-
An Implantable Low-Power Wireless Energy Harvesting Neural Recording and Stimulation System
-
Wireless Power Transmission to mm-size Biomedical Implants
-
A Low-Noise Amplifier for Recording Biomedical Signals
-
An Ultra-low Power Analog to Digital Converter
-
An Ultra-low Power Transceiver for Energy Harvesting Applications
An Implantable Low-Power Wireless Energy Harvesting Neural Recording and Stimulation System
I started working on this topic as my 599 research project. 599 project is equivalent to Ph.D. qualifier exam and I officially became a Ph.D. candidate at Rice University after clearing my qualifier exam at the end of my first year at Rice.
Diagnosing different disorders and even restoring physiological functions can be done by monitoring brain neural signals. Research on the anatomical, physiological, and computational bases of speech production has made important strides in recent years but has been limited by a glaring lack of information on the dynamics of the process of speech production. This is a methodological limitation related to the low Spatio-temporal resolution of widely available tools such as fMRI, EEG, behavioral, and stroke lesion based approach. At the current time, intracranial-electroencephalographic (icEEG) recordings in humans are carried out using either surface subdural grid electrodes (SDE) or depth electrodes (DE), both of which are designed for use in localizing seizures in patients with medically intractable epilepsy. These recording systems have yielded valuable insights into the ultra-small structural scale at which phoneme production and decoding in humans occurs.
While such recordings provide excellent temporal resolution, the spatial resolution is 
optimized for current clinical needs and not for fine scale decoding of cognitive and motor processes. Additionally, the systems capable of recording at fine scales (e.g. Utah Array) are penetrating and connected to a recording system via cables that are external to the head, limiting patient mobility and providing a portal 
for central nervous system (CNS) infections. We propose to design an exploratory wireless system with long-term potentials for use in humans for ultra-high resolution micro-electrocorticography (mECoG). Recordings and stimulation at this scale will pave the way for novel neural prosthetic approaches to remediate language impairment.
A conceptual view of the and a block diagram of the proposed system are shown.The system operation is fully wireless which means that there is no transcutaneous wire connection for power and data transfer. The power and data are transferred to the system using Radio Frequency waves. The systems harvest the required power for the operation and send the recorded signal to the reader which is located outside of the patient brain.
Wireless Power Transmission to mm-size Biomedical Implants
Power transfer is a key challenge in the application that there is no wire connection or a battery such as mm-size biomedical implants or sensor networks (IoT). Traditional wireless systems use an off-chip antenna for receiving electromagnetic waves and use it as the source of power. The off-chip antenna is few centimeter in size. Miniaturizing the system requires the receiver antenna to be implemented on a single silicon chip with other parts of the system and be limited to few mm in size. We have developed a system for harvesting required a power of biomedical implants with an on-chip antenna. In order to maximize the power transfer efficiency to the system, the optimum frequency is determined after simulating and studying the effect of intermediating biological tissues, antenna efficiency, and voltage rectifier conversion efficiency. An on-chip matching network matches the receiver antenna to the voltage rectifier and improves the power transfer efficiency. The power harvesting system has been fabricated in 180 nm CMOS SOI technology and the chip micrograph is shown below.
A Low-Noise Amplifier for Recording Biomedical Signals
The signals that are generated from activities of body organs in human or animals are usually low-frequency signals and their amplitude is limited to tens of mV. The amplitude is not enough to be passed to an Analog to Digital (ADC) converter for digitization. Therefore, a Low-Noise Amplifier (LNA) is used for amplifying biomedical signals before digitization. We have designed an extremely low-power LNA that can be used for a scalable recording system. The amplifier can reject the DC offset caused by electrochemical interactions between biological tissues and the recording electrodes while it amplifies frequency components higher than 0.1 Hz. The designed LNA has been fabricated in 180 nm CMOS SOI technology.
An Ultra-Low Power Analog to Digital Converter
In order to eliminate the effect of noise, biasing and environmental conditions on the biomedical signal, they need to be digitized with an appropriate resolution to be ready for digital processing. Converting the recorded analog signals to digital signals is done by an Analog to Digital Converter (ADC). We have designed an ADC based on Successive Approximation Register (SAR) structure. The main concern in the design process is to minimize the power consumption to make the designed ADC compatible for a scalable recording system. The designed ADC is capable of sampling the biomedical signals with a 10-bit resolution having 100 KSps while t only consumes 20 nWand it is about to be fabricated.
An Ultra-Low Power Transceiver for Energy Harvesting Applications
Data telemetry is one of the most power hungry tasks in wireless systems especially when the data rate is increasing. In a scalable recording and stimulation system, the recorded data from numerous recording nodes should be transferred via a wireless uplink and stimulation data should be received via a wireless downlink as well. We have started designing an ultra-low power transceiver for data telemetry purposes.