Transient execution vulnerabilities have emerged as a critical threat to modern processors. Hardware fuzzing testing techniques have recently shown promising results in discovering transient execution bugs in large-scale out-of-order processor designs. However, their poor microarchitectural controllability and observability prevent them from effectively and efficiently detecting transient execution vulnerabilities.

Memory corruption vulnerabilities pose a significant threat to system security. The traditional paging-based approach cannot protect fine-grained runtime data (e.g., function pointers), which are often mixed with other data in memory. To protect the runtime data, data space randomization is proposed to encrypt the in-memory data so that the attacker cannot control the decrypted result. Unfortunately, current hardware does not provide dedicated support for fine-grained data encryption. This article presents RegVault II, a cross-architectural hardware-assisted lightweight data randomization scheme for OS kernels. To achieve robust, fine-grained, and lightweight data protection, we first identify five required capabilities for efficient and secure data randomization. Guided by these requirements, we design and implement novel hardware primitives that provide cryptographically strong encryption and decryption, thus ensuring both confidentiality and integrity for register-grained data. At the software level, we propose identification- and annotation-based approaches to automatically mark sensitive data and instrument the corresponding load and store operations. We also introduce new techniques to protect the interrupt context and safeguard the sensitive data spilling. We implement RegVault II on an actual FPGA hardware board for RISC-V and on QEMU for Arm, applying it to protect six types of sensitive data in the Linux kernel. Our thorough security and performance evaluations show that RegVault II effectively defends against a broad range of kernel data attacks while incurring minimal performance overhead.

Concurrent indexes for persistent memory (PM) have been extensively investigated due to the appealing features of PM, such as data persistence and DRAM-comparable performance. Among them, Adaptive Radix Tree (ART) is a widely used index that performs well on variable-sized keys. Nevertheless, existing persistent ARTs still suffer from high PM overhead as well as inefficient concurrency control.In this paper, we present Heart, a persistent ART with low latency and high scalability. Heart proposes a unified node structure for different capacities with Hashed Node and Node Decoupling to reduce the PM access overhead. To achieve high scalability, especially in write-intensive scenarios, we propose an efficient concurrency control protocol with lock-free basic operations and lock-free node split. Furthermore, Heart employs Perceivable Transformation to avoid anomalies during node expansion and shrinkage. Compared to other state-of-the-art persistent ARTs, Heart achieves up to 21.6× higher performance under YCSB workloads with high memory utilization. We also deploy Heart in DRAM and obtain up to 33.4× speedup than the original in-memory ART.

This paper presents RegVault, a hardware-assisted lightweight data randomization scheme for OS kernels. RegVault introduces novel cryptographically strong hardware primitives to protect both the confidentiality and integrity of register-grained data. RegVault leverages annotations to mark sensitive data and instruments their loads and stores automatically. Moreover, RegVault also introduces new techniques to protect the interrupt context and safeguard the sensitive data spilling. We implement a prototype of RegVault by extending RISC-V architecture to protect six types of sensitive data in Linux kernel. Our evaluations show that RegVault can defend against the kernel data attacks effectively with a minimal performance overhead.

The mean shift algorithm shows a good performance in optical image segmentation. However, conventional mean shift algorithm performs poorly if it is used directly to synthetic aperture radar (SAR) image due to the large dynamic range and strong speckle noise. Recently, a generalized mean shift (GMS) algorithm with an adaptive variable asymmetric bandwidth was proposed for polarimetric SAR (PolSAR) image filtering. In this paper, it is further developed and extended for PolSAR image segmentation. The proposed algorithm can be used for PolSAR image superpixel segmentation directly without any preprocessing steps. Experiments using AirSAR and ESAR L-band PolSAR data demonstrate the effectiveness of the proposed superpixel segmentation algorithm.

As the development of marine economy and population explosion, coastal areas is suffering great pressure - because of the immigration from inland to the developed cities along east China. Island coastal zones, which is a specific ecosystem surrounded by the sea, is more sensitive to human activities, e.g. reclamations. It is essential to monitor the dynamic changes of the island coastal areas to retrieve the siltation pattern of the surrounding open-sea and their impacts to island coastlines using remote sensing technique. In this paper, a time-series monitoring using Landsat images is performed to monitor the changes of a sand dam in the island coast zone, aiming at analyzing the effects of human exploitation. The results show great potential of using remote sensing images for coast zone dynamic monitoring.

In mining areas, wastes are usually piled-up in the dump, making a man-made hill with height of more than tens of meters. Because of the loose structure of dump sites, landslides or debris flow may occur after heavy rainfall, threatening local people's lives and properties. Therefore, dump stability monitoring and early warning for possible hazards is crucial for ensuring local safety. In this paper, a collaborative stability analysis with high-resolution SAR data and ground measurements is conducted over Xudonggou dump of Anqian iron mine. A two-pass DInSAR analysis is carried out to derive the deformation distribution over the whole area. High-precision GPS and robotic measurements on several points are also conducted over the dump site. Based on the measurements, an slope stability analysis is carried out and eventually a landslide hazard zone is outlined.

The radiometric quality of polarimetric SAR (PolSAR)/SAR images is affected by terrain undulations, and the resultant radiometric distortions should be calibrated to facilitate quantitative applications as land cover classification. This paper presents a terrain-related radiometric calibration method to a quad-polarimetric Advanced Land Observing Satellite phased array type L-band synthetic aperture radar (ALOS PALSAR) image. A digital elevation model (DEM) was used for accurate detection of layover and shadow areas. Precise calibration was done subsequently. Polarimetric features were extracted and a supervised random forest (RF) classifier was then employed. Five classes were extracted as waterbody, bare soil, farmland, forest, and man-made objects. Accuracy assessment was performed and the results were analyzed. Improvement of overall accuracy from 78.67% to 82.67% and that of kappa coefficient from 0.73 to 0.78 was achieved using the radiometric-terrain-calibrated (RTC) features, which shows great necessity of RTC processing for PolSAR land cover classification in mountainous areas.

In this paper, we investigate the separability of targets with weak backscattering on synthetic aperture radar (SAR) images by using of an unsupervised classification method. This technique is a combination of the Cloude target decomposition and the likelihood ratio test based on complex Wishart distribution for the polarimetric covariance matrix. The polarimetric SAR (PolSAR) image is initially classified by the H - α plane into eight classes. The dissimilar distance measure is derived from the statistical test of equality of covariance matrices. Significant improvement of the classification results are observed for weak backscattering targets in iterations. The effectiveness of this algorithm is demonstrated using a Radarsat-2 PolSAR image in C band and an ALOS PALSAR PolSAR image in L band.