| Currently there are more than 40 million blind people in the world.Because of incurable retinal degenerative diseases like RP?Retinitis Pigmentosa?and AMD?Age-related Macular Degeneration?.These are the predominant diseases which can make the vision gradually decay until lost,but the internal layers of the eye are still healthy.Visual prosthesis is an electronic device that bypass the damaged parts of the retina by electrically stimulating the surviving active parts of retina i.e.ganglion cells using electrical pulses routed through an array of electrodes.Ganglion cells generate a nerve impulse and transmit to the optic nerve,which shifts it to the brain where the visual percepts are interpreted by electrical signals.In the visual prosthesis system,the transmission of data and energy through the high-order modulation and demodulation technology with the a pair of coils wireless transmission method not only solves the contradiction between the transmission rate and the efficiency,but also takes into account the size of the surgically implantable area of the retina.However,due to the influence of noise in the transmission channel,error bits are likely to occur.Therefore,the LDPC?Low-Density Parity-check Code?error correction code is mainly studied in this paper.Firstly,this paper makes a theoretical comparison between the encoding method and decoding algorithm of LDPC codes to determine the encoding and decoding scheme,and performs the collaborative simulation of the encoding and decoding algorithms.Then according to the simulation results,a method of adopting a fixed iteration number in the decoding process is proposed,and an offset of 0.125 for simulation based on the MS?minimum sum-product?algorithm is selected,So that the coding gain of the basic MS algorithm is increased by 0.4 dB.Secondly,in combination with the high-order modulation and demodulation scheme,the circuit model of the error correction code system is built by using SIMULINK and implemented in hardware.According to the check matrix of LDPC code and some calculation rules,the IEEE802.16e-based encoding algorithm is improved to simplify the computational complexity and eliminate some matrix inversion and sum operations.In the decoding circuit,it is proposed to store the address in advance and update the read/write information of the storage unit,thereby greatly saving the storage space and reducing the amount of calculation.Finally,the ber?bit error rate?simulation results show that the ber of the output signal of the error-correction code circuit?ECC?is lower than that of the demodulation circuit at the same bit-to-noise ratio.At a ber of1×10-5,the coding gain of the ECC circuit is increased by about 5 dB.Simultaneously,FPGA?Field Programmable Gate Array?experimental results show that when the demodulation circuit outputs errors,the ECC circuit can correct errors and output a correct code word.Therefore,the circuit realizes the function of correcting errors.The hardware implementation complexity of the ECC circuit is low and this circuit provides a guarantee for the communication reliability of the visual prosthesis system. |
PDF Full Text Request