Jiamei Lv

IoT applications often keep secrets and safety-critical decisions in an operating system exposed to physical or remote attackers. Wardrift is a prototype hardening pipeline that combines constrained large language model (LLM) assistance with static analysis to move selected C# application state and logic into ARM TrustZone. The LLM identifies security anchors from a natural-language protection goal; Roslyn analysis constructs confidentiality and integrity boundaries; and a generator produces host patches, bridge calls, and Rust OP-TEE artifacts. Across 14 C# projects, Wardrift achieved 0.923 anchor F1, 0.889 sensitive-path F1 with GPT-5.4 in the default hybrid boundary analyzer, and 8 of 14 end-to-end hardening successes.

The growth in mobile applications and user numbers has significantly increased the volume of app updates on the network. For example, TikTok has garnered 825 million downloads in 2024 and maintains a nearly weekly update frequency. To reduce the data traffic, incremental update has become a common solution, as it calculates and transmits only the differences between two versions of an app. However, the widespread adoption of application encryption has introduced new challenges for efficient incremental updates. Even minor changes in files or encryption keys can result in substantial differences between app versions that were initially highly similar. This paper presents EncDiffPatch, the first incremental update system tailored for encrypted mobile apps. EncDiffPatch leverages two key techniques to enable efficient updates. First, it uses an updating-aware key sharing mechanism, which utilizes file-level encryption and dynamically shares the encryption keys for files that remain unchanged across versions based on the content of these files. Second, it incorporates the decryption process with a selective decompression approach. This approach performs only the Huffman decoding process of decompression and decompresses only those bytes estimated to contribute to differencing benefits. Evaluation shows that EncDiffPatch achieves a 51% reduction in compression ratio without introducing any performance overhead. When incorporating the decryption process, EncDiffPatch achieves a 73% reduction in compression ratio, outperforming the state-of the-art algorithms while requiring only 45.3% of the reconstruction time. Our security analysis shows that EncDiffPatch maintains an equivalent level of security to existing mobile systems.

Bluetooth Low Energy (BLE) technology supports various Internet-of-Things (IoT) applications. However, because of their limited transmission power and channel interference, their performance is deficient over weak links. Extending physical layer symbols or using error correction code to the link layer is effective somehow. Introducing excessive BLE bits to both respectively can also decrease the network throughput. To optimize the BLE technology performance, we propose CPL, a combining PHY and link layer optimization technology that adaptively allocates BLE bits to both the physical layer and link layer. Then we propose the Cross-Layer BLE Bits Dynamic Allocation Model that unifies the gain of BLE bits in different layers. Finally, we propose a Interference-Aware Controlled CFO Fine-Tuning Method that calibrates the model according to different interference patterns. We implement CPL on Commercial-Off-The-Shelf (COTS) BLE chips and SDR. The experiment results show that under various interference conditions, CPL achieves 50× and 32.16% throughput improvement than RSBLE and Symphony. CPL reduces energy consumption by 60.42% to 97.95% compared to RSBLE, and 11.04% to 25.15% compared to Symphony.

The fifth-generation (5G) has empowerd various applications. Effective bandwidth and delay prediction in 5G cellular networks are essential for many applications, such as virtual reality and holographic video streaming. However, accurate bandwidth and delay prediction in 5G networks remains a challenging task due to the short-distance coverage and frequent handover properties of 5G base stations. In this paper, we propose HYPER, a hybrid bandwidth and delay prediction approach that uses an Auto Regressive Moving Average (ARMA) time series predictive model for intra-cell prediction and a Random Forest (RF) regression model for cross-cell prediction. Our ARMA model takes prior information as its input, while the RF model further uses related network and physical features to predict future performance. We conduct a measurement study in commercial 5G networks to analyze the relationship between these features and bandwidth/delay. Moreover, we also propose a handover window adaptation algorithm to automatically adjust the handover window size and determine which model to use during handover for accurate bandwidth and delay prediction. We use commercial 5G smartphones for data collection and conducted extensive experiments in diverse urban environments. Experimental results show that HYPER can reduce the prediction error by more than 13% compared to state-of-the-art prediction approaches.

The growing popularity of encrypted container images in registries poses unique challenges for storage management due to the necessity for deduplication amidst rising image volumes. Traditional deduplication struggles with encrypted content, which inherently disguises duplicate data as distinct due to its randomized nature. Current advanced methods tackle this issue by decompressing images and applying message-locked encryption (MLE). However, these techniques face considerable challenges. Minor content changes can impair deduplication effectiveness, and decompressing layers increases storage requirements. Furthermore, this process negatively impacts both the speed at which users access the images and the overall system throughput. We propose SimEnc, a high-performance and secure deduplication system for encrypted container images by exploiting multiple similarity spaces. SimEnc pioneers the integration of semantic hashing with MLE to effectively parse semantic relationships across layers, thereby increasing deduplication efficacy. This system incorporates a rapid selection mechanism for similarity spaces, offering enhanced flexibility over previous models that relied on full decompression. By adopting Huffman decoding to navigate new similarity spaces, SimEnc not only improves deduplication ratios but also enhances overall performance. Our experimental results demonstrate that SimEnc substantially reduces storage needs by up to 261.7% compared to encrypted serverless platforms and by 54.2% against plaintext registries, while also delivering superior pull latency metrics.

Recent years have witnessed the rapid growth of IoT (Internet of Things). Bluetooth Low Energy (BLE) is one of the most popular wireless protocols to implement IoT applications because of its energy efficiency and low-cost properties. However, the development of BLE applications is time-consuming and exhausting. Users are required to write programs for both sides of a BLE connection using complicated low-level APIs. Moreover, it needs much expertise for developers to set appropriate parameters in accordance to different application requirements, especially when there exist multiple concurrent BLE connections. To address these problems, we propose BLEdge , an edge-centric programming approach for BLE applications with multi-connection optimization. First, we propose a wireless bus abstraction for BLE programming. With this, users can write BLE applications in an edge-centric way, as if the BLE-connected peripherals are physically attached to the edge node. Second, we advocate an optimization approach for BLE connection parameters. This optimization approach considers the time slot collision problem under a multi-connection scenario. We conduct extensive experiments with the nRF52840DK platform. Experiment results show that BLEdge can reduce 62.50% to 90.55% LOC (Lines of Code) when developing BLE applications. Furthermore, our parameter optimization approach can reduce up to 42.23% energy consumption.