Exploiting pervasive leaked EM signals for communication, charging and sensing

Since the first successful wireless transmission across the Bristol Channel in 1897, wireless technologies have revolutionized the way we live, work and interact with the world. Besides the traditional communication function (e.g., Wi-Fi), wireless signals are further utilized for other functions such as localization, sensing and even charging. While promising in many aspects, one critical issue associated with these new functions is that they require dedicated signal transmissions, which interfere the original communication function. Take popular Wi-Fi sensing as the example. 200-1000 dedicated Wi-Fi packets per second need to be transmitted to enable Wi-Fi sensing which significantly decrease the throughput of ongoing Wi-Fi communication.

Thus, I ask this question: "can we utilize ambient, non-dedicated wireless signals to realize the aforementioned functions?" For my Ph.D thesis, I harness the pervasive ambient leakage signals traditionally considered detrimental to enhance the performance of wireless communication and enable new functions such as sensing and charging. My thesis is inspired by the key observation that there exist a large amount of leakage RF signals in our surroundings. For example, the powerlines continuously emit out 50/60 Hz electromagnetic (EM) signals due to the alternating current flowing inside. The operation of electric vehicles leaks RF signals in th frequency range of hundreds of hertz. One of the key wireless technologies in the next generation 6G networks, i.e., visible light communication (VLC) which relies on quickly turning ON/OFF the LED light also emit out RF leakages due to the quick ON/OFF state change. In my thesis, through novel designs across both hardware and software, I turn these ambient leakages from foes to friends.

Enhancing the security and throughput of wireless communication. Visible Light Communication is considered a key component of the 6G networks owing to its potential high data rates, pervasiveness of commodity LEDs, and minimal interference on existing RF communication. For the first time, we showed that during the transmission of VLC, the transmitter not only emits out visible light signals but also leaks out RF signals [1]. The underlying principle behind this leaked signal is the intensity modulation scheme commonly adopted in VLC systems. As illustrated in Figure 1, in VLC, data bits '0' and '1' are represented by switching 'OFF' and 'ON' the LED, which leads to quick alterations of the current flow. These changing electric currents induce EM signal leakages. We developed a theoretical model to quantify the relationship between the amplitude of these leaked RF signals and the ON/OFF frequency. We show that while LED light signals can be easily constrained within an area of interest (e.g., a room with walls), the leakage RF signal can penetrate through walls. What's more interesting is that the leakage RF signal contains a copy of the data transmitted in the light signals and this finding renders VLC-the generally believed most secure wireless technology-not secure any more as illustrated in Figure 2. Building upon this finding, we further demonstrate a novel utilization of these leaked signals for carrying extra data [2] to double the data rate of existing VLC systems. In this design, data is transmitted through the leaked RF signal without a dedicated active RF front-end, reducing both power and hardware costs. We show through experiment that the free leaked RF signals can be leveraged to transmit extra data at a data rate even higher than the primary VLC channel.

Harvesting the leaked energy for far-field charging. Following the successful utilization of VLC leakage signals for communication, we further view the leaked signals as a form of wasted energy and devote our effort to harvesting it [3]. The key observation enabling us to achieve this objective is that the surrounding objects can help significantly boost the amount of energy harvested. What is more exciting is that not just the ordinary objects such as walls and furniture can help enhance energy harvesting, the human body can also increase the amount of harvested energy. When the receiver antenna is in contact with a human body, the amount of harvested energy is increased by more than ten times. Based on this key observation, we propose our system, Bracelet+, which involves human body into the ecosystem of energy harvesting for the first time as depicted in Figure 3. We design the coil antenna as a bracelet so a person can wear it comfortably. The harvested power can reach up to micro-watts, holding promise for powering low-power body sensors and enabling body sensor network. Note that this far-field (i.e., a separation of a few meters between the target and the power source) wireless charging modality is very different from existing near-field wireless charging with the target and power source separated by just a few centimeters.

Enabling a novel sensing modality. After we successfully utilized leaked signals for communication and charging, we move forward to utilize the leaked signals (i.e., the powerline leakage) for sensing purposes [4]. This electromagnetic leakage, stemming from alternating current in powerlines, is governed by Maxwell's Equations. However, the ultra-low frequency (60 Hz) of the leakage renders existing wireless sensing models inapplicable. Specifically, current wireless sensing modalities rely on the fact that signal propagation is affected by human's motions and thus by analyzing the induced signal variation, the information of the human motion can be inferred. However, for leaked signal which is extremely low-frequency, the propagation of the signal is not affected by human motions. Owing to the antenna's unique property (i.e., coil shape) for low frequency signals, we design the antenna as a ring and combine the human target and ring antenna together as a "human antenna" as shown in 4(a) to sense the human target's gesture. Although the human target's gesture does not affect the signal propagation, it affects the antenna and accordingly affects the received signal. We can thus utilize the signal variation for sensing. One unique advantage of this sensing modality is that although the ring is only in contact with the target's finger, it can sense the motion of the whole body. This new sensing modality is demonstrated to be able to support a large range of sensing applications such as gesture recognition, sleep posture sensing, and fall detection.

Besides powerline leakage, we also leverage the EM leakage from electric vehicles (EVs) to enable in-vehicle sensing [5]. We observe that numerous components within the EVs including battery, powerline, and power inverter, emit EM signals during their operation. We thus use the leakage to sense the body motions of the driver/passenger without any dedicated signal transmitters. To address the unique challenge in car environment, i.e., the interfering car motions, we adopt a reference tag design, making the proposed sensing modality practical in real-world settings. Extensive experiments with a driving distance of 4000 kilometers under diverse real-world road conditions show that the proposed sensing system can achieve an accuracy over 90% in recognizing body motions solely utilizing ambient leakage signals.

To summarize, we innovatively utilize pervasive leakage signals for wireless communication, far-field charging, and wireless sensing, achieving the following contributions:

• We explored different types of leakage signals in our surroundings and demonstrated the application of the "bad" leakage signals in communication, charging and sensing.

• Through deeply understanding the signal characteristics, we propose models to lay the theoretical foundation for utilizing the leakage signals for communication, charging and sensing.