IEEE Trans Biomed Circuits Syst. 2020 Jun;14(3):441-451. doi: 10.1109/TBCAS.2020.2972733. Epub 2020 Feb 10.
This article reports an implantable transcutaneous telemetry for a brain machine interface that uses a novel optical communication system to achieve a highly energy-efficient link. Based on an pulse-based coding scheme, the system uses sub-nanosecond laser pulses to achieve data rates up to 300 Mbps with relatively low power levels when compared to other methods of wireless communication. This has been implemented using a combination of discrete components (semiconductor laser and driver, fast-response Si photodiode and interface) integrated at board level together with reconfigurable logic (encoder, decoder and processing circuits implemented using Xilinx KCU105 board with Kintex UltraScale FPGA). Experimental validation has been performed using a tissue sample that achieves representative level of attenuation/scattering (porcine skin) in the optical path. Results reveal that the system can operate at data rates up to 300 Mbps with a bit error rate (BER) of less than 10 , and an energy efficiency of 37 pJ/bit. This can communicate, for example, 1,024 channels of broadband neural data sampled at 18 kHz, 16-bit with only 11 mW power consumption.
这篇文章报道了一种可植入的经皮遥测脑机接口,它使用一种新颖的光学通信系统来实现高效能的链路。基于脉冲编码方案,该系统使用纳秒级激光脉冲实现数据速率高达 300Mbps,与其他无线通信方法相比,所需的功率水平相对较低。该系统使用离散组件(半导体激光器和驱动器、快速响应硅光电二极管和接口)与可重构逻辑(使用 Xilinx KCU105 板上的 Kintex UltraScale FPGA 实现的编码器、解码器和处理电路)相结合,在电路板级集成来实现。使用在光路上具有代表性衰减/散射水平(猪皮)的组织样本进行了实验验证。结果表明,该系统可以以数据速率高达 300Mbps 的速率工作,误码率(BER)小于 10 ,能量效率为 37pJ/bit。例如,它可以仅用 11mW 功耗来传输以 18kHz 采样的、16 位、带宽为 1024 个通道的宽带神经数据。