Time series classification is a crucial yet challenging task due to the diverse and complex nature of time series data. While deep learning methods, particularly Convolutional Neural Networks (CNNs), have been widely explored for this problem, transformer-based approaches remain relatively underutilized. With the rise of large language models, recent research has begun to investigate how various data types, including time series, can be integrated into language models. In this paper, we explore the transformation of time series data into symbolic representations for use with language models such as BERT. Symbolizing time series data is a foundational and non-trivial step, as the choice of symbolization method can significantly affect downstream performance. We examine three different symbolization techniques and analyze their impact on classification accuracy. Additionally, we investigate the effect of task-adaptive pretraining using an expanded symbolic time series corpus. By combining these elements, our study provides a analysis of applying language models to time series classification, contributing valuable insights into the effective utilization of symbolic representations and pretrained transformers in this domain.

The rapid and widespread deployment of Internet of Things (IoT) sensors is limited by issues related to maintenance costs and safe battery disposal. While battery-less systems supported by harvesting ambient energy offer a potential solution, the weakness and instability of ambient energy result in intermittent execution. To provide data consistency, I/O operations are usually performed in the atomic sections, resulting in low-energy efficiency and low-system progress in intermittent systems. To improve application responsiveness and prevent low-energy efficiency due to atomic I/O operations, we propose a framework named FASE, which provides energy isolation assistance to support multiple asynchronous atomic I/O operations in intermittent systems. Our proposal was evaluated on a real platform, and the results show that, compared to state-of-the-art works, our proposed algorithm can increase the rate of completion for latency-sensitive applications by more than 50%.

H.264, a state-of-the-art video compression standard, is used across a range of products from cellphones to HDTV. These products have vastly different performance, power and cost requirements, necessitating different hardware-software solutions for H.264 decoding. We show that a design methodology and associated tools which support synthesis from high-level descriptions and which allow modular refinement throughout the design cycle, can share the majority of design effort across multiple design points. Using Bluespec SystemVerilog, we have created a variety of designs for the H.264 decoder tuned to support decoding at resolutions ranging from QCIF video (176 times 144 @ 15 frames/second) to 1080p video ((1280 times 1080)p @60 frames/second) in a 180 nm process. Some of these design points require major transformations of pipelining to increase performance or to reduce area. We also explore several common design issues surrounding memory structures, such as caches and on-chip vs. off-chip memories. We believe the design methodology used in this paper is directly applicable to many IP blocks involving algorithmic specifications. The same design capabilities also permit rapid microarchitecture exploration and changes in RTL late in the design process even in non-algorithmic IP blocks.

In this paper, we present a parallel computing technique for the feature extraction of hyperspectral images. The approach is based on the complete modular eigenspace (CME) scheme, which was designed to extract the simplest and most efficient feature modules by a newly defined multi-dimensional correlation matrix to optimize the modular eigenspace for high- dimensional datasets. The CME feature extraction scheme improves the performance of feature extraction by modifying the correlation coefficient operations. The proposed parallel CME (PCME) scheme is introduced to reduce the computational load of CME feature extraction using the parallel computing technique. It is implemented by parallel virtual machine (PVM) to solve the huge matrix problems of CME feature extraction. The performance of the proposed method is evaluated by applying to hyperspectral images of MODIS/ASTER (MASTER) airborne simulator during the Pacrim II project. The experiments demonstrate the proposed PCME approach is an effective scheme not only for the feature extraction but also for the band selection of high-dimensional datasets. It can improve the precision of hyperspectral image classification compared to conventional multispectral classification schemes.

This research made a non-contact vibration mode scanning probe microscope. The system operation principle was firstly by placing the microprobe in perpendicular to the sample surface, and using signal generator to deliver a sinusoidal drive voltage, then the microprobe would vibrate due to the Coulomb electrostatic force generated by the electrodes on the microprobe. The working frequency is right at the natural resonant frequency of the microprobe. Then let the sample carried by a Z-stage move up. When the sample gets closer to the microprobe, then the vibration amplitude of the microprobe would be reduced and which can be determined by a laser Doppler interferometer, because the Van Der Waal's force between the sample and microprobe would become larger. Thus one can detect the surface profile of the sample, by moving and holding the tip on a constant height, then the probe vibration amplitude is proportion to the surface profile that can be determined by a laser Doppler interferometer, and the topographic data can be obtained by recording the amplitude of vibration history.