A Practical and Efficient Node Blind SignCryption Scheme for the IoT Device Network

In recent years, Internet of Things (IoT for short) research has become one of the top ten most popular research topics. IoT devices also embed many sensing chips for detecting physical signals from the outside environment. In the wireless sensing network (WSN for short), a human can wear several IoT devices around her/his body such as a smart watch, smart band, smart glasses, etc. These IoT devices can collect analog environment data around the user’s body and store these data into memory after data processing. Thus far, we have discovered that some IoT devices have resource limitations such as power shortages or insufficient memory for data computation and preservation. An IoT device such as a smart band attempts to upload a user’s body information to the cloud server by adopting the public-key crypto-system to generate the corresponding cipher-text and related signature for concrete data security; in this situation, the computation time increases linearly and the device can run out of memory, which is inconvenient for users. For this reason, we consider that, if the smart IoT device can perform encryption and signature simultaneously, it can save significant resources for the execution of other applications. As a result, our approach is to design an efficient, practical, and lightweight, blind sign-cryption (SC for short) scheme for IoT device usage. Not only can our methodology offer the sensed data privacy protection efficiently, but it is also fit for the above application scenario with limited resource conditions such as battery shortage or less memory space in the IoT device network.

One for “All”: a unified model for fine-grained sentiment analysis under three tasks

Background Fine-grained sentiment analysis is used to interpret consumers’ sentiments, from their written comments, towards specific entities on specific aspects. Previous researchers have introduced three main tasks in this field (ABSA, TABSA, MEABSA), covering all kinds of social media data (e.g., review specific, questions and answers, and community-based). In this paper, we identify and address two common challenges encountered in these three tasks, including the low-resource problem and the sentiment polarity bias. Methods We propose a unified model called PEA by integrating data augmentation methodology with the pre-trained language model, which is suitable for all the ABSA, TABSA and MEABSA tasks. Two data augmentation methods, which are entity replacement and dual noise injection, are introduced to solve both challenges at the same time. An ensemble method is also introduced to incorporate the results of the basic RNN-based and BERT-based models. Results PEA shows significant improvements on all three fine-grained sentiment analysis tasks when compared with state-of-the-art models. It also achieves comparable results with what the baseline models obtain while using only 20% of their training data, which demonstrates its extraordinary performance under extreme low-resource conditions.

Improving thai-lao neural machine translation with similarity lexicon

It has been shown that the performance of neural machine translation (NMT) drops starkly in low-resource conditions. Thai-Lao is a typical low-resource language pair of tiny parallel corpus, leading to suboptimal NMT performance on it. However, Thai and Lao have considerable similarities in linguistic morphology and have bilingual lexicon which is relatively easy to obtain. To use this feature, we first build a bilingual similarity lexicon composed of pairs of similar words. Then we propose a novel NMT architecture to leverage the similarity between Thai and Lao. Specifically, besides the prevailing sentence encoder, we introduce an extra similarity lexicon encoder into the conventional encoder-decoder architecture, by which the semantic information carried by the similarity lexicon can be represented. We further provide a simple mechanism in the decoder to balance the information representations delivered from the input sentence and the similarity lexicon. Our approach can fully exploit linguistic similarity carried by the similarity lexicon to improve translation quality. Experimental results demonstrate that our approach achieves significant improvements over the state-of-the-art Transformer baseline system and previous similar works.

FastQR: Fast Pose Estimation of Objects Based on Multiple QR Codes and Monocular Vision in Mobile Embedded Devices

In recent years, the pose estimation of objects has become a research hotspot. This technique can effectively estimate the pose changes of objects in space and is widely used in many mobile devices, such as AR/VR. At present, mainstream technologies can achieve high-precision pose estimation, but the problem of that of multiple irregular objects in mobile and embedded devices under limited resource conditions is still challenging. In this paper, we propose a FastQR algorithm that can estimate the pose of multiple irregular objects on Renesas by utilizing homography method to solve the transformation matrix of a single QR code and then establish the spatial constraint relationship between multiple QR codes to estimate the posture of irregular objects. Our algorithm obtained a competitive result in simulation and verification on the RZ/A2M development board of Renesas. Moreover, the verification results show that our method can estimate the spatial pose of the multiobject accurately and robustly in distributed embedded devices. The average frame rate calculated on the RZ/A2M can reach 28 fps, which is at least 37 times faster than that of other pose estimation methods.

Comparison of Fault Simulation Over Custom Kernel Module Using Various Techniques

To test the behavior of the Linux kernel module, device drivers and file system in a faulty situation, scientists tried to inject faults in different artificial environments. Since the rarity and unpredictability of such events are pretty high, thus the localization and detection of Linux kernel, device drivers, file system modules errors become unfathomable. ‘Artificial introduction of some random faults during normal tests’ is the only known approach to such mystifying problems. A standard method for performing such experiments is to generate synthetic faults and study the effects. Various fault injection frameworks have been analyzed over the Linux kernel to simulate such detection. The following paper highlights the comparison of different approaches and techniques used for such fault injection to test Linux kernel modules that include simulating low resource conditions and detecting memory leaks. The frameworks chosen to be used in these experiments are; Linux Text Project (LTP), KEDR, Linux Fault-Injection (LFI), and SCSI.

Limiting-Stress-Elimination Hypothesis: Using Non-hormonal Biostimulant to Reduce Stress and Increase Savanna Cowpea [Vigna unguiculata (L.) Walp.] Productivity

An alternative decision axiom to guide in determining the optimal intervention strategy to maximize cowpea production is proposed. According to the decrement from the maximum concept of Mitscherlich, the decrement from the maximum for each stressor must be minimized to produce the absolute maximum production. In crop production, this means all deficient nutrients must be supplemented to ensure maximum yield and laid the foundation in fertilizer formulation. However, its implementation is not economically feasible in many situations, particularly where multiple environmental factors impact crop productivity as in the case of low resource conditions. We propose and test the hypothesis that yield allocation will increase when the most limiting stressor among prevailing stressors is eliminated at least until the next limiting stressor impacts productivity. We selected drought limiting savanna conditions and cowpea (Vigna unguiculata), adapted to nitrogen dependence. To determine the limiting condition, we measured the response of cowpea to D-sorbitol, nitrogen, and non-hormonal biostimulant (nhB) treatments. The nhB treatment increased total biomass by 45% compared to nitrogen, 13%, and D-sorbitol, 17%, suggesting osmotic stress is more limiting in the observed savanna conditions. The effect of the biostimulant is due to antioxidants and key amino acids that stimulate metabolism and stress resistance. Where nitrogen becomes the next constraining factor, biostimulants can contribute organic nitrogen. The study supports the use of biostimulants as candidate intervention under conditions where crop productivity is limited by multiple or alternating constraints during crop growth.