Bitsliced cipher implementations demonstrate enhanced performance and security on high-end processors through specialized instruction utilization. However, IoT processors, particularly 32-bit architectures, present significant implementation challenges due to limited register sizes and instruction sets, hindering efficient parallelism in bitsliced SPN ciphers. This study presents optimization strategies for implementing bitsliced SPN ciphers on 32-bit processors using common instruction sets. The linear layer optimization employs a decomposition algorithm that transforms complex permutation operations into minimal instruction sequences. This approach recursively identifies optimal instruction combinations while maintaining computational efficiency. For the non-linear layer, operations are constrained to basic logic instructions (NOT, AND, OR, XOR). A novel encoding method for the Bit-slice Gate Complexity (BGC) model is proposed to optimize S-box transformations within these constraints using Boolean satisfiability solvers. Additionally, a comprehensive benchmarking framework facilitates standardized performance evaluation across implementations. Experimental evaluation of the optimized implementations on ARM Cortex-M and Xtensa LX processors demonstrates significant performance improvements. The proposed techniques achieve reductions of 9.7% and 67.6% in Cycles Per Byte for AES and QARMAv2 implementations, respectively.

Current camouflaged object detection (COD) methods rely heavily on large-scale datasets with pixel-level annotations. We propose a semi-supervised iterative learning network (SILNet) to address the reliance on large-scale pixel-level annotations in COD. SILNet employs a co-training strategy with convolutional networks and Transformers as encoders, followed by a binary gated decoder (BGD) for feature fusion. To optimize the use of labeled data, we introduce an optimal representative election mechanism (OREM) to identify key sequences of unlabeled images, guiding iterative learning and pseudo-label generation. To reduce noise in pseudo-labels, we incorporate a long-range representation module (LRM) leveraging Mamba’s background modeling. Experiments show that SILNet trained with only 10% of the labeled data outperforms state-of-theart unsupervised and weakly supervised methods, achieving performance competitive with fully supervised models.

The rapid advancement of autonomous driving technology, particularly in autonomous trucking on highways, shows great value for enhancing efficiency and reducing costs in the logistics industry. In this work, we define the full-trip speed planning problem for autonomous trucks under delivery time and fuel consumption constraints, referred to as the Operational Speed Planning (OSP) problem. To support and accelerate research on the OSP problem, we have developed a comprehensive dataset using a fleet of over 400 trucks. The dataset contains rich, diverse information covering more than 22 million kilometers of real-world highway driving data. In addition to this static dataset, we have developed a closed-loop simulator that allows for the interactive evaluation of OSP solutions, enabling researchers to test speed planning strategies in a realistic environment. Furthermore, we provide an OSP baseline method based on dynamic programming to optimize speed planning, balancing the delivery time requirements and fuel consumption. Our extensive experiments demonstrate both the accuracy of the simulation and the effectiveness of the OSP baseline in planning optimal speeds, proving its capability to meet time constraints while improving fuel efficiency. The dataset, simulator, and baseline will be made publicly available to foster further research and innovation in this area.

The rise of high-resolution cameras and deep learning models has propelled video analytics but also poses new challenges, especially in crowded outdoor scenes. Processing high-resolution video frames demands substantial computational resources, often surpassing edge device capabilities, resulting in high latency and energy costs. Detecting small objects in crowded regions further hinders accuracy in critical applications such as traffic surveillance and monitoring. To address these issues, we propose PAROD, a real-time system for high-resolution object detection in crowded outdoor environments. PAROD integrates three key technologies: a background modeling-based adaptive frame partitioning algorithm that dynamically identifies regions of interest (RoIs), parallel offloading of partitioned RoIs to multiple edge servers for inference, and an object counting model to detect crowded regions and enable differentiated inference strategies. By strategically allocating simpler models to ordinary regions and more complex models to crowded regions, PAROD enhances accuracy while minimizing latency. Furthermore, by resizing the input partitions to a fixed size before inference, PAROD reduces latency fluctuations and optimizes overall performance. Evaluations on a public dataset show that PAROD achieves a 3.4× speed-up over traditional edge offloading methods, with only a 1% accuracy reduction, offering a scalable solution for real-time, high-resolution video analytics in crowded outdoor scenes.

Fuel efficiency is a crucial aspect of long-distance cargo transportation by oil-powered trucks that economize on costs and decrease carbon emissions. Current predictive control methods depend on an accurate model of vehicle dynamics and engine, including weight, drag coefficient, and the Brake-specific Fuel Consumption (BSFC) map of the engine. We propose a pure data-driven method, Neural Predictive Control (NPC), which does not use any physical model for the vehicle. After training with over 20,000 km of historical data, the novel proposed NVFormer implicitly models the relationship between vehicle dynamics, road slope, fuel consumption, and control commands using the attention mechanism. Based on the online sampled primitives from the past of the current freight trip and anchor-based future data synthesis, the NVFormer can infer optimal control command for reasonable fuel consumption. The physical model-free NPC outperforms the base PCC method with 2.41% and 3.45% more significant fuel saving in simulation and open-road highway testing, respectively.