FPGAs are being virtualized to improve resource utilization in data centers. Memory access performance is essential to FPGA hypervisors for shared-memory FPGA platform, where accelerators access memory spontaneously. DMA remapping with IOMMU provides a handy solution; however, fixed IOMMU can not benefit from the reconfigurability of FPGAs. In this work, we propose MEGATRON, a hybrid address translation service consisting of a hardware TLB and a software page table walker. By integrating MEGATRON into an existing FPGA hypervisor, we conduct a comprehensive analysis of link performance of a multi-link CPU-FPGA platform, and demonstrate the competitiveness of the customizable translation service.

GPUs have become ubiquitous in the Cloud-rendering areas due to the outstanding rendering performance. However, many existing Cloud-rendering systems suffer from low GPU utilization caused by the CPU bottleneck. Recent proposals (e.g., API-forwarding and c-GPU) for GPU-usage optimization are promising but fail to address the system-resource redundancy issues (i.e., each instance tends to occupy all the system resources exceeding their requirements), leading to unnecessary CPU consumption and lowering GPU utilization. We conducted an experiment by testing real-world applications on the percentage of unused resources to demonstrate the severity of this issue. Nearly 50% of resources are unused.

Traditional GPU resource allocation approaches, widely adopted in today's data centers, only focus on the server-side functions while ignoring the client-side. These approaches waste client-side hardware resources. To solve this problem, remote API-forwarding architectures appear. Through running applications on the client-side, remote API-forwarding architectures offload some workloads to the client. However, many remote API-forwarding systems suffer from one big issue: shared-resource interference, stemming from two reasons: (a) GPU resource racing caused by resource overuse for a single client, and (b) CPU resource racing caused by resource shortage among clients. This paper presents gRemote, an open-source GPU-remoting system that can address this issue. To mitigate the CPU resource shortage, gRemote improves CPU configurations by expanding CPU resources from the server-side to both server- and client-side. To maintain the reasonable GPU usage for individual tasks, we innovate a new resource-sharing mechanism called GPU throttle. gRemote supports 1,228 OpenGL commands with around 10% shared-resource interference.

Cloud rendering is an emerging technology in which rendering-heavy applications run on the cloud server and then stream the rendered contents to the end-user device. High density and high scalability of the cloud rendering services are crucial to support millions of users concurrently and cost-effectively. However, it is still challenging to run Android OS in cloud smoothly with high density and high scalability without compromising user experience. This paper presents DroidCloud, the first open-source Android\footnoteAndroid is a trademark of Google LLC. cloud rendering solution focusing on the scalable design and density aspect optimization to the best of our knowledge. To cloudify Android OS, DroidCloud utilizes thevHAL technology in order to support remote devices and keep transparent to Android applications. And aFlexible rendering scheduling policy is introduced to break the boundary of GPU physical locations. Thus, both remote GPUs and local GPUs can accommodate render tasks by forwarding rendering tasks and making it possible to support multiple Android OSes with GPU acceleration. Besides, to further improve the density, DroidCloud optimizes the resource cost both in a single instance and across instances. We show that DroidCloud can run hundreds of Android OSes on a single Intel Xeon server with GPU acceleration simultaneously, increasing the density at the scale of one order of magnitude compared to current cloud gaming systems. Further experimental results demonstrate that DroidCloud can transparently run Android applications at native speed with lower CPU, memory, and storage utilization.

A compact dual band antenna operating simultaneously at 5.4 GHz and 15 GHz is presented in this paper. It is a planar array composed of a grid of 6×8 (48) patch antennas working at 5.4 GHz and 7×9 (63) patch antennas working at 15 GHz. The whole antenna array occupies an area of 150 mm × 180 mm which fits the ordinary size requirement of wireless access point. All the antenna patch antennas have a reflection coefficient lower than −6 dB across 320 and 2000 MHz bandwidth at 5 and 15 GHz. The electromagnetic coupling between any two antennas is lower than − 10 dB within the specified bandwidth. Measurement results show good agreement with the simulation of input matching, mutual coupling and radiation pattern. Compared to the ordinary antenna array design with half-wavelength spacing between two neighboring antennas, our proposed dual band array design accommodates 1.6 times more antennas at 5.4 GHz and extra 63 antennas at 15 GHz, while maintaining sufficient radiation efficiency, matching bandwidth and decoupling.

Online tracking is a challenging task to develop effective and efficient models to account for appearance change. However, most tracking algorithms only consider the holistic or local information and do not make full use of the appearance information. In this study, a novel tracking algorithm with sparse representation is proposed and the online classifier is learned to discriminate the target from the background. To reduce visual drift problem which is encountered in object tracking, a two‐stage sparse representation method is proposed. The holistic information is used to estimate the initial tracking position, and the local information is used to determine the final tracking position. To improve the performance of the classifier and robustness of the algorithm, the kernel function is applied on the sparse representation. Moreover, the dimension of the target is reduced via compressive sensing. Besides, a simple and effective method for dictionary update is proposed. Both qualitative and quantitative evaluations on challenging image sequences demonstrate that the proposed algorithm performs favourably against several state‐of‐the‐art algorithms.

In this paper, non-linear predictors based on different artificial neural network (ANN) are compared. General radical basis function (GERBF) neural network, a modified RBF network, is introduced. From the comparison with BP, RNN and RBF, it is obvious that GERBF has priority in the prediction of speech signal. The experiment results show: the speech coding systems based on ANN have better synthesized speech than ITU's G721 and the speech coding system based on GERBF has higher mean segmental SNR but less computation than that of other systems.

A novel methodology utilizing the direct data domain approach to space-time adaptive processing (STAP) in airborne radar environments is presented in this paper. The deterministic least squares adaptive signal processing technique operates on a snapshot-by-snapshot basis to determine the adaptive weights for nulling interferences and estimating signal of interest (SOI), and eliminates the requirement for the secondary data from neighboring range cell. This is in contrast to conventional adaptive techniques where processing is done by taking the time averages as opposed to spatial averages. Simulation results illustrate the efficiency of interference suppression in nonhomogeneous environment.