• ETRI Journal
• /
• v.39 no.4
• /
• pp.525-534
• /
• 2017

A hardware-acceleration architecture that separates virtual network functions (VNFs) and network control (called HSN) is proposed to solve the mismatch between the simple flow steering requirements and strong packet processing abilities of software-defined networking (SDN) forwarding elements (FEs) in SDN/network function virtualization (NFV) architecture, while improving the efficiency of NFV infrastructure and the performance of network-intensive functions. HSN makes full use of FEs and accelerates VNFs through two mechanisms: (1) separation of traffic steering and packet processing in the FEs; (2) separation of SDN and NFV control in the FEs. Our HSN prototype, built on NetFPGA-10G, demonstrates that the processing performance can be greatly improved with only a small modification of the traditional SDN/NFV architecture.

• Electronics and Telecommunications Trends
• /
• v.38 no.5
• /
• pp.12-22
• /
• 2023

The development of neural networks is evolving towards the adoption of transformer structures with attention modules. Hence, active research focused on extending the concept of lightweight neural network algorithms and hardware acceleration is being conducted for the transition from conventional convolutional neural networks to transformer-based networks. We present a survey of state-of-the-art research on lightweight neural network algorithms and hardware architectures to reduce memory usage and accelerate both inference and training. To describe the corresponding trends, we review recent studies on token pruning, quantization, and architecture tuning for the vision transformer. In addition, we present a hardware architecture that incorporates lightweight algorithms into artificial intelligence processors to accelerate processing.

• Journal of Information Processing Systems
• /
• v.8 no.1
• /
• pp.133-144
• /
• 2012

In order to make cognitive radio systems a practical technology to be deployed in real-world scenarios, the core Software Defined Radio (SDR) systems must meet the stringent requirements of the target application, especially in terms of performance and energy consumption for mobile platforms. In this paper we present a feasibility study of hardware acceleration as an energy-efficient implementation for SDR. We identified the amplifier function from the Software Communication Architecture (SCA) for hardware acceleration since it is one of the functions called for most frequently and it requires intensive floating-point computation. Then, we used the Virtex5 Field-Programmable Gate Array (FPGA) to perform a comparison between compiler floating-point support and the on-chip floating-point support. By enabling the on-chip floating-point unit (FPU), we obtained as high as a 2X speedup and 50% of the overall energy reduction. We achieved this with an increase of the power consumption by no more than 0.68%. This demonstrates the feasibility of the proposed approach.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.6 no.3
• /
• pp.45-53
• /
• 1998

Traction control systems(TCS) improve vehicle acceleration performance and stability, particularly on slippery roads through engine torque and/or brake torque control. This research mainly deals with the engine control algorithm based on adjustment of the engine throttle valve opening. Hardware-in-the-loop simulation(HWILS) is carried out where the actual hardware is used for the engine/automatic transmission and TCS controller, while various vehicle dynamics are simulated on real-time basis. Also, use of the dynamometer is made in order to implement the tractive force that a road applies to the tire. Although some restrictions are imposed mainly due to the capability of the synamometer, simplified HWILS results show that the slip control algorithm can improve the vehicle acceleration performance for low-friction roads.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.41 no.5
• /
• pp.562-572
• /
• 1992

In this paper, we propose software algorithms which provide acceleration/deceleration characteristics essential to high dynamic performance at the transient states where industrial robots or CNC machine tools start and stop. Software acceleration/deceleration methods are derived from the mathematical analyses of typical hardware systems controlling acceleration/deceleration. These methods make servo motors, which drive axes of motion, start and stop smoothly without vibration in the repeated tools. The path error, which is one of the most significant factors in the performance evaluation of industrial robots or CNC machine tools, is analyzed for linear, exponential, and parabolic acceleration/deceleration algorithms in case of circular interpolation. The analyses show that path error consists of the distance between the required path and generated one through acceleration/deceleration, and that between the generated one through acceleration/deceleration algorithm and the actual one of the end effector of the industrial robot or tool of the CNC equipment.

• Journal of the Korean Society for Precision Engineering
• /
• v.17 no.1
• /
• pp.34-44
• /
• 2000

In this paper, a modified skyhook control for the semi-active Macpherson suspension system is investigated. A new model for the semi-active type suspension, which incorporates the rotational motion of the unsprung mass, is introduced and an output feedback control law using the skyhook control method is derived. The gains in the skyhook controller are adaptively adjusted by estimating the road conditions. Because two vertical acceleration sensors, one for the sprung mass and another for the unsprung mass, are used rather than using the angle sensor for the rotational motion of the control arm, the relative velocity of the rattle space is filtered using the acceleration signals. For testing the control performance, the actual damping force has been incorporated via the hardware-in-the-loop simulations. The performances of a passive damper and a semi-active damper are compared. Simulation results are provided.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.907-910
• /
• 1993

This paper describes the use of fuzzy control and decision making to simulate the control of traffic flow at an intersection. To show the value of fuzzy logic as an alternative method for control of traffic environments. A traffic environment includes the lanes to and from an intersection, the intersection, vehicle traffic, and signal lights in the intersection. To test the fuzzy logic controller, a computer simulation was constructed to model a traffic environment. A typical cross intersection was chosen for the traffic environment, and the performance of the fuzzy logic controller was compared with the performance of two different types of conventional control. In the hardware verifications, fuzzy logic was used to control acceleration of a model train on a circular path. For the software experiment, the fuzzy logic controller proved better than conventional control methods, especially in the case of highly uneven traffic flow between different directions. On the hardware si e of the research, the fuzzy acceleration control system showed a marked improvement in smoothness of ride over crisp control.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2002.09a
• /
• pp.61-71
• /
• 2002

Color-code imaging should be one of the best display tools for the quick perception of both damage area and its intensity, In this study we proposed the standard color code corresponding to maximum observed acceleration value. Since the acceleration value changes with frequency contents, we suggested to limit the frequency band less than 10Hz. The background considerations of this limitation, magnitude vs. dominant frequency, natural characteristics of artificial noises and the relationship between velocity and acceleration, were reviewed in detail. The standard of color code was assigned from 0.001%g to 20%g based on the possible earthquakes at Korean peninsula and the color imaging given by RGB value was shown to be very consistent, irrespective of hardware configuration.

• Smart Structures and Systems
• /
• v.7 no.5
• /
• pp.393-416
• /
• 2011

Hybrid acceleration-impedance sensor nodes on Imote2-platform are designed for damage monitoring in steel girder connections. Thus, the feasibility of the sensor nodes is examined about its performance for vibration-based global monitoring and impedance-based local monitoring in the structural systems. To achieve the objective, the following approaches are implemented. First, a damage monitoring scheme is described in parallel with global vibration-based methods and local impedance-based methods. Second, multi-scale sensor nodes that enable combined acceleration-impedance monitoring are described on the design of hardware components and embedded software to operate. Third, the performances of the multi-scale sensor nodes are experimentally evaluated from damage monitoring in a lab-scaled steel girder with bolted connection joints.

• Proceedings of the KIEE Conference
• /
• summer
• /
• pp.117-119
• /
• 2004

An optimal flight motion simulator path generation method is proposed for hardware in the loop simulation of high maneuverable missile. The proposed method consists of open loop and closed loop path calculation algorithm based on the energy optimal control strategies. The optimal angle command is able to protect the simulator from high acceleration shock at initial control phase.