliminate inter-symbol interference under the high-speed transmission.The simulation results illustrate that with the smaller transmission half-angle and the receiving half-angle,the shorter communication distance,and the smaller number of sub-carriers,the bit error rate of the UVC system is smaller.Fourthly,this thesis proposes the enhanced Belief Propagation Bit Flip（BPF）decoding al-gorithm.In this thesis,the prior knowledge of information bits and the critical set is united to locate error-prone bits.A BPF decoding algorithm based on a new flip mechanism is designed,namely,CS-BPF decoder.This algorithm reduces the ranking complexity of the flipping set by half the number of flipping.Simulation results show that compared with the original BP decoding algorithm,the CS-BPF decoder can bring about 0.6d B of performance gain at a block error rate of 10-3.Compared to the CA-SCL decoder with list size of 4 on UVC systems,the CS-BPF decoder can reduce system latency to a large extent with similar decoding performance,verified as a strong practical application value. |
