Demo: Energy Neutral Backscatter Device Utilizing High Performance Commercial Components

Demo: Energy Neutral Backscatter Device Utilizing High Performance Commercial Components Kalle Koskinen, Jingyi Liao, Kalle Ruttik and Riku J¨ antti Department of Communications and Networking, Aalto University, 02150 Espoo, Finland Email: {kalle.koskinen, jingyi.liao, kalle.ruttik, riku.jantti}@aalto.fi Abstract—This demonstration presents an energy-autonomous ambient backscatter device that achieves full operational selfsufficiency by harvesting energy from ambient light. Unlike the 3GPP-defined Ambient Internet of Things (A-IoT) concept, which relies on RF-based energy harvesting and continuouswave illumination, our setup exploits existing LTE downlink signals as the communication carrier and leverages the user equipment (UE) channel estimator as the backscatter receiver. The prototype integrates a high-performance microcontroller unit (MCU) capable of executing complex signal processing tasks while maintaining energy neutrality through photovoltaic energy harvesting. Real-time power measurements demonstrate less than 30 µWconsumption for communication, and the energy harvested suffices for continuous operation. This proof-of-concept validates the feasibility of light-powered, infrastructure-compatible ambient backscatter communication, representing a complementary and sustainable direction for 6G A-IoT systems. Index Terms—Ambient Backscatter Communications, LTE, Cell-specific Reference Signals, Channel Estimation, Ambient Internet of Things I. INTRODUCTION The Ambient Internet of Things (A-IoT) concept, introduced in 3GPP Release 19, envisions the large-scale deployment of battery-free sensor and actuator devices as part of massive Machine-Type Communications (mMTC) [1]. Current A-IoT operation targets indoor environments, where nearby base stations provide continuous carrier-wave (CW) illumination for RF-based energy harvesting. While this enables battery-free operation, it confines deployment to dedicated indoor setups and yields low overall energy efficiency. Recent work on cellular-integrated backscatter systems has shown how A-IoT can extend to commercial networks without CW illumination, reusing existing downlink transmissions for communication [2]. However, since such transmitters are typically distant, the received power is insufficient for RF energy harvesting. To ensure energy autonomy, we employ photovoltaic (PV) energy harvesting, offering a sustainable and widely available power source. Earlier self-sustaining backscatter demonstrations have relied mainly on low-performance components to minimize power consumption [3], [4], with energy-use estimates often limited to averages [5], analytical models [4], or incomplete methodologies [3]. In this demonstration, we present an energy-neutral A-IoT device that combines a highperformance commercial microcontroller, PV harvester, and supercapacitor-based storage, together with a real-time monitoring framework that visualizes instantaneous power conSolar Panel Battery-free Base Station (BS) User Equipment (UE) as AmBC receiver Backscatter Device (BD) L0 L0 L1 Direct path BD Modulate LTE pilot signal Scattered path Channel Estimate LTE as ambient signal Fig. 1. Illustration of AmBC system. sumption across operational states. The setup integrates the device into an LTE downlink, showing that harvested light energy suffices for both sensing and backscatter communication. Building on our earlier LTE-based AmBC demonstration using a UE channel estimator as receiver [6], the present work introduces a fully self-powered, light-harvesting A-IoT prototype that remains compatible with existing cellular infrastructure. To comply with frequency-licensing regulations, the demonstration is performed over coaxial cables, using recorded LTE base-station signals as illumination and an SDRbased UE channel estimator as the receiver. II. SYSTEM OVERVIEW A. Ambient backscatter communication Figure 1 illustrates the proposed AmBC signal model. The base Station (BS) continuously broadcasts an LTE cell specific reference signal (CRS) and the UE receives that CRS. This downlink is called the direct path. The ambient backscatter device (BD) is illuminated by that ubiquitous CRS. In the scattered path, the BD modulates the propagation channel using frequency-shift keying (FSK) modulation by changing antenna impedance matching [7]. This modulated channel state is recognized by the UE’s channel estimation. B. Energy neutral device implementation Figure 2 presents the main components of the energy neutral device. The microcontroller (MCU), STMicroelctronics STM32L562 [8] is based on Arm Cortex-M33. It incorporates RAM and ROM memories and several peripheral components, Fig. 2. Energy neutral device. 1: Microcontroller Unit (MCU), 2: Harvester, 3: PV Cell, 4: Energy Storage, 5: RF Switch. 1cm1pt Boot Set alarm Enter stop mode Alarm wakeup Measure SC voltage Over 3V? Transmit data no yes Fig. 3. Bacscatter device program flow. including the timers that generate the FSK waveform for the backscatter modulator. The main benefit of this MCU is the possibility to keep these timers operational while the processor core is kept in a low power stop state. Figure 3 shows the operation flow of the backscatter device. After the device boots up, a real time clock (RTC) alarm is set 20 s in the future to trigger a backscatter transmission. While waiting for the alarm, the MCU is placed in a low power consuming sleep state, where only the RTC consumes energy. Before transmission, the voltage of the supercapacitors is measured to verify that the device has enough energy for the operation. For transmission, one of the MCU timers is set up to output a square wave to drive the RF switch. Another timer is set to wake the MCU core for each new symbol. The core adjusts the first timer according to the symbol to be sent, and then returns to the low power stop mode. During backscatter transmission, only the processor timers are operating, and power consumption is minimized. C. Consumption measurement To measure the energy consumption of the device in real time, a measurement resistor Ris added to the MCU supply path (Figure 4a). The voltage across the resistor Ris amplified + − G= 100 um(t) R i(t) (a) 0 1 2 FSK drive (V) 0.2 0.22 0.24 0.26 0.28 0.3 0 50 100 150 Time (s) Current (µA) (b) Fig. 4. Current measurement setup (a) and current consumption during symbol changes (b). Fig. 5. Demonstration scenario. with an instrumentation amplifier with gain G, and the output voltage um(t)of the amplifier is measured with an oscilloscope. The current can be calculated from the voltage as i(t) = um(t) GR .(1) III. DEMONSTRATION OVERVIEW The setup of the experiment is depicted in figure 5. Two B205 Universal Software Radio Peripherals (USRPs) are set up to emulate an LTE BS and UE. The BS transmits LTE pilot signals, and the UE estimates the channel based on them. In order to avoid interfering with other equipment, the demonstration channel is cable based. The backscatter device is connected to the channel via a directional coupler and uses the RF switch to modulate the channel using binary frequency shift keying. The keying frequencies are 250 Hz and 625 Hz. The AmBC reader demodulation algorithm runs on the Laptop 1 displaying the reception results in real time. During the demonstration, the backscatter device is energy autonomous, harvesting energy using the PV cell. An oscilloscope measures the backscatter device MCU supply voltage and current. Laptop 2 is used to display the real-time power consumption. The BD requires less than 30 µWduring backscatter transmission. IV. CONCLUSION In this work, we demonstrated the feasibility of an energyneutral communication device based on a high-performance commodity microcontroller. We also presented a simple but still accurate real-time power consumption measurement method for low-power IoT devices. ACKNOWLEDGMENT This work is in part supported by the European Union Horizon Europe project AMBIENT-6G (grant 101192113) and Business Finland Project eMTC (Dnro 8028/31/2022). REFERENCES [1] 3rd Generation Partnership Project (3GPP), “Study on ambient IoT (Internet of Things) in RAN,” 3rd Generation Partnership Project (3GPP), Technical Specification TS 38.848, Jun. 2023, release 18. [2] R. J¨ antti, J. Liao, T. Zhang, B. Xie, M. U. Sheikh, K. Ruttik, G. Phan, D.- T. Phan-Huy, and A. Hassan, “Integration of backscatter-based ambient internet of things to cellular communication systems,” IEEE Communications Standards Magazine, 2025. [3] M. Lazaro, A. Lazaro, and R. Villarino, “Long-Range LoRaWan backscatter based sensors for medical and wearable applications,” in 2021 51st European Microwave Conference (EuMC), Apr. 2022, pp. 777–780. [4] S. N. Daskalakis, A. Georgiadis, M. M. Tentzeris, G. Goussetis, and G. Deligeorgis, “The New Era of Long-Range “Zero-Interception” Ambient Backscattering Systems: 130 m with 130 nA Front-End Consumption,” Sensors, vol. 22, no. 11, p. 4151, Jan. 2022. [5] A. Vena, N. Samat, B. Sorli, and J. Podlecki, “Ultralow Power and Compact Backscatter Wireless Device and Its SDR-Based Reading System for Environment Monitoring in UHF Band,” IEEE Sensors Letters, vol. 5, no. 5, pp. 1–4, May 2021. [6] J. Liao, K. Ruttik, R. J¨ antti, and D.-T. Phan-Huy, “Ue assisted ambient iot in lte downlink, in real-time and open source,” in Proceedings of the 21st Annual International Conference on Mobile Systems, Applications and Services, 2023, pp. 588–589. [7] J. Liao, X. Wang, K. Ruttik, R. J¨ antti, and D.-T. Phan-Huy, “In-band ambient FSK backscatter communications leveraging LTE cell-specific reference signals,” IEEE Journal of Radio Frequency Identification, vol. 7, pp. 267–277, 2023. [8] STMicroelectronics NV, “Ultra-low-power Arm Cortex-M33 32-bit MCU + TrustZone + FPU, 165DMIPS, up to 512KB Flash, 256KB SRAM, SMPS, AES + PKA,” STM32L562xx datasheet, Sep 2020, rev 4.