TLB-pilot: Mitigating TLB Contention Attack on GPUs with Microarchitecture-Aware Scheduling

Co-running GPU kernels on a single GPU can provide high system throughput and improve hardware utilization, but this raises concerns on application security. We reveal that translation lookaside buffer (TLB) attack, one of the common attacks on CPU, can happen on GPU when multiple GPU kernels co-run. We investigate conditions or principles under which a TLB attack can take effect, including the awareness of GPU TLB microarchitecture, being lightweight, and bypassing existing software and hardware mechanisms. This TLB-based attack can be leveraged to conduct Denial-of-Service (or Degradation-of-Service) attacks. Furthermore, we propose a solution to mitigate TLB attacks. In particular, based on the microarchitecture properties of GPU, we introduce a software-based system, TLB-pilot, that binds thread blocks of different kernels to different groups of streaming multiprocessors by considering hardware isolation of last-level TLBs and the application’s resource requirement. TLB-pilot employs lightweight online profiling to collect kernel information before kernel launches. By coordinating software- and hardware-based scheduling and employing a kernel splitting scheme to reduce load imbalance, TLB-pilot effectively mitigates TLB attacks. The result shows that when under TLB attack, TLB-pilot mitigates the attack and provides on average 56.2% and 60.6% improvement in average normalized turnaround times and overall system throughput, respectively, compared to the traditional Multi-Process Service based co-running solution. When under TLB attack, TLB-pilot also provides up to 47.3% and 64.3% improvement (41% and 42.9% on average) in average normalized turnaround times and overall system throughput, respectively, compared to a state-of-the-art co-running solution for efficiently scheduling of thread blocks.

Design and Implementation of Football Player Training Management System Based on Intelligent Image

This article is aimed at studying the design and implementation of a football player training management system based on smart images. Based on the analysis of the importance of informatization for scientific football training, system performance requirements and intelligent image detection technology, the football player training management is designed. The overall architecture of the system, and the detailed design of each functional module of the system. It mainly includes football player information management module, football player training plan viewing module, training goal formulation module and training information feedback module. The realization of the training management system relies on intelligent image technology to detect and track athletes. Finally, the performance of the system was tested. The test results show that the expected response time of each module of the system when different numbers of users are accessed is within 3 seconds. The longest actual time is 2.64 s, and the actual shortest time is 1.18 s. It can be seen that the response time of the system meets the demand. At the same time, the system throughput rate meets the requirements of this article, and the user pass rate is also above 95%, indicating that the performance of the football player training management system designed in this article is better.

Online Maintenance Prioritization via Monte Carlo Tree Search and Case-based Reasoning

When maintenance resources in a manufacturing system are limited, a challenge arises in determining how to allocate these resources among multiple competing maintenance jobs. We formulate this problem as an online prioritization problem using a Markov decision process (MDP) to model the system behavior and Monte Carlo tree search (MCTS) to seek optimal maintenance actions in various states of the system. Further, we use Case-based Reasoning (CBR) to retain and reuse search experience gathered from MCTS to reduce the computational effort needed over time and to improve decision-making efficiency. We demonstrate that our proposed method results in increased system throughput when compared to existing methods of maintenance prioritization while also reducing the time needed to identify optimal maintenance actions as more experience is gathered. This is especially beneficial in manufacturing settings where maintenance decisions must be made quickly.

High-speed 3D imaging using a chaos lidar system

We characterize a new chaos lidar system configuration and demonstrate its capability for high-speed 3D imaging. Compared with a homodyned scheme employing single-element avalanche photodetectors (APDs), the proposed scheme utilizes a fiber Bragg grating and quadrant APDs to substantially increase the system throughput, frame rate, and field-of-view. By quantitatively analyzing the signal-to-noise ratio, peak-to-standard deviation of the sidelobe level, precision, and detection probability, we show that the proposed scheme has better detection performance suitable for practical applications. To show the feasibility of the chaos lidar system, while under the constrain of eye-safe regulation, we demonstrate high-speed 3D imaging with indoor and outdoor scenes at a throughput of 100 kHz, a frame rate of 10 Hz, and a FOV of 24.5$$^\circ $$ ∘ $$\times $$ × 11.5$$^\circ $$ ∘ for the first time.

Orthognal Zones for Interference Migration in 2.4 GHz Mesh Backhaul

Managing interference in the multi-radio networks is critical challenge; problem becomes even more serious in 2.4 GHz band due to minimal availability of orthogonal channels. This work attempts to propose a channel assignment scheme for interference zones of 2.4 GHz backhaul of Wireless Mesh Networks (WMN). The static nodes of Infrastructure based Backhaul employing directional antennas to connect static nodes, orthogonal channel zones introducing Interference are formatted with the selection of single tire direct hop and two tier directional hopes. The effort maintain the orthogonality of channels on system thus reduce the co-channel interference between inter flow and intra flow links. Group of non-overlapping channels of selected band are obtained by a mathematical procedure, interference is modeled by directed graph and Channel assignment is carried out with the help of greedy algorithms. Experimental analysis of the technical proposal is done by simulation through OPNET 14. Our framework can act as an imperative way to enhance the network performance resulting a leading improvement in system throughput and reduction in system delay

Application of Reinforcement Learning and Deep Learning in Multiple-Input and Multiple-Output (MIMO) Systems

The current wireless communication infrastructure has to face exponential development in mobile traffic size, which demands high data rate, reliability, and low latency. MIMO systems and their variants (i.e., Multi-User MIMO and Massive MIMO) are the most promising 5G wireless communication systems technology due to their high system throughput and data rate. However, the most significant challenges in MIMO communication are substantial problems in exploiting the multiple-antenna and computational complexity. The recent success of RL and DL introduces novel and powerful tools that mitigate issues in MIMO communication systems. This article focuses on RL and DL techniques for MIMO systems by presenting a comprehensive review on the integration between the two areas. We first briefly provide the necessary background to RL, DL, and MIMO. Second, potential RL and DL applications for different MIMO issues, such as detection, classification, and compression; channel estimation; positioning, sensing, and localization; CSI acquisition and feedback, security, and robustness; mmWave communication and resource allocation, are presented.

Throughput Performance of A Delay Constraint Transmission For Half-Duplex And Full-Duplex Cognitive Relay Networks

The performance of a relay based Half-Duplex (HD) and Full-Duplex (FD) cooperative cognitive radio (CR) network with a RF energy harvesting (EH) is studied in this paper. Co-operative environment includes a network with multiple primary users (PUs), and CRs. The relay node is considered as an EH node which harvests energy (HE) from RF signal (RFS) of source and loop-back interference. The network performance is studied for instantaneous transmission and delay constraint transmission for decode and forward (DF) relaying protocol. The performance is investigated under a relay energy outage constraint and the expression of throughput is redesigned. Expressions of energy outage, data outage and throughput for HD and FD are developed. The impact of several parameters such as transmitting SNR, fractional harvesting time parameter, fractional transmission time parameter, and loop-back interference on the system throughput has been investigated.

Performance Analysis of MEC Based on NOMA under Imperfect CSI with Eavesdropper

Mobile edge computing (MEC) is becoming more and more popular because of improving computing power in virtual reality, augmented reality, unmanned driving, and other fields. This paper investigates a nonorthogonal multiple access- (NOMA-) based MEC system, which is under imperfect channel state information (ipCSI). In this system model, a pair of users offloads their tasks to the MEC server with the existence of an eavesdropper (Eve). To evaluate the impact of Eve on the performance of the NOMA-MEC system, the secrecy outage probability (SOP) expressions for two users with the conditions of imperfect CSI and perfect channel state information (pCSI) are derived. In addition, both throughput and energy efficiency are discussed in the delay-limited transmission mode. Simulation results reveal that (1) due to the influence of channel estimation errors, the secrecy outage behaviors of two users under ipCSI conditions are worse than those of users with pCSI; (2) the secrecy performance of NOMA-MEC is superior to orthogonal multiple access- (OMA-) aided MEC systems; and (3) the NOMA-MEC systems have the ability to attain better system throughput and energy efficiency compared with OMA-MEC.

A Review of Routing Algorithms Based on Multi-Layer Satellite Networks

Compared with single layer satellite network, satellite distribution of multi-layer satellite networks in the double layer or multilayer orbital plane, combines all the advantages of engaging subjects of the satellite, therefore contains its low vulnerability, good robustness, high stability, high spectrum efficiency and system throughput, channel characteristics such as large capacity, in the satellite network is a very promising direction. Multilayer satellite network also has some inevitable disadvantages, that is, the number of nodes and the number of links is large, so it has the characteristics of frequent changes in network topology, which requires higher routing algorithm. There are many researches on routing algorithms based on multi-layer satellite networks at home and abroad. This paper introduces the general situation of multi-layer satellite network, classifies the existing routing algorithms of multi-layer satellite network according to different standards, analyzes several typical routing algorithms in detail, points out the advantages and disadvantages, and summarizes the future development trend.

Throughput performance for full-duplex DF relaying protocol in hybrid wireless power transfer systems

Most of the existing studies on energy harvesting (EH) cooperative relaying networks are conducted for the outdoor environments which are mainly characterized by Rayleigh fading channels. However, there are not as many studies that consider the indoor environments whereas the state-of-the-art internet of things (IoT) and smart city applications are built upon. Thus, in this paper, we analyze a namely hybrid time-power splitting relaying (HTPSR) protocol in a full-duplex (FD) decode-and-forward (DF) battery-energized relaying network in indoor scenarios modelled by the unpopular log-normal fading channels. Firstly, we formulate the analytical expression of the outage probability (OP) then the system throughput. Accordingly, we simulate the derived expressions with the Monte Carlo method. It is worth mentioning that in our work, the simulation and the theory agree well with each other. From the simulation results, we know how to compromise either the power splitting (PS) or the time splitting (TS) factors for optimizing the system performance.