Search

This article presents a fully energy-autonomous temperature-to-time converter (TTC), entirely powered up by a triboelectric nanogenerator (TENG) for biomedical applications. Existing sensing systems either consume too much power to be sustained by energy harvesting or have poor accuracy. Also, the harvesting of low-frequency energy input has been challenging due to high reverse leakage of a rectifier. The proposed dynamic leakage suppression full-bridge rectifier (DLS-FBR) reduces the reverse leakage current by more than 1000 $\times $ , enabling harvesting from sparse and sporadic energy sources; this enables the TTC to function with a TENG as the sole power source operating at $\mu \text{m}$ 1P6M CMOS technology, consuming 0.14 pJ/conversion with 0.014-ms conversion time.

We present a fully energy-autonomous temperature-to-time converter (TTC) for biomedical applications. This is the first work in literature to power the entire converter purely by a triboelectric energy harvester (TEG). The dynamic leakage suppression full-bridge rectifier (DLS-FBR) reduces reverse leakage current to 1/100, which enables the TEG operated by human motion at $\mu \text{m}$ 1P6M CMOS consumes 0.14 pJ/conversion while powered up purely by a TEG, achieving energy-autonomous operation.

Body-coupled powering (BCP) via the human body channel has demonstrated improved pathloss and body-area coverage over conventional approaches. This work focuses on dynamic optimization of the transmitter (hub-side) output power and frequency, to avoid power overdraw for prolonged transmitter (hub) lifetime. Power evaluation and feedback allow for the transmitter to dynamically adapt its power and frequency, based on the node-specific channel and power condition, with only ~90nW overhead. This saves the overall TX power by 7×. Moreover, HV driver with simplified driving circuits is proposed in standard CMOS process, which allows for 20× improved power delivery (at 105cm) while saving 23% of the TX power consumption. The 40nm CMOS body-coupled power transceiver (TRX) ICs delivers up to 1.1mW and 22µW power to the receiver located at 20cm and 105cm away on the body.

A 1225-Channel Neuromorphic Retinal Prosthesis (RP) SoC is presented. Existing RP SoCs directly convert light intensity to electrical stimulus, which limit the adoption of delicate stimulus patterns to increase visual acuity. Moreover, a conventional centralized image processor leads to the local hot spot that poses a risk to the nearby retinal cells. To solve these issues, the proposed SoC adopts a distributed Neuromorphic Image Processor (NMIP) located within each pixel that extracts the outline of the incoming image, which reduces current dispersion and stimulus power compared with light-intensity proportional stimulus pattern. A spike-based asynchronous digital operation results in the power consumption of 56.3 nW/Ch without local temperature hot spot. At every 5×5 pixels, the localized (49-point) temperature-regulation circuit limits the temperature increase of neighboring retinal cells to less than 1 °C, and the overall power consumption of the SoC to be less than that of the human eye. The 1225-channel SoC fabricated in 0.18 μm 1P6M CMOS occupies 15mm2 while consuming 2.7 mW, and is successfully verified with image reconstruction demonstration.

Retinal prosthesis (RP) electrically stimulates the retinal cells of the blind to restore visual loss, and it is essential to increase the number of channels for higher spatial-resolution information [1]. Conventional RP SoC with an external camera and an image processor has a limitation of increasing the number of channels due to bandwidth bottleneck, and power consumption from high supply voltage [2]. The photo-diode based RP SoC [3] can reduce the data rate by using incident light directly. However, conductive human retinal environment blurs the electrical stimulus pattern based on the light-intensity difference; thus it is challenging to meet the minimum spatial-resolution information (>1000 Ch) to recognize faces [1], [4]. This clinical result raises the necessity for integration of an image processor within the photo-diode based RP SoC. Also, cell damage due to temperature increase must be regulated (no more than 1°C increase allowed). To date, no work in the literature addresses the localized temperature regulation. In order to address these issues, we propose a 1225-channel Neuromorphic Retinal Prosthesis (NRP) SoC integrated with 56.3nW/Ch Neuromorphic Image Processor and the localized (49-point) temperature regulation circuit.

As the number of body area electronics increases, the power source is becoming a major bottleneck. Using a battery for each node is bulky and inconvenient due to the need of frequent battery replacement/recharging. Conventional harvesting methods are largely placement confined and thus unable to sustain multiple nodes of heterogeneous placements simultaneously [1], [2]. For simultaneous power delivery to multiple nodes on body, [3] utilizes the near-field inductive link for power delivery to sensor nodes placed on the eTextiles shirt and thus exhibits placement constraints, whereas the far-field RF transfer based [4] suffers from antenna pattern distortion and body shadowing effect (RF energy blocked by human body), leaving sensor nodes under such effect practically uncovered due to the 20~40 dB more channel degradation [5]. Such placement-induced nonidealities, along with distance/environment induced impedance variations and voltage/power degradations, challenges the powering system design for wider body coverage. This paper presents the full-body coverage “body-coupled power delivery” and the placement-independent “body-coupled ambient energy harvesting” ICs, achieved by the Bulk Adaptation Rectifier (BAR) and the Detuned Impedance Booster (DIB), with the power sustainability supported by the Dual Mode Buck-Boost Converter (DM-BBC).