Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Wearable and implantable medical devices (IMDs) have come a long way in the past few decades and have contributed to the development of many personalized health monitoring and therapeutic applications. Sustaining these devices with reliable and long-term power supply is still an ongoing challenge. This review discusses the challenges and milestones in energizing wearable and IMDs using the RF energy harvesting (RFEH) technique. The review highlights the main integrating frontend blocks such as the wearable and implantable antenna design, matching network, and rectifier topologies. The advantages and bottlenecks of adopting RFEH technology in wearable and IMDs are reviewed, along with the system elements and characteristics that enable these devices to operate in an optimized manner. The applications of RFEH in wearable and IMDs medical devices are elaborated in the final section of this review. This article summarizes the recent developments in RFEH, highlights the gaps, and explores future research opportunities.
Keywords: implantable medical devices, rectenna, RF energy harvesting, wearable medical devices, wireless power transfer
Biomedical devices (wearable and implantable) have expanded in popularity in integrating digital health for a diversity of biomedical applications, which includes the sensing, tracking, and monitoring of vital signals for continuous monitoring, thus making it possible for early intervention. In the healthcare sector, the advancement in these technologies is a result of several factors such as the increasing population of people with chronic diseases (such as diabetes and obesity), the aging population, and the rising need for real-time health monitoring, including fitness and wellness [1]. Recent developments in integrated circuits, medical sensors, wireless communications, and fabrication methods enable small, low-power devices with special interfacing capabilities to interact with human tissues and biological objects [2]. The platforms offered by these interfaces are appropriate for the quantitative measurement, continuous observation, and documentation of physiological and biomedical parameters, as well as for the modification of cells, tissues, or organs. Figure 1 shows different biomedical devices used in various body locations. The devices are usually powered by conventional batteries that are either rechargeable or can be replaced periodically [3]. Battery technology has advanced and now offers compact sizes with high energy capacities. However, the functional lifespan of the batteries is still relatively short and demands frequent replacement. Additionally, it is adverse and impractical to entirely power these biomedical devices with conventional batteries, especially the stretchable devices which are too small and thin to even accommodate a battery. Another issue is that the process involved in replacing the battery of these devices may trigger unnecessary side effects such as infection, inflammation, internal bleeding, and prolonged recovery [4]. Consequently, the significant challenge of these small-scale wearable and IMDs is to have a sufficient sustainable power source that is stable, biocompatible, and with no side effects. Renewable energy sources, such as RFEH technology, have great potential to fulfill the above requirements because the ambient or near-field RF energy are ubiquitous and always available. RFEH technology is based on scavenging the electromagnetic energy from RF signals. This technology enables self-sustained or battery-assisted biomedical devices with an extended lifespan. The RF wave is available in both outdoors and indoors, in urban, suburban, and rural regions. RF sources can be categorized into the following categories: ambient sources (e.g., near cell phone towers) and dedicated/nearfield sources such as Powercast TX91501-3WID in a wireless powered communication network [5]. An ambient source emits stable output power to supply wireless sensors irrespective of the transmission frequency. On the other hand, a dedicated source provides a predefined energy supply to the device by optimizing the frequency and maximum power transfer to replenish sufficient power for the sensor devices.
Implantable and wearable medical devices for various body locations.
The exponential growth in the broadcasting infrastructures and wireless communication with the advancement in the broadcasting technologies have led to an increase in the availability of ambient RF power density. The open-space medium is populated by sources of electromagnet signal such as wireless fidelity (Wi-Fi) signals [6,7,8,9,10,11], AM/FM radio [12,13], mobile base stations [14,15,16,17,18,19], television (TV), and digital television (DTV) [20,21,22,23], as described in Figure 2 . A comparison of various power sources available for energy harvesting is shown in Table 1 , considering the characteristics of these sources for energy harvesting: power density, efficiency, advantage, and bottleneck. The advantages of the RFEH technology over the other energy harvesting methods are as follows:
The widespread availability of RF energy makes it suitable to supply sustainable energy for various biomedical devices when compared with other energy harvesting technologies such as thermal, solar, and vibration.
Unlike solar, RFEH is a cost-effective technology and can be used to power the IMDs.In comparison with solar, kinetic, and thermal energy, RFEH does not depend on the light source, body motion, or temperature.
Characteristics of various sources for energy harvesting.
Reference | Energy source | Power density | Efficiency (%) | Advantage | Bottleneck |
---|---|---|---|---|---|
[30,31] | Perovskite solar cells | 35.0 μ W/cm 2 | 25.2 | Flexible and lightweight; suitable for wearable applications | Require light |
[32,33,34,35] | Thermoelectric | Human: 100 μ W/cm 3 Industrial: 100 mW/cm 3 | 10–15 | Cost-effective technology; does not require body motion or light | Low power source |
[36,37] | Acoustic | 1.436 mW/cm 2 at 123 dB | 0.012 | Require minimum maintenance; suitable to be used in remote or inaccessible locations | Hard to capture energy from the sounds wave source |
[35,38,39,40] | Pyroelectric | 3.5 μ W/cm 3 at the temperature rate of 85 °C/s @ 0.11 Hz | 1–3.5 | Cost-effective technology; ubiquitous and serves as a low-grade waste | Low output power |
[41,42,43] | Piezoelectric | 29.2 μ W/mm 3 | 83.3 | Does not require RF waves or light | low power source; require body activity |
[44,45,46] | Biofuel cells | 3.7 mW cm − 2 | 86 | The integration of the power module and sensing module results in better compactness; does not require RF waves, body activity, or light | The analyte concentration influences the power density |
[47,48] | Triboelectric | 2.5 W/m 2 | —— | Simple fabrication process and low cost | low power source; require body activity |
[38] | RFEH | GSM: 0.1 μ W/cm 2 WiFi: 0.01 μ W/cm 3 | 50–70 | Does not require light or body motion and is continuously available | Low output power; distant dependent |
Schematic illustration of the RF energy harvesting system.
A typical RFEH system comprises a receiving antenna, a matching circuit, a rectifier, a power management unit (PMU), and the load, as illustrated in Figure 2 . The receiving antenna captures the incident RF power, whereas the role of the rectifier circuit is to convert the received RF to DC. In addition, a matching circuit is required to maximize the received input power. PMU stocks and manages the harvested RF energy and also supplies the load.
The recently published review papers on energy harvesting for biomedical devices [24,25,26,27,28,29] investigate multiple technologies to power up these devices. To the best of our knowledge, this is the first review that extensively discusses and focuses on recent RF energy harvesting technologies to power up wearable and IMDs. Figure 3 illustrates the organization of this review, in which Section 2 presents the operation principle of the RFEH system, antenna design and characteristics, impedance matching network, and rectifier topologies. Section 3 highlights the application of RFEH in wearable and IMDs, followed by the conclusion.
Organization of the paper.
RFEH is enabled by ambient, dedicated, or unidentified RF signal sources. The amount of energy harvested using RFEH is dependent on the transmission power, the RF signal’s wavelength, and the distance between the RF power source and the harvesting node. The harvested power at the receiving antenna can be computed using the Friis equation as given in (1) for a transmitter and receiver antenna with a line of sight propagation [49].
P r = P t G t G r λ 2 ( 4 π d ) 2 Lwhere P r signifies received power and P t indicates transmitted power, G t denotes the transmitter of the antenna and G r represents receiver gain of the antenna, λ refers to transmitted wave wavelength, L denotes the path loss factor, and d defines the distance between transmission and receiving antenna.
Free space loss represents the loss in the signal strength and is computed by determining the distance between the transmitter and receiver, the transmitting frequency, and the antenna gain [50]. The equation to compute the free space path loss is given in (2–3):
P L = ( 4 π d ) 2 G t G r λ 2 = ( 4 π f d ) 2 G t G R c 2 P L ( d B ) = 32.44 + 20 l o g 10 ( f ) + 20 l o g 10 ( d ) − G t − G Rwhere PL stands for the free space path loss, c signifies the speed of light, and f denotes the frequency of the transmitted wave.
The performance parameters of an energy harvester (based on different applications requirements) are power conversion efficiency (PCE), operation distance, output power, and sensitivity. The PCE measures the rectifier efficiency to convert the RF signal to DC. Equations (4) and (5) are used to compute the PCE of a rectenna system [51]:
η = O u t p u t P o w e r I n p u t P o w e r × 100 η = P D C P I N = V D C 2 R L · 1 P Lwhere V DC indicates the DC output voltage, and P IN indicates the incident power at the rectenna input.
The operational distance of a harvester is reliant on the operating frequency. The DC output power describes the output of the RFEH system. The output is technically described as the minimum and maximum load voltage and current of the power management system. Due to the dependency of the load voltage and the current on the load impedance, measurement in an open load condition will accurately describe the performance of the RFEH system. In some applications, such as the sensors, the load voltage is a more critical parameter than the current, but in others, such as the LED, the predominant parameter is current. For the RFEH system, the sensitivity can be described as the optimum input power required to operate. The threshold voltage of complementary metal–oxide–semiconductor (CMOS) technology impacts the sensitivity tradeoffs and in an advanced technology node, the leakage current adversely affects the efficiency.
The free space RF energy is divided into two categories: near field and far field. The frequency range of energy transfer is from 3 kHz to 3000 MHz. RFEH can be achieved in a near field by magnetic resonance coupling [52] or inductive coupling [53]. Energy transfer during magnetic resonance coupling is achieved by the activity of two coils that resonate via magnetic coupling at the same frequency. This is in contrast to inductive coupling where energy transfer is achieved through the evanescent wave coupling between two systems that resonate at the same frequency. In such systems, the key features are the incident power density and the conversion efficiency. However, the coupling coefficient K is essential in calculating the PCE [53]. The distance of power transfer is restricted, due to coupling coefficient dependency, to the distance of the coils. In addition, the configuration requires suitable standardization and arrangement between the coils. Equation (4) is used to calculate the coupling coefficient K [54].
K = M L 1 L 2where M is the mutual inductance, and L 1 and L 2 refer to the self-inductance of coil 1 and coil 2, respectively.
Due to varying rate of attenuation, the near-field inductively coupled power propagates for a shorter distance than the far field power. The attenuation rate of the far-field propagation is 20 dB/decade, which is lower in strength than the near-field energy [55]. Multiple applications such as wearable electronics and mobile phones adopt inductive coupling wireless charging, whereas magnetic resonance coupling is utilized to charge consumer electronics and commercial appliances.
Far-field energy transfer is achieved when the distance of RF power transfer is more than λ / 2 π and enables a wide area of coverage [49]. Nevertheless, low input power and the distance between the transmitter and receiver would limit the conversion efficiency [56]. The main challenge with this method is boosting the DC power level at the rectenna’s output without boosting the transmit power, as well as energizing devices that are tens to hundreds of meters away from the transmitter [57]. As such, numerous technical efforts have been devoted to design an efficient rectenna. Table 2 compares near-field and far-field energy transmission methods.
Power transmission characteristics [58].
Features | Resonant Coupling | Inductive Coupling | Far-Field Transfer |
---|---|---|---|
Field | Resonance | Magnetic method | Electromagnetic |
Scheme | Resonator | Coil | Antenna |
Efficiency | High | High | Low to high |
Distance | Medium | Short | Short to long |
Frequency | KHz to MHz | KHz to MHz | GHz |
Power | High | High | Low to high |
Typical load | Fixed impedance | Varying impedance | Fixed impedance |
Regulation | Under discussion | Under discussion | Radio wave |
Pros | Medium efficiency in a short distance | High efficiency | Long distance |
Cons | Difficulties in preserving high Q | Very short distance | Low efficiency and safety issues |
The receiving antenna plays a pivotal role in the RFEH system. The antenna is responsible for capturing the received RF signal from the extended sources and significantly impacts the conversion efficiency. The optimization of an antenna to resonate can be at single or multiple frequency bands. A high gain receiving antenna design is required to meet the requirement of high-power input for standard applications such as recharging the mobile phone battery remotely, waking up sensors in a sleep mood, and energizing the wireless sensor network [59]. The antennas need to supply sufficient power for the applications from limited ambient signals. The desirable characteristics in the antenna design for RFEH are high gain, wide bandwidth, miniaturization while upholding the efficiency, and circular polarization for minimum mismatch loss. In biomedical applications, several other aspects need to be considered in designing wearable and implantable antennas, such as flexibility, stretchability, lightweight, and safety. Therefore, this subsection discusses the design characterization and specification of wearable and implantable antennas that are used in the RFEH system.
The integration of an antenna in wearable medical devices must be conformal and nonobstructive. The mechanical deformations of the wearable antennas affect the bandwidth and center frequency, especially when the antenna operates in a narrow band, where even a small change could detune the frequency of optimization. The wearable antenna must be isolated from the human body loading effect and absorptive losses by the ground plane. Different types of dielectric and conductive materials are used to construct wearable antennas. The selection of these materials is driven by their ability to offer a sufficient number of mechanical deformations (twisting, bending, and wrapping) with little influence from various environmental factors (rain, snow, ice, etc.), as well as adequate protection from EM radiation. Recently, several types of fabric and nonfabric materials have been used to fabricate wearable antennas; however, these fabric materials need proper characterization before being used for such purpose [60]. Given that textiles are widely used and available, several wearable antennas for RFEH are fabricated based on textile materials such as felt fabric, polyester cotton, and Cordura [61,62,63]. These textile antennas are designed with different topologies such as monopole antenna and patch antenna, operated with a single band, and dual-band. On the other hand, numerous wearable antennas for RFEH used nonfabric flexible polymer-based materials due to their stable dielectric properties. For example, the work reported by [64] proposed a dual-band dipole antenna printed on Kapton, while in [65], a monopole antenna printed on a polyethylene terephthalate (PET) film. A dualband transparent patch antenna fabricated on polydimethylsiloxane (PDMS) substrate was reported by [66].
Furthermore, wearable antennas with potential use in RFEH must be stretchable to accommodate natural skin deformations or large human motions [67]. The detuning effect of the stretchable antennas significantly reduces their resonant frequency when they are stretched [68], thereby restricting their usage in strain sensing [69,70,71]. Additionally, their integration with the commercially available chips is difficult. Besides that, the radiation efficiency of the stretchable antennas degraded significantly due to the lossy human tissues [72,73]. Consequently, stretchable antennas with high performance and insensitive resonance frequencies are required for RFEH and communication [74,75,76,77,78,79]. In this context, Zhang et al. [80] designed a stretchable microstrip antenna with different 3D layouts for RFEH. The proposed antenna was fabricated by patterning a meshed patch attached with an elastomeric substrate (Ecoflex00–30) and a meshed ground with an ultraviolet laser. The 3D microstrip antenna achieved better stretchability, strain-insensitive resonance, and improved peak gain when compared to its 2D equivalent. However, further optimizations in the design of asymmetric 3D structures are required to achieve high on-body performance resistant to various mechanical deformations. Conventional and stretchable metals are suitable radiation elements due to their inherent electromagnetic properties and high radiation efficiency. Thus, the work reported [81] proposed a laser stretchable wideband dipole antenna fabricated using conductive laser-induced graphene (LIG) patterns; the surface was selectively coated with a metal and mounted on an elastomeric substrate. The radiation performance of the stretchable antenna was evaluated depending on its tensile strain using a modified stretcher to reflect its combined mechanical–electromagnetic properties. In addition to stretching, the performance of the stretchable wideband dipole antenna is resistant to deformation modes such as twisting and bending.
The safety of the wearable antenna must be taken into account during the design phase, in addition to ensuring that a wearable antenna can blend in with the body. The specific absorption rate (SAR) is used to evaluate the electromagnetic safety of wearable antennas. The International Commission on Non-Ionizing Radiation Protection and the Federal Communications Commission (FCC)/International Electrotechnical Commission (IEC) limits the maximum general exposure to SAR to 1.6 W/kg (SAR 1 g) or 2 W/kg (SAR 10 g), respectively. SAR is computed by averaging over a sample volume, which is typically 1 or 10 g (ICNIRP), respectively.
Implantable antennas are an essential element in electromagnetic energy conversion, whether they are utilized for RFEH or data transmission in IMDs [82]. Additional difficulties beyond those of wearable antennas must be considered in the design of the implantable antennas as they will be situated inside the body. Body tissues and fluids which act as dielectric loading to the antenna are the primarily limiting factor, adversely degrading the performance of the implantable antennas. Miniaturization techniques are important in the design of implanted antennas since they can minimize their size as much as possible, increasing the possibility that they will integrate with the body. According to various scenarios, researchers have proposed numerous typologies of miniaturized implantable antennas for RFEH. The work reported by [83] developed a circular dual-band implantable antenna with a radiating slot patch; this antenna has an additional external metallic reflector that was positioned behind the human arm to improve its power transmission link. The size of the radiating metallic patch was 10.8 mm, the same as the substrate and the superstrate; this is to ensure that it would touch the surrounding tissue. A dual-band compact planar inverted F-antenna (PIFA) for RFEH was designed in [84]; the developed PIFA is equipped with a matching layer on the arm to improve the wireless power link. The size of the antenna is 16 × 14 × 1.27 mm 3 . The dual-band operation and miniaturized size of the antenna were facilitated by the use of the slit/slot loading techniques on the radiator. Another study by [85] described a self-diplexing implantable antenna for RFEH antenna that had two ports and a minimal size of 9.4 mm 3 . A common ground plane connects the two semicircular patches that make up this antenna. However, the complexity of human tissues provides that the aforementioned antennas have small impedance bandwidths; the antennas used in IMDs should have a broad bandwidth to prevent straying outside of the required band. Furthermore, antenna with wider bandwidths often present more RF power at different frequencies. Thus, the authors in [82] reported an implantable, broad dual-band antenna for RFEH. The design of the antenna involved the introduction of multiple radiating branches and etching of a C-shaped slot; multiple resonance frequencies were also produced to accomplish dual and broad bands. The size of the suggested antenna was 7.9 × 7.7 × 1.27 mm 3 . The circular polarized (CP) implantable antennas are recommended for various biomedical applications due to less polarization mismatch, minimizing multipath interference, and improvement of bit error rates [86], but when there is a need for compact size, greater performance, and biocompatibility, the design of a CP implantable antenna becomes significantly more challenging. In [87], they reported the use of C-shaped slots and a complementary split-ring resonator (CSRR) for the miniaturization of a wide-beamwidth CP implantable antenna; the size of the antenna was 8.5 × 8.5 × 1.27 mm 3 . A miniaturized CP implantable antenna was designed in [88] for far-field wireless power transmission (WPT). The antenna has a size of 11 × 11 × 1.27 mm 3 and features stub loading and capacitive coupling among the stubs. A CP implantable antenna for WPT was reported by [89]. The antenna has a miniaturize size of 7.5 mm × 7.5 mm × 1.27 mm 3 , designed by etching four C-shaped open slots on the patch. Shaw et al. [86] proposed a wideband, biocompatible, flexible CP slot antenna using a single-layer skin tissue model; the antenna was designed for use as a receiving (Rx) element; the CP antenna has a reduced size of 12 × 12 mm.
Similar to miniaturization, biocompatibility is another important aspect that should be considered during the design of implantable antennas. Biocompatibility can be defined as the capability of the material to work with proper host response in a particular application. Two methods have been proposed to ensure the biocompatibility of the implantable antennas. The first method is based on covering the implantable antenna with biocompatible materials (PDMS) to accomplish the desirable biocompatible specifications [90,91]. The second method is to use biocompatible substrates such as alumina, zirconia, and Macorin the fabrication of the implantable antenna [90,92].
Although different groups have determined SAR for implantable antennas using different phantoms, input powers, etc., the safe limits established by the IEEE, FCC, and ICNIRP remain unchanged. Implantable antennas have similar SAR values to wearable antennas; however, the depth of the implantable device inside the body is a further degree of freedom for SAR determination [93].
A comparative study along with performative analysis of wearable and implantable antennas for RFEH are presented in Table 3 , taking into account antenna characteristics: size, realized gain, operating frequency, SAR, and power conversion efficiency.
Comparison of antenna performance for wearable and implantable medical devices.
Reference | Antenna Type | Size | Substrate | Gain (dBi) | Frequency (GHz) | SAR (1 g Average) W/Kg | PCE (%) |
---|---|---|---|---|---|---|---|
[83] | Implantable dual-band miniaturized circular antenna | 10.8 mm | Rogers RO 3210 | −23.2 | 0.42–0.91 | 0.36 | 58 |
[94] | Implantable slot antenna array | 30 × 30 mm | Rogers 3010 | −26 | 0.915 | 175 | 50 |
[95] | Silicon Carbide implantable antenna | 4.5 × 4.5 mm | Semi insulating (4H-SiC) | —- | 10 | —— | 47.4 |
[96] | Quad-Band Implantable Antenna | 8.43 mm 3 | RO3010 | –34, –29.6, –28.2, –22.4 | 0.403, 0.915, 0.147, 2.4 | 0.87 | 0.67 |
[97] | Broadband Implantable Antenna | 91.44 mm 3 | Rogers 6010 | –32, –34 | 0.72–1.504 | 921 | —— |
[98] | Compacted Conformal Implantable Antenna | 48.98 mm 3 | Rogers ULTRALAM | −30.8, −19.7, −18.7 | 0.402, 0.915, 1.2 | 293.7 | —— |
[99] | Broadband Substrate-Independent Textile Wearable Antenna | 0.312 × 0.312 λ 2 | Felt and Polycotton | 2.2 | 0.9 | 1.52 | 40 |
[100] | A circular microstrip patch wearable antenna | 42.92 × 42.92 mm | Duroid 6010LM | —— | 2.45 | — | 25.5 |
[101] | Wearable Bandenna | 35 mm (outer radius) | silicone | 5 | 2.45 | —— | —— |
[102] | Folded Dipole Wearable Antenna | 0.212 × 0.212 λ 2 | Kapton | −0.3 | 0.94 | —— | 78.5 |
The impedance matching network is employed to prevent power leakage, and it guarantees maximum power transmission between the source of energy and the load. To achieve maximum power transfer, the impedance of the output antenna should be identical to the load impedance. The incident wave at the load will be reflected when an impedance mismatch leads to reduced efficiency. In the RFEH system, the receiver antenna acts as an RF source, while the rectifier represents the load. The impedance matching network concurrently functions as a low-pass filter in suppressing higher-order harmonics generated by the rectifying circuit, which can be re-radiated by the antenna, causing additional loss [103]. A perfect matching network for the RFEH should achieve impedance matching between the source and the load with minimum insertion loss and at any input power, frequency, and load resistance. Matching networks are classified into three configurations, which are L, Pi, and T configurations, as shown in Figure 4 . The commonly used configuration is L, with two components that ease the control process and design. Moreover, the quality factor Q is sustained in the L configuration. In comparison to the L network, the T and Pi matching designs are more complex. Additionally, organizing the T and Pi configurations into numerous stages preserves the final matching outcomes but will alter the Q factor. This method is advantageous for improving passive voltage boosting.
Different forms of common impedance matching networks.
The rectifier functions by converting the harvested RF signal by the receiving antenna to DC output voltage. The main design bottleneck in the rectifier’s design is to generate DC voltage from low power input. There are two types of rectifiers: diode-based rectifiers and metal–oxide–semiconductor field-effect transistor MOSFET rectifiers. When compared to MOSFET circuits, the diode-based rectifier has a lower forward voltage drop; hence, it is preferred in applications. The ability of low forward voltage diodes to achieve higher PCE makes it an optimal choice. For rectenna applications, Schottky barrier diodes are popularly adopted [104]. The half-wave rectifier is the basic rectifier topology; it is composed of a single diode D1, as depicted in Figure 5 a. However, for most applications, a half-wave rectifier is insufficient despite its simplicity of design. Therefore, the full-wave rectifier is preferred. Figure 5 b depicts the circuit design of a full-wave rectifier. Alternatively, the bridge rectifier (depicted in Figure 5 c) exhibits better reverse breakdown voltages, higher power efficiencies, and fewer output ripples compared to its counterparts [105].
Various rectifier topologies: (a) circuit diagram of half-wave rectifier; (b) circuit diagram of full-wave rectifier; (c) circuit diagram of bridge rectifier; (d) circuit diagram of three-stages Cockcroft–Walton voltage multiplier; (e) circuit diagram of four-stages Dickson voltage multiplier; (f) circuit diagram of four-stages Dickson voltage multiplier utilizing CMOS technology.
Applications driven by high power need are enabled through a rectifier, known as the voltage multiplier, in boosting of AC to DC conversion. The voltage multiplier encompasses numerous single rectifiers connected in series [106]. The Cockcroft–Walton voltage multiplier is the commonly used topology (depicted in Figure 5 d). This rectifier is similar to a full-wave rectifier, which integrates numerous stages to achieve higher voltage gains. A modified version of the Cockcroft–Walton’s configuration is the Dickson multiplier, shown in Figure 5 e, in which a shunt capacitor is integrated to minimize parasitic effects. The Dickson multiplier is popularly considered for low-power devices despite the tradeoffs in high power conversion efficiency due to the leakage current by the high threshold voltage within the diodes. Furthermore, the output voltage shows a significant reduction under high resistive load, thereby reducing the current supply. Nevertheless, the MOSFET-based diode has a fast switching speed but suffers from electromagnetic interference and thermal runaway; a high threshold voltage is also needed, restricting RFEH circuits’ performance [107]. The Dickson multiplier is suitable for low-power applications enabled by MOSFET technology; it can be realized in system on chip (SoC) solutions by substituting the diode with a negative channel metal oxide semiconductor (NMOS), as depicted in Figure 5 f.
The widespread availability and the diversity of RF energy sources (external and ambient) and the exponential growth in Internet of Things (IoT) and wireless sensor nodes has encouraged recent reported research work to employ RFEH techniques to energize low-power wearable and implantable medical devices. Biomedical devices gained widespread application in healthcare sectors due to the favorable feature of continuous monitoring of vital signals from the human body or apparent spurs from the internal body organs. RF energy is a prominent integral block for wearable and IMDs in sustaining the lifetime of the battery in use for an extended period, relaxing the need to change or recharge the battery frequently. This is enabled through the development and integration of ultra-low-power electronics, which decreases the power consumption of the primary sensor components to less than a milliwatt [108]. The current power consumption range of the wearable and IMDs is (nW to mW), as shown in Table 4 . Therefore, RFEH technology is an appropriate solution to power these devices. In this section, we will discuss the implementation of the RF energy harvester in wearable and IMDs.
Average power requirements and specifications of various biomedical devices.
Reference | Biomedical Device | Size | Power Consumption | Application |
---|---|---|---|---|
[109] | Pulse oximeter | 38.69 cm 3 | 294 mW | Measures the oxygen saturation level |
[109] | Hearing aid | 2.45 cm 3 | 1.82 mW | Amplifies the sound for the patient with hearing loss |
[110] | Cochlear implant | 9.58 × 9.23 mm | 100–2000 μ W | Stimulates the cochlear nerve electrically |
[109] | Pacemaker IC | 4.9 cm 3 | 0.28 mW | Monitors heart rate |
[111] | Drug pump for ophthalmic use | 9.9 × 7.7 × 1.8 mm | 400 μ W | Controls drug delivery |
[112] | Neural activity monitoring recorder | 3 × 3.5 mm | 1–10 mW | Records brain activities |
[109] | Combo insulin pump | 97.61 cm 3 | 15 mW | Delivers insulin using an insulin pump while also monitoring blood glucose levels and giving bolus instructions using a blood glucose meter |
[113] | Wireless intraocular pressure monitor | 0.5 × 1.5 × 2 mm 3 | 3.65 nW | Frequent measurement of intraocular pressure |
[114] | Health monitoring sensor on wristband | —— | 0.83 mW | Monitors chronic respiratory disease |
[109] | Wireless EKG system | 40.13 cm 3 | 60 mW | Monitors and records vital signs and cardiac information |
[115] | Electrocardiogram amplifier | —— | 2.76 μ W | Transforms the weak electrical signals from the heart into signals that can be transmitted to a monitoring system. |
[114] | Electronic-nose sensor system | —— | 250 μ W | Connected to a pattern-recognition system that responds to odors passing over it |
[114] | Spirometer | 6 × 6 mm | 0.01 mW | Measures FEV1, PEF, and FVC |
[116] | Cardiac activity sensing | 6 mm | 0.3 μ W | Monitors vital signs |
[117] | Retinal prostheses | Diameter = 3mm Pixel size = 25 μ M | 250 mV | Stimulates the retina |
[114] | ECG chest patch | 84 × 39 × 8.3 mm | 0.96 mW | Measures electrocardiogram ECG, skin impedance, photoplethysmography (PPG), motion, and acoustic signals |
Wearable medical devices contribute significantly to improving the current medical system and supporting physical activities due to their ability to offer extensive monitoring of different physiological functions, emergency alerting, and computer-aided rehabilitation. These devices can be used in the healthcare sector and military sector for evaluating the mental and physical health of soldiers in combat [118,119]. This current global pandemic outbreak has significantly impacted the global healthcare system and, as such, has necessitated the deployment of remote monitoring systems to monitor its symptoms and perform essential tracking [120]. Researchers have over the years introduced several IoT-based devices and wearables for accurate diagnosis and monitoring of patients in the prodromal phase; these devices can monitor the temperature, respiratory rate, and heart rate of the patients and provide real-time data for accurate treatments. Such devices are low in power consumption (ranges from pW to μ W) and can be efficiently powered using ambient RF sources. Low-power devices include temperature sensor devices (power consumption ranges from 113 pW–1.4 μ W) [97,121,122,123,124], biosensors, voltage, and current sensors (power consumption ranges from 9.3 nW–436 μ W) [125,126,127]. Suggestions have been made for ambient RF power harvesting to power these devices instead of batteries [128,129]. RFEH systems provide a controllable and continuous source of energy. However, the implementation of RF harvesters to power wearable devices faces some issues, such as their ability to supply an adequate amount of energy and the deterioration in power transfer efficiency of far-field RFEH systems. Several innovative methods to design RFEH systems for energizing various types of low-power wearable medical devices are reviewed in the following subsections.
The environmental friendliness and low-cost fabrication processes of additive manufacturing, such as 3D printing and inkjet printing, have increased its industrial relevance, as these emerging fabrication techniques are designed to significantly reduce the required number of manufacturing steps, such as the elimination of the etching processes, thereby improving the fabrication efficiency. Most of the time, wearable sensor devices are preferable in areas where disposable or single-time use is hygienically sorted, such as in hospital settings. This application type requires the creation of numerous circuit components within the device, using additive manufacturing to reduce the device cost. The improvements recorded recently in fabrication and performance have endeared inkjet printing technology to sensor and RF applications. In this regard, a design of enabling near-field RFEH for wearable sensors with an additional fabrication process was proposed in [130]. The work fabricated circuit prototypes through a combination of conductive traces developed with conductive inkjet printing and masking technologies with lumped circuit components. Furthermore, the S-parameters were used to estimate the input power for the RF to DC conversion circuit; the circuit produced a peak output power of 0.146 W with a H-field harvester and 0.0432 W with an E-field harvester. The harvesters (E- and H-field) were subjected to several operation tests using a microcontroller communication module and an LED in both on-bottle and on-body bent/flex conditions to validate the functionality based on the energy from a two-way talk radio. The outcome of these tests ensures the compatibility of the developed inkjet-printed flexible energy harvesters to be adopted in wearable biosensors. Besides that, the work in [131] presented a flexible and wearable energy-autonomous on-body sensor network with complete RFEH operability using a handheld 464.5 MHz UHF band two-way talk radio, which was created using 3D printing and inkjet processes. The system is equipped with two different types of energy harvesters; the first one was attached to the sensing-capable backscattering RFID tags for the harvesting of the 464.5 MHz signal energy that will power the tags; the second type of energy harvester was attached to the hands of the wearer for the harvesting of the same 464.5 MHz signal to provide both the carrier signal and the DC power. The efficiency of this second energy harvester is higher than that of standard ambient energy harvesters. Being that the rectifier’s DC and second harmonics were used to support two extra functions, energy harvesting designs were used in the system. The estimated DC power of the energy harvester is 17.5 dBm, while the second harmonic output (SHO) is 1.43 dBm when a two-way talk radio is placed at a distance of 9 cm. Using the harvester-powered RF amplifier, the measured SHO power is boosted to 13 dBm.
To achieve particular structures, traditional 3D printing technologies often demand a large amount of time and support material. Therefore, the work by [132] proposed a fabrication approach to design an origami RF harvester system based on origami folding principles for use in high-frequency applications. This approach was devised such that it considerably decreases the amount of time required for the fabrication and removes the necessity for support material. The method entails fabricating a planar structure using 3D printing technology and then using inkjet printing to directly produce conductors on its surface. The inkjet-printed on-package conductive features were manufactured successfully and integrated with RFEH electronics to demonstrate the effectiveness of using origami techniques to build completely 3D RF systems. At the input power of 0 dBm, the RF energy harvester achieved a DC output voltage of 1.2 V. Furthermore, 3D printing can be a good way of solving huge parasitic problemsfor traditional package structures that have a major impact on the performance of the system [133,134]. An embedded-on-package 5G RF energy harvester that operates at 26 GHz was presented by [135]; this system performs within the 3D printed multilayer flexible packaging. The entire system is made using low-cost, quick-to-prototype additive manufacturing processes such as 3D printing and inkjet printing. In this system, the use of the isotopically radiated power EIRP that is permitted for 5G communication (which is 75 dBm) [136] rules out the issue of low power density for RF energy [137]. At a 20 cm gap from the source, an energy output voltage of 0.9 V was harvested with a transmitted equivalent (EIRP) of 59 dBm, whereas a >1 m range is anticipated when applying the full 75 dBm EIRP for 5G communication. Another work by [138] used a fabrication process that relies on a masking process that is inkjet-printed and then etched. The work focused on improving the capability of typical shunt Schottky diode topologies that are deployed for energy harvesting in rectenna systems. This was accomplished by providing both flat and high-power conversion capabilities across the whole frequency range of interest. The system is a dual-tapered transmission-line-based matching network for the improvement of the rectification capability of the embedded Schottky diode. The rectenna is made up of a rectifier and a tiny monopole antenna that is optimized for a liquid crystal polymer (LCP) substrate. At an input power of 0 dBm, the flexible rectenna achieved a maximum PCE of 40%.
Wearable devices may benefit considerably from embedded tiny electronics and conductive materials to usher in the smart clothing paradigm [139]. Rectennas made of textiles can be used in body-centric wireless communication systems that do not require batteries. Studies have reported the fabrication of many antennas made of textile materials for wireless communication [140,141]. To create a wearable device for power harvesting, some works have paired a textile antenna with a rectifier circuit manufactured on a printed circuit board (PCB) [61]. Several rectennas have been designed for single- or dual-band operation. For example, the work reported by [142] described a single-band textile rectenna that operates in the 4.65 GHz frequency band made with jeans cotton as the substrate, while the radiating element is copper tap. The textile rectenna achieved a DC output voltage of 400 mV with a maximum PCE of 55% at an input power of −5 dBm. A dual-band operating textile rectenna was developed by [143] in which the antenna is a sub-1 GHz (785–875 MHz) broad-beam rectenna with a 2.4 GHz off-body antenna. The PCE of this rectenna was 63.9% and the DC output was 650 mV from a power density of less than 0.8 μ W/cm 2 . Next, in [144], a design of a complete textile-based rectenna system was presented. This rectenna is laden with a novel power management platform enabled independently through RF power harvesting. This rectenna operates at a frequency band of 900 MHz and 1800 MHz, and the Wi-Fi band at 2.4 GHz. The RF properties of the textile rectenna are verified through nonlinear techniques, where textile materials and antenna layout are numerically characterized through electromagnetic simulations. The system is entirely autonomous and is operational even without the use of a battery at a low RF power level of −15 dBm.
To increase the output power of a rectenna, numerous research efforts have been devoted to designing a textile rectenna array. The authors in [145] introduced textile-based rectenna arrays for the provision of power in wearable electronic devices. The system is operated at 2.45 GHz and consist of 2 × 2 and 2× 3 rectenna components. Each of the elements consist of a rectifier and a patch antenna created from fabrics using conductive thread embroidery. At −10 dBm input power level, this rectenna achieved a PCE of 35%, whereas the average DC power of the 2 × 3 rectenna was roughly 80 μ W at 60 cm from the source. However, the patch antennas and rectifying circuit were on the same layer, resulting in a considerable element-to-element distance. Thus, Chi et al [62] presented a 2 × 2 wearable textile-based rectenna array with a stack architecture for each rectenna to minimize the dimensions of the elements in the array. The rectenna element comprises a linearly polarized patch antenna and a single-stage full-wave Greinacher rectifier which are fabricated using Cordura textile material. The rectenna array was positioned on the human body at a distance of 150 cm away from the indoor access point (Wi-Fi). The maximum DC output voltage of the rectenna array was 1.05 V at an input power of 20 dBm. Juan et al. [146] developed a 2 × 2 textile rectenna array elements for RFEH based on the combination of the electromagnetically coupled microstrip patch antenna and a simple and precise construction method to improve the performance of multilayer microstrip textile patch antennas. Therefore, the patches and rectifier circuit pads were laser-cut, and the various layers were joined together with double-sided, thermally activated adhesive sheets. The patch antennas and the rectifiers were fabricated on pure copper polyester taffeta fabric. The on-body performance of the 2 × 2 textile rectenna array was verified through perfect alignment with the Tx antenna, as depicted in Figure 6 . The on-body measurement of the proposed system obtained a maximum output power of 1.287 mW with a maximum PCE of 40% at an input power of 12 dBm. Estrada et al [147] proposed 16- and 81-element broadband rectenna arrays based on a strongly coupled bowtie antenna screenprinted on a cotton T-shirt for harvesting power densities of 4–130 μ W/cm 2 between 2 and 5 GHz. The broadband rectenna achieved a PCE of 32% at an incident power density of 100 μ W/cm 2 .