This paper proposes an inductor-first two-inductor two-flying-capacitor (2L2C) multi-path hybrid DC-DC Converter. The proposed converter operates with a switching frequency of 2MHz, an input voltage range of 3 V to 5 V, and an output range between 0.8 V and 1.2 V using only 2-V NMOS power transistors. This hybrid DC-DC topology enables continuous input current and thus alleviates electromagnetic interference (EMI) issues. The two inductors operate in an interleaved manner, reducing the output current ripple. The switched-capacitor network decreases the average inductor current. This work is simulated in a 180-nm BCD process. Both inductors are 470nH with 29-mΩ DCR, while both flying capacitors are 10μF with 10-mΩ ESR. We obtain a peak efficiency of 96.89% at 400 mA load for a 4V-to-1V conversion.

This paper presents a 12V-input 1.8V-0.8V-output multiple-output hybrid buck (MOHB) DC-DC converter with a shared flying-capacitor (CF0) and real-time CF0voltage calibration, while the main control loop uses a pulse-width modulation (PWM) control scheme. With the proposed time-interleaving operation scheme, the MOHB converter has no cross regulation during load transient. The real-time CF0voltage calibration ensures that the voltage across CF0 equals VIN/2. The MOHB converter is designed with four outputs and simulated in an 180nm BCD process. Each output employs a 1μH power inductor with 19mΩ DCR and one 4.7μF flying capacitor. With 12V input, one output equals to 1.8V and other outputs equal to 1V, the peak efficiency 95.2%.

Switched-capacitor-inductor dual-path hybrid DC-DC converter can achieve high power density and efficiency, with reduced inductor voltage and current stresses and thus reduced inductor conduction loss and volume. However, it suffers from a large output ripple due to the hard-charging of the switched-capacitor (SC). To take advantage of recent in-substrate or on-chip high-density capacitor technologies, this paper proposes a multiphase interleaving operation for the dual-path SC hybrid converter, increasing the equivalent switching frequency and reducing the amplitude of the current ripple. It achieves a significant output ripple reduction within negligible efficiency degradation. The analysis and simulation results prove that the proposed scheme is a better method for reducing output ripple than increasing operating frequency or reducing power transistor size in the hard-charging path.

Visual object tracking aims to locate a targeted object in a video sequence based on an initial bounding box. Recently, Vision-Language~(VL) trackers have proposed to utilize additional natural language descriptions to enhance versatility in various applications. However, VL trackers are still inferior to State-of-The-Art (SoTA) visual trackers in terms of tracking performance. We found that this inferiority primarily results from their heavy reliance on manual textual annotations, which include the frequent provision of ambiguous language descriptions. In this paper, we propose ChatTracker to leverage the wealth of world knowledge in the Multimodal Large Language Model (MLLM) to generate high-quality language descriptions and enhance tracking performance. To this end, we propose a novel reflection-based prompt optimization module to iteratively refine the ambiguous and inaccurate descriptions of the target with tracking feedback. To further utilize semantic information produced by MLLM, a simple yet effective VL tracking framework is proposed and can be easily integrated as a plug-and-play module to boost the performance of both VL and visual trackers. Experimental results show that our proposed ChatTracker achieves a performance comparable to existing methods.

Adapters are lightweight bottleneck modules introduced to assist pre-trained self-supervised learning (SSL) models to be customized to new tasks. However, searching the appropriate layers to insert adapters on large models has become difficult due to the large number of possible layers and thus a vast search space (2Npossibilities for N layers). In this paper, we propose a technique that achieves automatic insertion of adapters for downstream automatic speech recognition (ASR) and spoken language understanding (SLU) tasks. Our approach is based on two-stage training. First, we train our model for a specific downstream task with additional shallow learnable layers and weight parameters to obtain the weighted summation over the output of each layer in SSL. This training method is established by the SUPERB baseline [1]. This first-stage training determines the most important layers given their respective weights. In the second stage, we proceed to insert adapters to the most important layers, retaining both performance and neural architecture search efficiency. On the CommonVoice dataset[2] we obtain 20.6% absolute improvement in Word Error Rate (WER) on the Welsh language against the conventional method, which inserts the adapter modules into the highest layers without search. In the SLURP SLU task, our method yields 4.0% intent accuracy improvement against the same conventional baseline.

The Hybrid DC-DC converter, especially with multiple inductors targeting high current delivery, has the advantages of high efficiency and power density with a large voltage conversion ratio (VCR), due to the combination of the benefits of both switched-capacitor-based and inductor-based buck converters. However, a higher number of inductors means that the energizing time for each inductor has more limitations, resulting in a relatively narrower VCR range. To reduce the conduction loss and extend the voltage conversion ratio scope, this paper presents a symmetrical double step-down (SDSD) converter with a VCR range up to 1/3, regulating an output voltage interval of 0.5 V-0.8 V from a 2.7 V-4.2 V Lithium-ion battery. This converter, implemented in 65 nm CMOS, occupies a core active area of 1.53 mm2. This work obtains 86.5% peak efficiency and 326 mA/mm2 maximum current density, with an effective switching frequency of 3 MHz.

In recent years, fatigue driving has been a serious threat to the traffic safety, which makes the research of fatigue detection a hotspot field. Research on fatigue recognition has a great significance to improve the traffic safety. However, the existing fatigue detection methods still have room for improvement in detection accuracy and efficiency. In order to detect whether the driver has fatigue driving, this paper proposes a fatigue state recognition algorithm. The method first uses MTCNN (multitask convolutional neural network) to detect human face, and then DLIB (an open-source software library) is used to locate facial key points to extract the fatigue feature vector of each frame. The fatigue feature vectors of multiple frames are spliced into a temporal feature sequence and sent to the LSTM (long short-term memory) network to obtain a final fatigue feature value. Experiments show that compared with other methods, the fatigue state recognition algorithm proposed in this paper has achieved better results in accuracy. The average accuracy of the proposed method in detecting key points of the face is as high as 93%, and the running time is less than half of the ordinary DLIB method.

Quick response (QR) code, one of the two‐dimensional barcodes, is now being widely used in all fields. The effectiveness of decoding, however, needs to be improved in real‐time application. In most cases, the decoding procedure is time consuming, in which the detection of QR code plays an essential part. Therefore, this study proposes a fast detection method of QR code based on run‐length coding: firstly, a novel approach is proposed to detect the minimum region containing position detection pattern (PDP) in QR code. Second, coordinates of central PDP in QR code are calculated by using run‐length coding. The highlight in this step is the calculation, which utilises modified Knuth–Morris–Pratt algorithm. By this means, the computational complexity can be reduced tremendously. Finally, QR code can be detected successfully with the coordinates. The experimental results show that the proposed method is time saving and suitable for real‐time application.

We consider the problem of jointly describing extreme events at a multitude of locations, which presents paramount importance in catastrophes forecast and risk management. Specifically, a novel Ensemble-of-Latent-Trees of Pairwise Copula (ELTPC) model is proposed. In this model, the spatial dependence is captured by latent trees expressed by pairwise copulas. To compensate the limited expressiveness of every single latent tree, an mixture of latent trees is employed. By harnessing the variational inference and stochastic gradient techniques, we further develop a triply stochastic variational inference (TSVI) algorithm for learning and inference. The corresponding computational complexity is only linear in the number of variables. Numerical results from both the synthetic and real data show that the ELTPC model provides a reliable description of the spatial extremes in a flexible but parsimonious manner.

Cyber crimes often involve complicated scenes. In this paper, we investigate unidentified crimes committed through anonymous communication networks. We developed a long Pseudo-Noise (PN) code based Direct Sequence Spread Spectrum (DSSS) flow marking technique for invisibly tracing suspect anonymous flows. By interfering with a sender's traffic and marginally varying its rate, an investigator can embed a secret spread spectrum signal into the sender's traffic. Each signal bit is modulated with a small segment of a long PN code. By tracing where the embedded signal goes, the investigator can trace the sender and receiver of the suspect flow despite the use of anonymous networks. Benefits of the Long PN code include its resistance to previous discovered detection approaches. We may also use the vast number of long PN code at different phases to conduct parallel tracback without worrying about the interference between codes. Using a combination of analytical modeling and experiments on Anonymizer, we demonstrate the effectiveness of the long PN code based DSSS watermarking technique.

An effective detection algorithm based on the randomness is devised in this paper for stego-webpages with different steganographies. The parts where secret information embedded in a webpage can generally be represented as two states, which can be described in binary code string. The randomness of the states varies a great deal depending on the webpage part carrying secret information or not. This paper presents a procedure to transform the binary code string into octal string to capture the randomness, from which some statistical features have been discovered. The theoretical description and proof are given that these features can be employed as a criterion to test whether a webpage contains secret information or not. Experiments show that this algorithm can effectively detect the stego-webpages based on letter changing in tags and invisible characters embedding.