Rich and context-aware activity logs facilitate user behavior analysis and health monitoring, making them a key research focus in ubiquitous computing. The remarkable semantic understanding and generation capabilities of Large Language Models (LLMs) have recently created new opportunities for activity log generation. However, existing methods continue to exhibit notable limitations in terms of accuracy, efficiency, and semantic richness. To address these challenges, we propose DailyLLM. To the best of our knowledge, this is the first log generation and summarization system that comprehensively integrates contextual activity information across four dimensions: location, motion, environment, and physiology, using only sensors commonly available on smartphones and smartwatches. To achieve this, DailyLLM introduces a lightweight LLM-based framework that integrates structured prompting with efficient feature extraction to enable high-level activity understanding. Extensive experiments demonstrate that DailyLLM outperforms state-of-the-art (SOTA) log generation methods and can be efficiently deployed on personal computers and Raspberry Pi. Utilizing only a 1.5B-parameter LLM model, DailyLLM achieves a 17% improvement in log generation BERTScore precision compared to the 70B-parameter SOTA baseline, while delivering nearly 10× faster inference speed.

Detailed activity logs are crucial for health monitoring and personalized interventions. Traditional methods rely on manual editing or raise privacy concerns due to the use of camera recordings. This paper proposes ContextLLM, an innovative system that utilizes a large language model (LLM) to understand sensor data from smartphones and smartwatches and automatically generate contextualized activity logs. Compared to the state-of-the-art, it incorporates key contextual information and physiological indicators, enabling more fine-grained semantic descriptions. Preliminary results show that the automatically generated activity logs achieve 80.26% similarity to human annotations, demonstrating the feasibility.

Federated learning is a distributed learning method by training the model in locally multiple clients, which has been used in numerous fields. Current convolutional neural networks (CNN)-based federated learning approaches face challenges from computational cost, communication efficiency, and robust communication. Recently, Hyper Dimensional Computing (HDC) has been recognized as a promising technique to address these challenges. HDC encodes data as high-dimensional vectors and enables lightweight training and communication through simple parallel vector operations. Several HDC-based federated learning methods have been proposed. Although existing methods reduce computational efficiency and communication cost, they are difficult to handle complex learning tasks and are not robust to unreliable wireless channels. In this work, we innovatively introduce a synergetic federated learning framework, FHDnn. With advantage of the complementary strengths of CNN and HDC, FHDnn can achieve optimal performance on complex image tasks while maintaining good computational and communication efficiency. Secondly, we demonstrate in detail the convergence of using HDC in a generalized federated learning framework, providing theoretical guarantees for HDC-based federated learning approach. Finally, we design three communication strategies to further improve the communication efficiency of FHDnn by 32×. Experiments demonstrate that FHDnn converges 3× faster than CNN-based federated learning methods, reduces the communication cost by 2,112×, and the local computation and energy consumption by 192×. In addition, it has good robustness to unreliable communication with bit errors, noise, and packet loss.

Secure protocol-independent communication is increasingly demanding to support information exchange among neighbor Internet of Things (IoT) devices. For example, recent works utilize ultrasound at the resonant frequency range of a gyroscope to build communication between a speaker and the gyroscope. However, they are vulnerable to eavesdropping attacks and may have limitations in communication delays. In this work, we present WordWhisper, an efficient, word-level, and speaker-to-gyroscope communication system, with which only the target device can receive the correct information. We theoretically analyze Micro-Electro-Mechanical System (MEMS) gyroscope resonance and propose a hardware-dependent mechanism to defend against eavesdropping, making non-target gyroscopes receive ineffective information. Note that WordWhisper is free of costly data collection from gyroscopes, we train and update our decoding model based on the synthesized data (generated from theoretical MEMS resonance analysis) rather than the costly collected data from gyroscopes. Meanwhile, we address the challenge of eavesdropping when it comes to multiple attackers. We evaluate WordWhisper over 50 MEMS gyroscopes and 100 words. Extensive evaluations demonstrate that WordWhisper can achieve word-level communication with 99.33% accuracy while the recognition accuracy drops to a random guess for the non-target. Our decoding delay is lower than 0.63 seconds.

In recent years, technologies such as Artificial Intelligence have been of great help in driving smart production in industry. However, some links with high complexity and flexibility still rely on manual labor. This paper focuses on the design process of electrical equipment. Based on the realistic needs of enterprises, we design a real-time system for automated recognition and analysis of blueprints of electrical equipment, AI-Blueprint. It automatically extracts multi-modal information from the user’s blueprints, comprehends it, and quickly matches it with the best-standardized schemes in the enterprise database. On average, each blueprint is processed in as little as ${1 . 2 3}$ seconds, which significantly reduces labor costs. We carefully designed several specific modules to build our system, including the Document Element Detector, Diagram Extractor, Table Extractor, and Feature Extractor. To achieve good performance and fast response of AI-Blueprint, we proposed some innovative algorithms and tested their impact on the system performance under different parameters. Numerous experiments have shown that AI-Blueprint can accurately understand information such as circuit diagrams and tables in electrical equipment blueprints, and recognize and classify them precisely and quickly.

Despite the potential for throughput enhancement with multiple tags, existing WiFi backscatter systems have been limited by inband interference among various tags. In response, we propose MultiRider, the first WiFi backscatter system that can tame in-band interference and support multi-tag parallel communication on commercial OFDM protocol. The principle behind MultiRider lies in its ability to demodulate and reconstruct tag data using just one uncorrupted subcarrier in the spectrum domain. To address the inherent challenges of preamble corruption and data collision due to in-band interference, we design three modules: 1) preamble recovery based on a concurrency-driven backscatter packet structure; 2) subcarrier-level demodulation using uncorrupted subcarriers; and 3) iterative interference cancellation for multiple tags. We prototype and evaluate MultiRider under 802.11g OFDM WiFi signals with commercial adapters and software-defined radios. Comprehensive evaluations illustrate that MultiRider can efficiently solve in-band interference. Notably, it can expand the network capacity of WiFi backscatter by 4× and use 8 channels in the 2.4GHz WiFi band for concurrent communication. Further results reveal that MultiRider can gain 10× network capacity in 35MHz bandwidth and reach 2.29 Mbps system throughput.

User authentication is essential for threat defense and data protection. Existed authentication systems have some known limitations, such as spoofing attacks, privacy leakage, and user-unfriendliness. In this paper, we propose BackLip, a novel anti-spoofing authentication system based on lip reading. We employ Wi-Fi backscatter-based technology to recognize users lip reading due to its various advantages, e.g. privacy protection, low power consumption, and low cost. Our system is touch-free and passphrase-independent, making it user-friendly, especially for the elderly and disabled. We first filter out irrelevant interference and enhance the signal-to-noise ratio of lip-reading signals by modulating the backscatter tags and constructing a series of suitable filters. Then, we study the energy distribution and steady-state characteristics of the backscattered signal caused by lip-reading movement at different frequencies to extract passphrase-independen lip-reading fingerprints. We build a theoretical model to analyze and prove the feasibility of our method and design an adaptive correction method to resist the interference caused by distance and angle changes. Additionally, we propose an adaptive segmentation algorithm to label lip-reading motions automatically. Extensive experiments demonstrate that BackLip has an average accuracy of 92.1% and is improved to 96.5% when users use the same passphrases.

The great popularity of smartwatches leads to a growing demand for smarter interactive systems. Hand gesture is suitable for interaction due to its unique features. However, the existing single-modal gesture interactive systems have different biases in diverse scenarios, which makes it intractable to be applied in real life. In this paper, we propose a multi-modal smartwatch interactive system named Mudra, which fuses vision and Inertial Measurement Unit (IMU) signals to recognize and identify hand gestures for convenient and robust interaction. We carefully design a parallel attention multi-task model for different modals, and fuse classification results at the decision level with an adaptive weight adjustment algorithm. We implement a prototype of Mudra and collect data from 25 volunteers to evaluate its effectiveness. Extensive experiments demonstrate that Mudra can achieve 95.4% and 92.3% F1-scores on recognition and identification tasks, respectively. Meanwhile, Mudra can maintain stability and robustness under different experimental settings.

Due to the complexity of the real world, effective consideration of the ambiguity and reliability of information is a challenge that must be addressed by the correct decision of the expert system. Z-number provides us with a good idea because it describes the probability of the random variable and the possibility measure. Recently, Yager presented a soft likelihood function that effectively combines probabilistic evidence to deal with the conflict information. This article generalizes Yager's soft likelihood function based on Z-numbers and proposes a Z-numbers soft likelihood function (ZSLF) decision model. The application examples show the rationality and effectiveness of the method. The comparison and discussion further show the advantages of the ZSLF decision model.

Information fusion is an important research direction. In this field, there are plenty of ways to combine evidence. Initially, Yager proposed a soft-likelihood function based on the ordered weighted average (OWA) operator to effectively fuse compatible probabilistic evidence. Recently, Song et al proposed a new soft-likelihood function based on the power ordered weighted average (POWA) operator. However, through analysis, we find Song et al's method has the following two shortcomings: (a) The weight of POWA cannot comprehensively reflect the relation between probability and OWA operator. (b) The soft-likelihood function does not reflect the preferences of decision makers. To overcome the above problem, we propose a modified soft-likelihood function. The effectiveness of the proposed method is demonstrated from the perspective of theoretical analysis and numerical examples.