Federated learning (FL) through its novel applications and services has enhanced its presence as a promising tool in the Internet of Things (IoT) domain. Specifically, in a multiaccess edge computing setup with a host of IoT devices, FL is most suitable since it leverages distributed client data to train high-performance deep learning (DL) models while keeping the data private. However, the underlying deep neural networks (DNNs) are huge, preventing its direct deployment onto resource-constrained computing and memory-limited IoT devices. Besides, frequent exchange of model updates between the central server and clients in FL could result in a communication bottleneck. To address these challenges, in this article, we introduce GWEP, a model compression-based FL method. It utilizes joint quantization and model pruning to reap the benefits of DNNs while meeting the capabilities of resource-constrained devices. Consequently, by reducing the computational, memory, and network footprint of FL, the low-end IoT devices may be able to participate in the FL process. In addition, we provide theoretical guarantees of FL convergence. Through empirical evaluations, we demonstrate that our approach significantly outperforms the baseline algorithms by being up to 10.23 times faster with 11 times lesser communication rounds, while achieving high-model compression, energy efficiency, and learning performance.
Traditional high-performance computing and modern artificial intelligence computing are converging with workflows as a common paradigm. We predict nine principles of heterogeneity and serverless computing for this convergence, from high-level programming to low-level hardware.
1 April 2025 in IEEE Transactions on Mobile Computing
Recent advances in speech and language processing have led to the rise of smart voice services like Alexa, Google Home, and Siri. However, these advancements also increase security risks due to sophisticated voice domain attacks. Instead of relying on acoustic clues to detect replayed or synthesized speech, we utilize microphones and motion sensors in head-wearable devices to authorize legitimate users through bone-conducted vibrations, enabling multi-factor authentication (MFA) for spoken voice. Our proposed two-stage authentication system, AirBone, captures air and bone conduction (AirBone) signals and exploits two authentication factors sequentially. The first stage, called temporal consistency scoring (TCS), employs signal processing to verify the recorded AC and BC signals are concurrent and originate from the same vocalization process. Statistical tools are employed to distinguish legitimate attempts against false-triggering or acoustic attacks. The second stage leverages deep learning to verify the user’s unique bone conduction patterns in the vibration domain. Specifically, we enhance the robustness through data augmentation with constant-Q transform and adversarial training, improving the model’s ability to detect impersonation and machine-induced vibrations. Thanks to these designs, AirBone authentication offers enhanced security via MFA with no extra cost of user effort. In addition, our experimental results demonstrate a $96.3\%$ overall accuracy, robustness against AirBone noise and room impulse responses, and $0.3\%$ Equal Error Rate (EER) against acoustic and cross-domain attacks.
25 October 2024 in IEEE Internet of Things Journal
Federated learning (FL) over mobile devices is an emerging distributed learning paradigm for numerous delay sensitive applications. In FL, the training delay is composed of the computing and communication delay. Some of the participating mobile devices may have slow local computing or wireless communications, which results in high FL training delay. Intuitively, if fast devices help slow ones, the FL training delay can potentially be reduced. However, helping each other among devices requires frequent transmissions and may cause additional delay. Fortunately, we observe that device-to-device (D2D) transmission, a fast and direct transmission, may be applied between device pairs to mitigate the additional delay from frequent transmissions. Inspired by those observations, we develop the D2D transmission assisted FL (DAFL), a novel FL scheme to improve the training delay over mobile devices. Briefly, we first put the eligible mobile devices into pairs, assigning each pair to one of the four types of relation: 1) similar computing, large communication gap; 2) similar communication, large computing gap; 3) one with faster computing and the other with faster communication; and 4) one with both faster computing and communication. We design the process for each type of device pair to: 1) improve the transmission delay of each pair, by letting the fast device help with the model parameters transmission to the server and 2) improve the computing delay by splitting learning task between paired devices. The emulation results demonstrate that DAFL surpasses existing peer designs in terms of reducing training delay by more than 20%.
Provisioning language models as inference services can be significantly challenging to cloud providers because of the need to balance service-level objectives (SLOs) with capital and operational costs. The heterogeneous nature of user queries, combined with temporal variations in workload patterns and the high energy consumption of GPUs warrants an intelligent resource-allocation strategy. We present OASIS (Optimal Allocation Strategy for Inference Services), a comprehensive methodology that combines static and dynamic optimizations to provision inference services efficiently. OASIS employs a two-phase approach: (1) pre-deployment profiling to determine optimal parameter configurations for different query types across various hardware options, and (2) run-time query routing and dynamic provisioning based on observed workload patterns. We evaluate OASIS using real-world BurstGPT traces on NVIDIA A100 and H100 GPUs. Our results show that: (1) GPU frequency scaling from 1320 MHz to 855 MHz reduces power by 39% with only 7% throughput loss; (2) query type classification reveals long-input-short-output workloads achieve 2.1× higher throughput than short-input-long-output; (3) MIG-based multi-tenancy achieves 42-43% power reduction when running two models simultaneously; and (4) runtime adaptation on real campus traces achieves statistically significant 12.4% power reduction while maintaining SLO compliance and identical throughput.