• The Journal of Korea Robotics Society
• /
• v.18 no.1
• /
• pp.88-92
• /
• 2023

Lithium-ion batteries are actively used in various industrial sites such as field robots, drones, and electric vehicles due to their high energy efficiency, light weight, long life span, and low self-discharge rate. When using a lithium-ion battery in a field, it is important to accurately estimate the SoC (State of Charge) of batteries to prevent damage. In recent years, SoC estimation using data-based artificial neural networks has been in the spotlight, but it has been difficult to deploy in the embedded board environment at the actual site because the computation is heavy and complex. To solve this problem, neural network lightening technologies such as network pruning have recently attracted attention. When pruning a neural network, the performance varies depending on which layer and how much pruning is performed. In this paper, we introduce an optimized pruning technique by improving the existing pruning method, and perform a comparative experiment to analyze the results.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.9
• /
• pp.3449-3467
• /
• 2015

A broadcast operation is the fundamental transmission technique in mobile ad-hoc networks (MANETs). Because a broadcast operation can cause a broadcast storm, only selected forwarding nodes have the right to rebroadcast a broadcast message among the one-hop and two-hop neighboring nodes of a sender. This paper proposes the maximum intersection self-pruning (MISP) algorithm to minimize broadcasting redundancy. Herein, an example is given to help describe the main concept of MISP and upper bounds of forward node have been derived based on induction. A simulation conducted demonstrated that when conventional blind flooding (BF), self-pruning (SP), an optimized link state routing (OLSR) multipoint relay (MPR) set, and dominant pruning (DP), are replaced with the MISP in executing Strong duplicate address detection (DAD), the performances in terms of the energy consumption, upper bounds of the number of forward nodes, and message complexity have been improved. In addition, to evaluate the performance in reference to the link error probability, P_e, an enhancement was achieved by computing a proposed retransmission limit, S, for error recovery based on this probability. Retransmission limit control is critical for efficient energy consumption of MANET nodes operating with limited portable energy where Strong DAD reacts differently to link errors based on the operational procedures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.4
• /
• pp.1931-1953
• /
• 2017

The reverse skyline query plays an important role in information searching applications. This paper deals with continuous reverse skyline queries in sensor networks, which retrieves reverse skylines as well as the set of nodes that reported them for continuous sampling epochs. Designing an energy-efficient approach to answer continuous reverse skyline queries is non-trivial because the reverse skyline query is not decomposable and a huge number of unqualified nodes need to report their sensor readings. In this paper, we develop a new algorithm that avoids transmission of updates from nodes that cannot influence the reverse skyline. We propose a data mapping scheme to estimate sensor readings and determine their dominance relationships without having to know the true values. We also theoretically analyze the properties for reverse skyline computation, and propose efficient pruning techniques while guaranteeing the correctness of the answer. An extensive experimental evaluation demonstrates the efficiency of our approach.

• ETRI Journal
• /
• v.44 no.5
• /
• pp.746-758
• /
• 2022

This paper focuses on a mobile edge-computing-enabled heterogeneous network. A battery level-aware task-scheduling framework is proposed to improve the energy efficiency and prolong the operating hours of battery-powered mobile devices. The formulated optimization problem is a typical mixed-integer nonlinear programming problem. To solve this nondeterministic polynomial (NP)-hard problem, a decomposition-based task-scheduling algorithm is proposed. Using an alternating optimization technology, the original problem is divided into three subproblems. In the outer loop, task offloading decisions are yielded using a pruning search algorithm for the task offloading subproblem. In the inner loop, closed-form solutions for computational resource allocation subproblems are derived using the Lagrangian multiplier method. Then, it is proven that the transmitted power-allocation subproblem is a unimodal problem; this subproblem is solved using a gradient-based bisection search algorithm. The simulation results demonstrate that the proposed framework achieves better energy efficiency than other frameworks. Additionally, the impact of the battery level-aware scheme on the operating hours of battery-powered mobile devices is also investigated.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.6
• /
• pp.850-858
• /
• 2022

To deploy Gate Recurrent Units (GRU) on resource-constrained embedded devices, this paper presents a reconfigurable FPGA-based GRU accelerator that enables structured compression. Firstly, a dense GRU model is significantly reduced in size by hybrid quantization and structured top-k pruning. Secondly, the energy consumption on external memory access is greatly reduced by the proposed reuse computing pattern. Finally, the accelerator can handle a structured sparse model that benefits from the algorithm-hardware co-design workflows. Moreover, inference tasks can be flexibly performed using all functional dimensions, sequence length, and number of layers. Implemented on the Intel DE1-SoC FPGA, the proposed accelerator achieves 45.01 GOPs in a structured sparse GRU network without batching. Compared to the implementation of CPU and GPU, low-cost FPGA accelerator achieves 57 and 30x improvements in latency, 300 and 23.44x improvements in energy efficiency, respectively. Thus, the proposed accelerator is utilized as an early study of real-time embedded applications, demonstrating the potential for further development in the future.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.5
• /
• pp.169-176
• /
• 2020

Recently, deep neural networks (DNNs) have brought revolutions to many mobile and embedded devices by providing human-level machine intelligence for various applications. However, high inference accuracy of such DNNs comes at high computational costs, and, hence, there have been significant efforts to reduce computational overheads of DNNs either by compressing off-the-shelf models or by designing a new small footprint DNN architecture tailored to resource constrained devices. One notable recent paradigm in designing small footprint DNN models is sharing parameters in several layers. However, in previous approaches, the parameter-sharing techniques have been applied to large deep networks, such as ResNet, that are known to have high redundancy. In this paper, we propose a parameter-sharing method for already parameter-efficient small networks such as ShuffleNetV2. In our approach, small templates are combined with small layer-specific parameters to generate weights. Our experiment results on ImageNet and CIFAR100 datasets show that our approach can reduce the size of parameters by 15%-35% of ShuffleNetV2 while achieving smaller drops in accuracies compared to previous parameter-sharing and pruning approaches. We further show that the proposed approach is efficient in terms of latency and energy consumption on modern embedded devices.