Recent research has advanced the transmission rate and quality of video streaming in backscatter communications. However, these studies often overlook the challenge of deploying such systems in environments without dedicated excitation signals. To address this, we propose VideoBack, a high-quality video backscatter system using ambient WiFi signals for easy monitoring. VideoBack employs JPEG compression on high-resolution images to reduce transmission payload, enabling adaptation to the lower rates of WiFi backscatter. We design a customized packet structure for tags to backscatter image data using multiple uncontrolled ambient WiFi packets. To ensure power efficiency, we incorporate envelope detection with energy harvesting. To enable single-receiver deployment, we design a pilot-subcarrier-based recovery method to reconstruct the original WiFi signal. Our system maintains video reliability through a low bit error rate (BER) decoding strategy that includes phase error tracking and self-correction with multi-subcarriers. We prototype VideoBack using a commercial WiFi adapter, achieving a transmission rate of nearly 250 kbps with a BER below 0.05% up to 9 meters. VideoBack can send one frame of an image after just 2 seconds of energy harvesting within 3 meters.

Counterfeiting and product tampering pose significant challenges in high-value packaged goods, leading to brand value degradation, regulatory violations, and potential public health risks. Although radio-frequency identification (RFID)-based authentication technologies have been widely adopted for item-level verification, existing systems struggle to precisely locate anomalous items within densely packed packages, often necessitating manual inspection. This paper presents PackLoc, which achieves high-precision RFID-based item ordering entirely using commercial off-the-shelf (COTS) hardware. PackLoc recovers the ordering and spatial layout of tags in dense packaging through a fixed multi-antenna configuration. By combining RSSI clustering, phase recovery, and linear fitting, the system overcomes technical bottlenecks such as low-dimensional observations, hardware-induced phase offsets, and$\pi$-radians phase jump. Experimental results show that PackLoc achieves over 90% ordering accuracy in dense layouts without requiring mechanical motion or environment-specific calibration, enabling rapid, noninvasive detection of counterfeit or misplaced items.

Bluetooth-low-energy (BLE)-based low-power cameras have expanded the applications of battery-powered low-power Internet of Things (IoT) vision systems. The adaptive tuning of connection and image encoding settings plays a vital role in optimizing the efficiency of wireless vision systems. This article introduces a framework for adaptive adjustment in BLE-based low-power vision systems. To reduce power consumption, we model BLE power usage and design a dynamic BLE parameters and resolution adjustment method for low-power IoT vision system. Additionally, we design an edge-end combined approach that delegates the task of key region prediction to the receiving device to achieve adaptive intraframe JPEG encoding. Our approach leverages commercially available BLE devices without requiring physical layer modifications, ensuring compatibility with standard devices. We design a wireless hardware prototype using off-the-shelf CMOS sensors and BLE Systems on a Chip to evaluate our framework. Extensive experiments demonstrate a 66.9% increase in energy efficiency compared to traditional fixed parameter methods.

Low-power perception technologies, notable for their micro-watt-Ievel energy consumption, offer the capability for years of continuous data sensing and communication without the necessity for battery replacement. These characteristics make them extremely viable for large-scale deployment and position them as a pivotal trend in the future development of the Internet of Things. Low-power video transmission is a critical application. Low-power devices often have weaker signals, which can lead to consecutive packet loss, resulting in severe image defects at the receiver end. To address this, we propose a novel image transmission method at the transmitter side, which transmitting image blocks or rows using a specific sequence, significantly reducing the risk of consecutive block loss under severe channel interference or at low signal-to-noise ratio conditions, thereby en-hancing system stability under continuous packet loss conditions. Therefore, We adjust the deep learning model for real-world scenarios, achieving better image restoration in environments with high packet loss rates. We constructe a prototype based on backscatter communication cameras. Experiments conducte on this platform demonstrate that we can recover the majority of image information, even with a packet loss rate of 75%, by using traditional algorithms or deep learning approaches. This significantly enhances the visibility of low-power visual perception systems.

Accurate and high-speed object movement tracking systems often face significant challenges due to signal instability caused by the object’s movement and rotation. To address these issues, we present STABack, a novel object tracking system using backscatter tags and accelerometer sensors, designed for high-accuracy, high-speed movement tracking. We developed an amplitude stabilization algorithm which uses envelope detection to reduce the impact of high-frequency movement on the signal, and uses dynamic threshold output to lower the BER in backscatter demodulation. Our method reduces the BER by 0.3757 compared to the regular demodulation method, resulting in a final BER of 0.07. Our evaluation of the STABack prototype shows that it achieves a median distance measurement accuracy of 6.45 cm with a standard deviation of 6.95 cm, under the condition of a speed of 120 cm. The attitude angle estimation’s mean error is under 7 degrees. The system’s accuracy in detecting the target object’s trajectory is as high as 99%, and it can still decode with a bit error rate of no more than 0.034 at a speed of 166 cm/s. The power consumption of our system prototype is only 38.54 μW based on our experimental results. Overall, our results demonstrate that STABack can accurately estimate the movement and rotation of target objects in unstable backscatter channels.

Recent works have achieved considerable success in increasing the transmission rate of video streaming backscatter communications. However, they do not consider the ease of deployment in real-world environments where these dedicated excitations are not readily available. In addition, these works are transmitted with lower image resolution, and the images received by users are not eye-friendly. In this paper, we propose VideoBack, a high quality video backscatter system with commercial WiFi excitation for easy-to-use monitoring. The key idea is to perform JPEG compression on high-pixel images to adapt to intermittent ambient WiFi transmission. This approach allows ambient commercial routers to be used for excitation. Furthermore, our work enables high-quality video transmissions by devising a low bit error rate decoding scheme. We build a prototype of VideoBack and use a commercial WiFi adapter to excite the tag to realize the transmission and reception of video. The prototype uses RF signals to charge and store energy. Our evaluations show that VideoBack can transmit images at a throughput of nearly 250 kbps with bit error rate below 0.0005 within 9 meters, while the images only have minor distinctions. VideoBack can support sending one frame of image after just 2 seconds of power acquisition within 3m.

In this paper, we present a novel object tracking system, named STABack, that utilizes backscatter tags and accelerometer sensors. The system is designed to support high-speed movement tracking with high accuracy. One of the challenges faced in this system is the instability of the received signal due to the motion and rotation of the backscatter tag. To address this issue, we propose an amplitude stabilization algorithm to eliminate the interference caused by the tag's motion. The algorithm uses envelope detection to remove the impact of high-frequency motion on the signal, and a dynamic threshold output to further reduce the bit error rate of backscatter demodulation. Additionally, we perform outliers removal and interpolation on the three-axis accelerometer data and compute the attitude angle using 3D geometric quadrants. Finally, we implement the prototype of STABack and evaluate its performance. Our method improves BER up to 0.3757 compared with the regular demodulation method. Our evaluation of the STABack prototype shows that it achieves a median distance measurement accuracy of 6.45 cm with a standard deviation of 6.95 cm. The mean error of the attitude angle estimation is less than 7 degrees, and the average relative error of three-axis acceleration tracking is only 0.097. The system's accuracy in detecting the target object's trajectory is as high as 98%, and it can still decode with a bit error rate of no more than 0.034 at a speed of 166 cm/s. The power consumption of backscatter communication is$38.54\ \mu\mathrm{W}$. Based on our experimental results, Overall, our results demonstrate that STABack can accurately estimate the movement and rotation of target objects in unstable backscatter channels.