IEEE 802.11b/g Practical Assessment Using a Low-Cost Quasi-Yagi Rectenna for Indoor RF Energy Harvesting

This article presents an experimental analysis of the use of signals transmitted by Wireless Local Area Networks (WLAN) based on the IEEE 802.11b/g standards for Radio Frequency Energy Harvesting (RFEH) applications in indoor environments employing a Quasi-Yagi Rectenna (QYR) topology. This analysis is a crucial point to distinguish the use of the IEEE 802.11b/g standards for ambient RFEH applications and contributes to the debate on the use of different types of energy available for free in modern society. For it, a dedicated setup, developed in a controlled environment to avoid any external interference, was built to carry out the programmed measurements. The results obtained indicate that IEEE 802.11b/g standards have great potential for applications in RFEH, with the IEEE 802.11b average power four times higher than IEEE 802.11g at the rectifier output.

IEEE 802.11b and IEEE 802.11g networks was analyzed and the time to charge a supercapacitor was optimized based on the available energy, but without highlighting the differences in the use of each standard.
This shows that several RFEH studies using non-CW transmissions, based on the IEEE 802.11b and IEEE 802.11g standards, are still in progress. Typically, the choice of IEEE 802.11g over IEEE 802.11b is justified, for communication applications, among other things, by its higher data rate.
However, this feature cannot be preferable for RFEH applications. Thus, due to the current high penetration of IEEE 802.11b and IEEE 802.11g standards for WLAN applications in different indoor environments (e.g. homes, businesses, and industries), it is essential to evaluate them separately to verify their operation from the point of view of the rectenna and RFEH performance.
In this context, to analyze the differences between these two standards for indoor RFEH applications, this work uses a low-cost Quasi-Yagi Rectenna (QYR) prototype and a common commercial WiFi router (compatible with both standards) to measure, inside an anechoic chamber, the variations of the output voltage (Vo) of the rectenna over time at different distances. Thus, this study can be used as a starting point for the analysis of RFEH from signals of communication systems operating in bursts and for the study of traffic optimization of these systems for RFEH applications.
Therefore, the comparison between the IEEE 802.11 standards aims to show how differences in signal transmission, such as frame structure, transmission rate, and modulation, influence RFEH performance.
This article was organized in the following order. In addition to this introduction, a description of the QYR prototype and its performance evaluation in a controlled CW environment are presented in section II. In section III, the essential characteristics of the IEEE 802.11b and IEEE 802.11g standards for RFEH applications are highlighted and the measurement system setup is characterized. In section IV, the obtained results are presented and analyzed in detail. Finally, in section V, the main conclusions are presented.

II. QYR PROTOTYPE
In this experimental analysis, the design of a QYR based on a Quasi-Yagi antenna (QY) was chosen, since this antenna has good characteristics of radiation pattern and gain [24] , which makes it a good option for indoor applications.
The overall QYR structure is composed of the QY associated with a series rectifier circuit, connected by a SMA male to SMA male connector, as shown in Fig. 2. The use of a series rectifier circuit was chosen to increase the rectenna efficiency in low levels of the RF received power [25]. All the QYR elements are designed over the same low-cost 1.6 mm-thickness FR-4 substrate with a tan δ (loss tangent) of 0.02 and εr (relative permittivity) of 4.5. The rectifier and QY designs were optimized with the Keysight Advanced Design System (ADS) software.

A. Rectifier
The rectifier input impedance and efficiency, for the specific frequency and power operational ranges, are critical parameters for any rectenna development. Aiming these characteristics as the main design goal, the chosen circuit topology is based on the series rectifier, as shown in Fig. 3a. This rectifier is often addressed in the literature [5] , [26] , [27] and it presents a higher efficiency when compared to voltage multiplier rectifier topologies, for low input power levels (Pin) [25]. Additionally, when it is compared with more complex rectifier topologies, as voltage doubler and full wave rectifiers, it requires a lower number of components, also contributing to decrease the prototype cost.
The layout of the rectifier is shown in Fig. 3b (for more design details, see [28]). The series rectifier output filter is composed by a choke inductor (Lo) connected in series with the load (Rl), and a microstrip radial stub capacitance (Co). The SMS7630-079LF diode (D) parasitic capacitance and inductance are, respectively, defined as Cp and Ls [29], [30]. The input impedance is set to 50 Ω, making use of a microstrip transmission lines (Zline), and the QY is modeled by its equivalent Thévenin circuit, formed by Vth and Zth. Both, Rl (optimum load of 3.3 kΩ) and Lo (10 μH), are general purpose Surface Mounted Device (SMD) components usually used in RF front-end circuits [28].
The RF to DC conversion efficiency (ηRF−DC), which is a function of the frequency, Pin, and Rl [30], is defined by (1), where Po is the output power over Rl.
Consequently, to evaluate ηRF−DC, the Keysight N931A RF signal generator is connected to the rectifier, and Vo is measured using the Tektronix DMM4040 precision multimeter. In addition, to confirm the output power levels of the N9310, whose small variations can cause a strong impact on the efficiency measurements, the power levels are calibrated using a Rohde & Schwarz NRP-Z91 power sensor. Furthermore, to demonstrate that the circuit is also matched for Pin levels around the design point, it becomes important to measure the reflection coefficient (S11) parameter related to Pin, using a Rohde & Schwarz ZVB 8 Network Analyzer and sweeping Pin from -30 to -10 dBm, as shown in Fig. 4, with the best |S11| matching at -18 dBm. The ηRF−DC of the series rectifier is also presented in Fig. 4, resulting in efficiency over 18% at -20 dBm. The rectifier is matched for the whole analyzed Pin range at 2.45 GHz, considering |S11| ≤ -10 dB, where more than 90% of the power supplied will be transmitted to the rectifier [31].

B. QY Antenna Design
The QY is a classical radiating element that was proposed by the first time in [32] . Its structure is based on the classic Yagi-Uda antenna topology, featuring a dipole as a driver element, a parasitic structure that works as a director and a truncated ground plane reflector [24] . The combination of resonant elements as the driver, director, and truncated ground plane reflector designed for slightly different frequencies, in addition with a broadband microstrip to Coplanar Stripline (CPS) transition (balun) make this element a microwave broadband device [33][34][35].
The QY layout is adapted from previous work [36] , presenting a total area of 110x105 mm 2 . The QY layout with its final dimensions is shown in Fig. 5. As a broadband antenna, the QY design optimization criteria are not associated to just one specific frequency, but to the entire operational band. Therefore, to attest its good performance, a current density analysis was implemented and the results can be seen in Fig. 6., including the adopted 2.45 GHz RFEH frequency (Fig. 6c). The |S11| of the QY and the rectifier circuit as a function of frequency, within the range of interest, can be seen in Fig. 7. It is important to note that due to the nonlinear behavior of the diode impedance, the rectifier measurements were done for three different Pin levels, centered at -20 dBm with a 20 dB span. The bandwidth of the QY, with three resonant deeps, according to Fig. 7 is approximately 1 GHz (light shade). This bandwidth (region |S11| ≤ -10 dB for 50Ω impedance) is adequate to ensure that any displacement of the S11, produced by Pin variations due to the non-linear behavior of the Schottky diode, does not cause any significant impedance mismatching between the antenna and rectifier [31] .
In addition, the bandwidth of the IEEE 802.11 standard operating in 2.4 GHz (dark shading) is entirely covered by the bandwidths of the rectifier and the QY in the Pin adopted range, as desired in a real operating condition.
The QY antenna gain was measured in the anechoic chamber (ETS-Lindgren -model Spacesaver are presented in the Fig. 8 and Fig. 9, respectively. In Fig. 8, the E-plane and H-plane (X-Y and Z-Y, respectively, in reference to Fig. 2) are in evidence. In the Fig. 9a is presented the total field analysis, while Fig. 9b shows the low cross-polarization for this prototype, which is less than -15 dB. These results endorse the adoption of this planar antenna for the proposed analysis. To calculate the ηRF-DC of the QYR (Fig. 10b), the Pin values used in (1) (Fig. 10a) In order to carry out the proposed analyzes, a router with high configuration capacity is required.
Thus, the D-Link DIR-610N + commercial router was chosen, which allows changing several parameters of both evaluated standards, such as the choice of the standard to be used, the operating channel, the bandwidth and the maximum Effective Isotropic Radiated Power (EIRP). In this study, for a fair comparison, the same parameters values were selected for both standards.
Aiming to analyze the differences in RFEH performance of using the IEEE 802.11b and IEEE 802.11g standards, it is essential to characterize the measurement setup in an interference-free environment, avoiding a wrong understanding of the characteristics of the standards for the specified boundary conditions and to ensure repeatability.
In this way, the measurement setup was also built inside of the anechoic chamber of the LIC (previously used for CW measurements), as described in Fig. 11a. In order to ensure that the router transmits signal continuously, two laptops were connected to the WiFi network, forming a support link so that one of them transmits data continuously to the other. Both were positioned 1.5 m away from the router and in the opposite direction of the rectenna. Its position and the use of absorbers ensure that the signals from the laptops reach the QYR with attenuation higher than 50 dB in relation to the signal received from the router, in addition to maintaining the signal power transmitted by the router at maximum.
A client-server software was developed to establish a communication link between the two laptops over the WiFi network (support link) using a specific IP address (Internet protocol). Using this software, it is possible to send any desired information from one computer to another via the WiFi router. To ensure continuous data transmission through the router during measurements, a long, high entropy data file was used.
On the measurement link, the QYR and the router were aligned, so that the direction of highest gain of the QYR pointed towards the router. The Vo measurements were performed for a separation of 0.25, 0.5, 1 and 1.5 m between the QYR and the router. The multimeter, used to measure Vo, was positioned about 1 m from the link. The measurement and support links are shown in Fig. 11b and Fig.   11c, respectively. IV. ANALYSIS OF THE RESULTS As an initial evaluation, the signal transmitted by the router for a bandwidth of 20 MHz was measured on a Rohde & Schwarz FS315 spectrum analyzer in order to determine, for both evaluated standards, the RF power received by an isotropic antenna at 1 m from the router, in addition to their respective PSDs. Among the various channel options available for these standards, channel 9 (central frequency at 2.452 GHz) was chosen due to its proximity to the central frequency of the standards (2.45 GHz). Fig. 12a and Fig. 12b show the instantaneous PSD of each evaluated standard. As can be seen in Fig. 12a and Fig. 12b, each standard has a different PSD. As IEEE 802.11b (Fig.   12a) uses CCK spread spectrum modulation, its PSD after the wireless channel (gray shading) has larger spectral components at central frequencies that decrease as they move away from the central frequency (according to a sinc(| f |) 2 function), which for simplicity can be represented by a spectral mask similar to a bowler hat. On the other hand, as the IEEE 802.11g (Fig. 12b)     The study presented in Fig. 13, for a distance of 0.5 m, can be expanded to provide a more in-depth quantitative discussion taking into account other distances. The results obtained are presented in It is important to note that in practice the active period of the router can vary randomly, as it depends, among other things, on the random characteristics of the wireless channel and, consequently, on the bit error rate (which can cause an increase in the number of retransmissions). Therefore, the power ratio between the standards may also vary randomly depending on the characteristics of the wireless channel.

V. CONCLUSIONS
This work presents a complete RFEH analysis of the WiFi network signals, according to the IEEE 802.11b and IEEE 802.11g standards. A QYR prototype was designed and characterized and measurements were performed inside an anechoic chamber, varying the distance between a WiFi router and the QYR to fully assess its functioning from the point of view of RFEH when each of the evaluated standards is used.
The evidence from this research implies that the IEEE 802.11b standard is advantageous compared to the IEEE 802.11g standard when the focus is on RFEH. It was demonstrated that Vo and Pavg obtained with the use of IEEE 802.11b are, respectively, of the order of two and four times higher than those obtained with the use of IEEE 802.11g. Additionally, due to the lower data transmission rate of IEEE 802.11b and, consequently, the longer exposure time for the transmission of information, the energy transmitted during the communication period will also be higher when compared to the IEEE 802.11g.
In this way, considering these characteristics, it can be concluded that an interesting advantage can be obtained from the RFEH point of view if a control method is developed on the router to enable the IEEE 802.11b (or another standard employing similar characteristics) when the priority is RFEH or when there is no data information to be transmitted (e.g., for a faster recharge of the battery of low power devices) and to enable IEEE 802.11g only when the priority is to transmit data at higher rates.