Design And Implementation Of Encoding And Decoding For BLE 5.0 Standard PHY Layer For

As an important component of digital information,channel coding and decoding,have been widely applied to various communication systems,such as anti-interference and error control.In the BLE(Low-Power Bluetooth)4.2 protocol,there is no encoding and decoding operations in the PHY layer.If bit error or interference occurs during transmission,the quality of the original signal will be affected.Therefore,in order to improve the accuracy of data transmission,a channel codec module is added in the BLE 5.0 standard.The error correction performance of channel encoding and decoding in the BLE 5.0 standard PHY layer is researched.Viterbi algorithm is a kind of maximum likelihood decoding algorithm for convolutional codes.It not only has faster decoding rate,but also has higher efficiency than other decoding algorithms.Therefore,it is considered as the best decoding algorithm for convolutional codes.This thesis focus on implementing Viterbi decoder of(2,1,4)convolutional codes.The thesis introduces the basic principles and description methods of convolutional code and presents the theoretical knowledge of the Viterbi algorithm.The design of Viterbi algorithm is investigated.The progress of the thesis is as follows:1.A method for convolution coding and decoding is presented for the small data packets of the BLE,which mainly solves the noise interference problem in the data channel transmission.And the following are two innovations proposed in the hard decision and soft decision algorithm:(1)In the decoding process,an optimized method in data calculation,storage,and comparison in the decoder unit is proposed.The Add-Compare-Select unit calculates the path metric values,and stores the results in the previous state from the fourth moment,each jump branch path with the corresponding state sent by Branch-Metric-Generation unit will be added with the path metric.A smaller path metric will appear.This method can not only reduce the bit error rate in signal transmission,but also reduce the space complexity of the Viterbi decoder.Simulations have shown that the hard decision algorithm can obtain an additional gain of 1-2 dB compared with the traditional Viterbi algorithm.(2)Soft decision algorithm was simulated by Simulink.The "Complex to Real-Imag" module is used to filter out the imaginary part of the signal,it leaves only the real part of the signal carrying useful information and removes the imaginary part of the original signal.The "Uniform Encoder" module performs 8-level,3-bit quantized normalization of the data.Simulations have shown that the soft-decision algorithm can gain about 3d B more than the hard-decision algorithm.2.The design methods of the Branch Metric Generate(BMG)unit,the Add-CompareSelect(ACS)unit,the Path metric Memor(PMU)unit,the Survivor path Memory(SUM)unit and the Traceback(TB)unit in the Viterbi decoder were emphasized,and its functions were verified.3.Based on the algorithm and logic design of the encoder and Viterbi decoder,the method of the joint simulation of Modelsim and Matlab was used to verify the Viterbi algorithm.The design and verification of the entire encoder and Viterbi decoder were completed by FPGA.It is based on the design of Viterbi algorithm and in the integrated environment Quartus of Altera Corporation.After the synthesis,as an automatic layout and routing tool,the Cadence SOC Encounter is used to generate the layout.