MobiCom 2024 Papers

Microwave ovens have become an essential cooking appliance owing to their convenience and efficiency. However, microwave ovens suffer from uneven distribution of energy, which causes prolonged delays, unpleasant cooking experiences, and even safety concerns. Despite significant research efforts, current solutions remain inadequate. In this paper, we first conduct measurement studies to understand the energy distribution for 10 microwave ovens and show their energy distribution in both 2D and 3D is very skewed, with notably lower energy levels at the center of the microwave cavity, where food is commonly placed. To tackle this challenge, we propose a novel methodology to enhance the performance of microwave ovens. Our approach begins with the development of a measurement driven model of a microwave oven. We construct a detailed 3D model in the High Frequency Structure Simulator (HFSS) and use real temperature measurements from a microwave to derive critical parameters relevant to the appliance's functionality (e.g., operating frequency, waveguide specifications). We then develop a novel approach that optimizes the design and placement of a low-cost passive metasurface for a given heating objective. Using extensive experiments, we demonstrate the efficacy of our approach across diverse food, optimization objectives, and microwave ovens.

This paper presents μTag, an ultra-low-power backscatter sensor that supports high-frequency sensing of a large number of targets simultaneously. The core of μTag is an RF "gene editing" technique that embeds both the identity of the sensor and the real-time motion state of the attached target intensively in the transient features of the sensor's RF signal, in a collision-resilient manner. We provide practical techniques which i) generate such "genetic signal" with purely analog and extremely simple circuits; and ii) separate the signals from a large scale of sensors reliably. Our experimental results show that our design can support concurrent tracking of 150 targets with a 12kHz per-tag sampling rate. We also demonstrate with multiple sensing applications that μTag can achieve high-speed and large-scale motion tracking and rotation frequency sensing. The PCB power consumption of μTag is 38~107μW, according to the operating frequency of the tag. Our ASIC simulation based on the 40nm CMOS process shows that the power consumption can be further reduced to 0.13~0.52μW.

A crucial task in autonomous driving is to continuously detect nearby vehicles. Problems thus arise when a vehicle is occluded and becomes "unseeable", which may lead to accidents. In this study, we develop mmOVD, a system that can detect fully occluded vehicles by involving millimeter-wave radars to capture the ground-reflected signals passing beneath the blocking vehicle's chassis. The foremost challenge here is coping with ghost points caused by frequent multi-path reflections, which highly resemble the true points. We devise a set of features that can efficiently distinguish the ghost points by exploiting the neighbor points' spatial and velocity distributions. We also design a cumulative clustering algorithm to effectively aggregate the unstable ground-reflected radar points over consecutive frames to derive the bounding boxes of the vehicles.

Leaf Wetness Duration (LWD), the time that water remains on leaf surfaces, is crucial in the development of plant diseases. Existing LWD detection lacks standardized measurement techniques, and variations across different plant characteristics limit its effectiveness. Prior research proposes diverse approaches, but they fail to measure real natural leaves directly and lack resilience in various environmental conditions. This reduces the precision and robustness, revealing a notable practical application and effectiveness gap in real-world agricultural settings. This paper presents Hydra, an innovative approach that integrates millimeter-wave (mm-Wave) radar with camera technology to detect leaf wetness by determining if there is water on the leaf. We can measure the time to determine the LWD based on this detection. Firstly, we design a Convolutional Neural Network (CNN) to selectively fuse multiple mm-Wave depth images with an RGB image to generate multiple feature images. Then, we develop a transformer-based encoder to capture the inherent connection among the multiple feature images to generate a feature map, which is further fed to a classifier for detection. Moreover, we augment the dataset during training to generalize our model. Implemented using a frequency-modulated continuous-wave (FMCW) radar within the 76 to 81 GHz band, Hydra's performance is meticulously evaluated on plants, demonstrating the potential to classify leaf wetness with up to 96% accuracy across varying scenarios. Deploying Hydra in the farm, including rainy, dawn, or poorly light nights, it still achieves an accuracy rate of around 90%.

5G new radio (NR) employs frequency range 2 (FR2) in the millimeter-wave (mmWave) bands, which employs a much shorter slot duration compared to FR1 (sub-7 GHz) systems and, therefore, poses significant challenges for softwarized baseband processing in virtualized radio access networks (vRANs). Existing systems supporting software baseband processing focus on enabling (massive) multiple-input and multiple-output (MIMO) using multi-core edge server(s). These solutions may fail to meet the more stringent processing deadline in FR2 or require more intensive computational resources. In this paper, we present Savannah, an efficient mmWave baseband processing framework using minimal and heterogeneous computing resources including CPU and eASIC. Savannah addresses the challenges associated with baseband processing in FR2 by applying techniques for vectorizing matrix operations and memory access patterns, supporting heterogeneous computation via offloading LDPC decoding to an eASIC, and enabling single-core operation. We show that Savannah, using a single CPU core and the ACC100 accelerator, can support a 2×2 MIMO link with 100 MHz bandwidth, yielding a data rate of up to 487 Mbps.

Cloud-based live video analytics with tight latency bound is gaining importance to support emerging applications such as UAVs and augmented reality. However, existing systems often struggle to meet stringent latency constraints under fluctuating network conditions with packet losses and late-arriving packets. We propose a loss-tolerant live video analytics system called Logan, which effectively accepts packet losses while maintaining high accuracy by utilizing the inherent resilience in DNNs. We design i) Codec-aware Inpainting, which accurately recovers the frame error from packet losses ii) Fast-Forward Recovery that prevents the remaining un-recovered error from propagating over future frames indefinitely. Our results show a 3× improvement (33.2%→99.9%) in SLO satisfaction rate compared to the reliable transmission scheme with <1% accuracy drop under a 5% packet loss rate.

Cellular networks provide the most accessible emergency services with ubiquitous coverage, yet their emergency-specific designs remain largely unexplored. To systematically explore potential design defects that lead to failures or delays in emergency services, we introduce M911-Verifier, an emergency-specific model checking tool. It reveals many counterintuitive findings regarding the ubiquitous access support for cellular emergency services. Our study shows that, despite sufficient wireless signal coverage, users may still experience prolonged emergency call setup times, call initiation failures, or call drops due to flaws in the design of cellular emergency services. These design defects arise from three major causes: problematic network selection for initiating emergency calls, emergency-unaware call operation, and network escalation forbidden during emergency calls. The impacts of these defects have been experimentally validated across three U.S. carriers and two Taiwan carriers using commodity smartphones. Finally, we propose solutions and evaluate their effectiveness.

Along with AIGC shines in CV and NLP, its potential in the wireless domain has also emerged in recent years. Yet, existing RF-oriented generative solutions are ill-suited for generating high-quality, time-series RF data due to limited representation capabilities. In this work, inspired by the stellar achievements of the diffusion model in CV and NLP, we adapt it to the RF domain and propose RF-Diffusion. To accommodate the unique characteristics of RF signals, we first introduce a novel Time-Frequency Diffusion theory to enhance the original diffusion model, enabling it to tap into the information within the time, frequency, and complex-valued domains of RF signals. On this basis, we propose a Hierarchical Diffusion Transformer to translate the theory into a practical generative DNN through elaborated design spanning network architecture, functional block, and complex-valued operator, making RF-Diffusion a versatile solution to generate diverse, high-quality, and time-series RF data. Performance comparison with three prevalent generative models demonstrates the RF-Diffusion's superior performance in synthesizing Wi-Fi and FMCW signals. We also showcase the versatility of RF-Diffusion in boosting Wi-Fi sensing systems and performing channel estimation in 5G networks.

The complexity and volatility of real-world radio environments often hamper wireless networks from achieving optimal performance. Recently, intelligent metasurfaces have been explored to dynamically reshape the radio propagation environment. However, existing systems are limited to standalone metasurfaces, only enabling one-time signal redirection/reshaping effects within their direct line-of-sight. They cannot effectively scale to cover larger areas. In this paper, we propose RFMagus, which employs a network of metasurfaces to overcome the limitation. We carefully optimize the configurations of the networked metasurfaces so that they can cooperatively and coherently propagate the analog signals towards the target regions. We have implemented the networked metasurfaces and deployed them in a variety of real-world environments. Experimental results demonstrate that RFMagus can effectively expand the coverage, improve the throughput, and operate transparently to different wireless standards.

In the current AI era, mobile devices such as smartphones are tasked with executing a myriad of deep neural networks (DNNs) locally. It presents a complex landscape, as these models are highly fragmented in terms of architecture, operators, and implementations. Such fragmentation poses significant challenges to the co-optimization of hardware, systems, and algorithms for efficient and scalable mobile AI.

In the wake of the rapid deployment of large-scale low-Earth orbit satellite constellations, exploiting the full computing potential of Commercial Off-The-Shelf (COTS) devices in these environments has become a pressing issue. However, understanding this problem is far from straightforward due to the inherent differences between the terrestrial infrastructure and the satellite platform in space. In this paper, we take an important step towards closing this knowledge gap by presenting the first measurement study on the thermal control, power management, and performance of COTS computing devices on satellites. Our measurements reveal that the satellite platform and COTS computing devices significantly interplay in terms of the temperature and energy, forming the main constraints on satellite computing. Further, we analyze the critical factors that shape the characteristics of onboard COTS computing devices. We provide guidelines for future research on optimizing the use of such devices for computing purposes. Finally, we have released the datasets to facilitate further study in satellite computing.

LoRa has been widely used for the Internet of Things (IoT) due to its low power consumption and long communication range. The standard LoRa demodulation process condenses the energy of LoRa chirps to combat noise. However, there is an intrinsic frequency jump in real-life LoRa signals that standard demodulation neglects, reducing communication range in practice. We thoroughly study the frequency jump phenomenon and observe that it affects LoRa demodulation mainly in two folds: First, it makes each section of the signal shorter than the standard FFT perception range, introducing additional noise; Second, it induces a random phase jump that causes destructive addition of signal power. To mitigate the influence of frequency jump on LoRa demodulation, we propose LoRaTrimmer, a novel, fast, and noise-resilient LoRa decoding algorithm that optimally condenses LoRa signal power. LoRaTrimmer contains two innovative designs: First, we trim the perception range of FFT at the frequency jump, trimming off the additional noise; Second, we bypass the phase jump induced by frequency jump by probabilistic modeling and add up signal power constructively. Furthermore, we performed theoretical analysis to guarantee the performance of our method. Thorough experiments in various real-life environments show 1.70 to 2.49 dB SNR gain over the state-of-the-art and 3.44 to 3.79 dB SNR gain over FFT-based methods, translating to at most 1.67 times gain of coverage area. LoRaTrimmer is also robust under complex noise patterns, and capable of real-time decoding, with the only overhead being a slight increase in computational cost (0.51 to 3.17 ms per packet, compared with 0.23 to 0.94 ms of baseline methods).

Wireless sensing is gaining increasing attention from both academia and industry. Various wireless signals, such as Wi-Fi, UWB, and acoustic signals, have been leveraged for sensing. While promising in many aspects, two critical limitations still exist: a) limited sensing coverage; and b) the requirement for dedicated sensing signals, which may interfere with the original function of the wireless technology. To address these issues, we propose to utilize GPS signals for sensing, as GPS signals are already pervasive and emitted from satellites 24/7 at pre-allocated frequency bands, causing no interference. To make GPS sensing possible, we reconstruct signals with amplitude and phase information which is critical for sensing using the raw measurements reported by commercial GPS receiver module. We also develop sensing models to tailor the unique properties of GPS signals such as extremely long transmission distance. Finally, we introduce the concept of distributed sensing and design signal processing methods to fuse signals from multiple satellites to improve sensing performance. With all these designs, we prototype the first GPS wireless sensing system on commercial GPS receiver modules. Comprehensive experiments demonstrate that the proposed system can realize meaningful sensing applications such as human activity sensing, passive trajectory tracking, and respiration monitoring.

Ultra-high precision motion sensing leveraging computer vision (CV) is a key technology in many high-precision AR/VR applications such as precise industrial manufacture and image-guided surgery, yet conventional CV can be challenged by moiré-based sensing mechanism, thanks to moiré pattern's high sensitivity to six degrees of freedom (6-DoF) pose changes. Unfortunately, existing moiré-based solutions, in their infancy, cannot deal with complicated curvilinear moiré patterns caused by various perspective angles. In this paper, we propose a generalized moiré-based mechanism, MoiréVision, towards practical adoptions; it relies on high-frequency gratings as visual marker to help extract the fine-grained feature points for ultra-high precision motion sensing. As the foundation of general moiré-based sensing, we propose a formulation to characterize "uncontrolled" curvilinear moiré patterns in practical scenarios. To deal with the problem of moiré feature interference in practice, we propose a Gabor-based algorithm to separate overlapped curvilinear moiré patterns from two dimensions. Furthermore, to extract fine-grained feature points for high-precision motion sensing, we propose a bending function-based model and a resolution-enhanced strategy to reconstruct detailed texture of moiré markers and extract moiré feature points at sub-pixel level. Extensive experimental results show that MoiréVision greatly enhances the usability and generalizability of moiré-based sensing systems in real-world applications.

Within the burgeoning 6G wireless network landscape, there is an intensified push toward achieving all-encompassing accessibility through integrated solutions spanning a multitude of domains. Notwithstanding recent advancements, the conventional relay-centric communication paradigms grapple with scalability and optimal performance issues. In this paper, we introduce MeAnt ---a versatile IoT platform uniquely architected to foster seamless cross-medium communication by leveraging the compact design of piezoelectric-based mechanical antennas (Piezo-MAs). By capitalizing on the propagation attributes of medium-frequency radios emitted from Piezo-MAs, MeAnt promises communication across diverse environments such as air, water, soil, concrete, and even biological tissue, all while maintaining a compact antenna footprint. Moreover, in light of challenges such as potential interference from AM broadcasts and the intrinsic unidirectional nature of Piezo-MAs, we have developed a finely crafted full-stack communication protocol. Comprehensive tests underscore the system's proficiency, demonstrating a penetration depth of up to 10 m in cross-medium environments and realizing a throughput of 8.7 kbps.

Accurate estimation of on-device model training time is increasingly required for emerging learning paradigms on mobile edge devices, such as heterogeneous federated learning (HFL). HFL usually customizes the model architecture according to the different capabilities of mobile edge devices to ensure efficient use of local data from all devices for training. However, due to oversimplification of latency modeling, existing methods rely on a single coefficient to represent computational heterogeneity, resulting in sub-optimal HFL efficiency. We find that existing methods ignore the important impact of runtime optimization of deep learning frameworks, which we call development-chain diversity. Specifically, layers of a model may have different algorithm implementations, and deep learning frameworks often have different strategies for selecting the algorithm they believe is the best based on a range of runtime factors, resulting in different training latencies and invalid predictions from existing methods. In this paper, in addition to considering this diversity to ensure synchronized completion time of model training, we also study how to select the best algorithm each time to reduce the latency of the per-round training, thereby further improving the overall efficiency of federated training. To this end, we propose LATTE, which consists of a novel selector that identifies the best algorithm at runtime based on relative runtime factors. By further integrating it into our training latency model, LATTE provides accurate training time estimation. We develop LATTE as middleware, compatible with different deep learning frameworks. Extensive results show significantly improved training convergence speed and model accuracy compared to state-of-the-art methods.

Deep learning-enabled Voice User Interfaces (VUIs) have surpassed human-level performance in acoustic perception tasks. However, the significant cost associated with training these models compels users to rely on third-party data or outsource training services. Such emerging trends have drawn substantial attention to training-phase attacks, particularly backdoor attacks. Such attacks implant hidden trigger patterns (e.g., tones, environmental sounds) into the model during training, thereby manipulating the model's predictions in the inference phase. However, existing backdoor attacks can be easily undermined in practice as the inserted triggers are audible. Users may notice such attacks when listening to the training data and remaining alert for suspicious sounds. In this work, we present a novel audio backdoor attack that exploits completely inaudible triggers in the frequency domain of the audio spectrograms. Specifically, we optimize the trigger to be a frequency-domain pattern with the energy below the noise floor (e.g., background and hardware noises) at any given frequency, thereby rendering the trigger inaudible. To realize such attacks, we design a strategy that automatically generates inaudible triggers in the spectrum supported by commodity playback devices (e.g., smartphones and laptops). We further develop optimization techniques to enhance the trigger's robustness against speech content and onset variations. Experiments on hotword and speaker recognition indicate that our attack can achieve attack success rates of more than 98.2% and 81.0% under digital and physical attack scenarios. The results also demonstrate the trigger's inaudibility with a Signal-to-Noise Ratio (SNR) less than -3.54 dB against background noises. We further verify that our attack can successfully bypass state-of-the-art backdoor defense strategies based on learning and audio processing.

Wireless signals have been widely utilized for human sensing. However, wireless sensing systems face a fundamental limitation, i.e., the wireless device must keep static during the sensing process. Also, when sensing fine-grained human motions such as respiration, the human target is required to stay stationary. This is because wireless sensing relies on signal variations for sensing. When device is moving or human body is moving, the signal variation caused by the target area (e.g., chest for respiration sensing) is mixed with the signal variation induced by device or other body parts, failing wireless sensing. In this paper, we propose MSense, a general solution to deal with motion interference from wireless device and/or human body, moving wireless sensing one step forward towards real-life adoption. We establish the sensing model by taking both device motion and interfering body motion into consideration. By extracting the effect of body and device motions through pure signal processing, the motion interference can be removed to achieve accurate target sensing. Comprehensive experiments demonstrate the effectiveness of the proposed scheme. The achieved solution is general and can be applied to different sensing tasks involving both periodic and aperiodic motions.

Sub-6GHz radio sensing offers several compelling advantages, such as resilience to poor lighting conditions, privacy preservation, and the ability to see through walls. However, in indoor environments, the sub-6GHz ISM spectrum is heavily occupied by WiFi devices, leaving little available spectrum for sensing purposes. In this paper, we introduce SiWiS, a new approach to integrate radio sensing capabilities into individual WiFi devices for fine-grained human activity detection. SiWiS comprises two main components: (i) a new hardware component that can be easily installed on an off-the-shelf WiFi device, and (ii) a dual-branch deep neural network (DNN) optimized for concurrent human mask segmentation and pose estimation. We have built a prototype of SiWiS and installed it on a commercial WiFi router for evaluation. Extensive experimental results demonstrate a significant performance improvement over WiFi channel state information (CSI) based sensing methods. More importantly, zero-shot experiments confirm that SiWiS can be directly transferred to unseen real-world environments.