Body-Coupled Communication for Microscale Implants In Human Brain

This thesis investigates the application of body-coupled communication (BCC) for microscale implants inside the human brain. Brain-Computer Interface (BCI) platforms have received a lot of attention in the last decades, demonstrating to be an effective means of restoring sensory and motor functions (e.g. sight, movement, and cognitive abilities), in addition to their utility in neuroscience studies. Developing technologies to enable distributed massive neural interfaces is an active area of research to enhance the efficiency and efficacy of BCI systems. To this end, both wired solutions (e.g. Neuralink project), and wireless networks of micro-scale implants are explored. Most wireless solutions utilize far-field electromagnetic radiation, i.e. RF, for communications between implants. Alternatively, this thesis explores BCC as a promising technology for communication between distributed micro-scale implants. BCC intrinsically offers lower power consumption, higher efficiency, low interference, and higher security compared to RF solutions. In this thesis, the applicability of galvanic BCC is investigated, which is one of the popular BCC methods that involves coupling a rather low-frequency differential signal directly to the body through a pair of electrodes. Since the BCC transmission efficiency is highly dependent on the system’s geometries and the electromagnetic properties of the tissue, finite element models (FEM) in COMSOL Multiphysics are heavily employed in this thesis. This enables the study of the impact of various parameters such as electrode size and orientation, implant location, operating frequency, and the number of devices in a network of miniaturized implants. FEM simulations revealed that for an electrode size of 200µm, the highest power gain can be achieved at frequencies above 50MHz, which has a good margin to the model’s validity limit. Thus, this thesis presents some noteworthy findings all at 50 MHz frequency. Firstly, concurrently scaling the inter-electrode distance in implants ("De"), channel length ("Lc", distance between two implants) and electrodes' dimension with a constant ratio, revealed that the voltage gain ("Gv") decreases from around -37dB to -47dB as De is reduced from 5cm to 1mm. This suggests that scaling of BCC for a network of micro-scale implants comes at the cost of more power consumption or a more sensitive receiver, but it is at an affordable level. Next, keeping De constant at 1mm, as Lc is increased from 2mm to 5cm, Gv drops exponentially from about -47dB to -120dB due to the natural decay of the electric field intensity in the near-field region. On the other hand, increasing De from 0.5mm to 2mm, while maintaining Lc at 5mm, leads to a rise in Gv from approximately -81dB to -59 dB. Additionally, the study has found that the BCC design demonstrates good robustness against misalignment. An angular misalignment of up to 60°, and a lateral displacement of up to 1mm result in the maximum Gv attenuation of less than 5dB, and 2dB, respectively. Furthermore, the thesis covers a discussion on the frequency-dependent behavior of each tissue, the polarization impedance around the electrodes, the devices' equivalent circuits, and quasi-static approximation analysis. Based on the results obtained, the trade-offs between gain, maximum frequency and data rate, input impedance, and safety implications are explored and discussed.