We consider the directed minimum latency problem (DirLat), wherein we seek a path P visiting all points (or clients) in a given asymmetric metric starting at a given root node r, so as to minimize the sum of the clien...
详细信息
We consider the directed minimum latency problem (DirLat), wherein we seek a path P visiting all points (or clients) in a given asymmetric metric starting at a given root node r, so as to minimize the sum of the client waiting times along P. We give the first constantfactor approximation guarantee for DirLat, but in quasi-polynomial time. A key ingredient of our result, and our chief technical contribution, is an extension of a recent result of Kohne et al. (2019) [17] showing that the integrality gap of the natural Held-Karp relaxation for asymmetric TSP-Path (ATSPP) is at most a constant. We also give a better approximation guarantee for the minimum total-regret problem, where the goal is to find a path P that minimizes the total time that nodes spend in excess of their shortest-path distances from r, which can be cast as a special case of DirLat involving so-called regret metrics. (c) 2021 Elsevier Inc. All rights reserved.
We study scheduling problems with release times and rejection costs with the objective function of minimizing the maximum lateness. Our main result is a PTAS for the single machine problem with an upper bound on the r...
详细信息
We study scheduling problems with release times and rejection costs with the objective function of minimizing the maximum lateness. Our main result is a PTAS for the single machine problem with an upper bound on the rejection costs. This result is extended to parallel, identical machines. The corresponding problem of minimizing the rejection costs with an upper bound on the lateness is also examined. We show how to compute a PTAS for determining an approximation of the Pareto frontier on both objective functions on parallel, identical machines. Moreover, we present an FPTAS with strongly polynomial time for the maximum lateness problem without release times on identical machines when the number of machines is constant. Finally, we extend this FPTAS to the case of unrelated machines.
The Gaussian process (GP) algorithm is considered as a powerful nonparametric-learning approach, which can provide uncertainty measurements on the predictions. The standard GP requires clearly observed data, unexpecte...
详细信息
The Gaussian process (GP) algorithm is considered as a powerful nonparametric-learning approach, which can provide uncertainty measurements on the predictions. The standard GP requires clearly observed data, unexpected perturbations in the input may lead to learned regression model mismatching. Besides, GP also suffers from the lack of good generalization performance guarantees. To deal with data uncertainty and provide a numerical generalization performance guarantee on the unknown data distribution, this article proposes a novel robust noisy input GP (NIGP) algorithm based on the probably approximately correct (PAC) Bayes theory. Furthermore, to reduce the computational complexity, we develop a sparse NIGP algorithm, and then develop a sparse PAC-Bayes NIGP approach. Compared with NIGP algorithms, instead of maximizing the marginal log likelihood, one can optimize the PAC-Bayes bound to pursue a tighter generalization error upper bound. Experiments verify that the NIGP algorithms can attain greater accuracy. Besides, the PAC-NIGP algorithms proposed herein can achieve both robust performance and improved generalization error upper bound in the face of both uncertain input and output data.
Mobile edge computing has emerged as a prevalent computing paradigm to support applications that demand low latency and high computational capacity. Hardware reconfigurable accelerators exhibit high energy efficiency ...
详细信息
Mobile edge computing has emerged as a prevalent computing paradigm to support applications that demand low latency and high computational capacity. Hardware reconfigurable accelerators exhibit high energy efficiency and low latency compared to general-purpose servers, making them ideal for integration into mobile edge computing systems. This article investigates the problem of joint task offloading, access point selection, and resource allocation in heterogeneous edge environments for latency minimization. Given the heterogeneity of edge computing devices and the interdependence of the decisions required for offloading, access point selection, and resource allocation, it is challenging to optimize over them simultaneously. We decomposed the proposed problem into two disjoint subproblems and developed algorithms for each of them. The first subproblem is to jointly determine access point selection and communication resource allocation decisions, for which we have proposed an algorithm with a provable approximation ratio of 2.62/(1-8 lambda), where lambda is a tunable parameter balancing the approximation ratio and time complexity. Additionally, we offer a faster variant of the algorithm with an approximation ratio of (root 3+1)(2). The second subproblem is to determine offloading and computing resource allocation decisions jointly and is NP-hard, where we developed algorithms based on relaxation and rounding. We conducted comprehensive numerical simulations to evaluate the proposed algorithms, and the results demonstrated that our algorithms outperformed existing baselines and achieved near-optimal performance across various settings.
In this paper, we investigate the assignment of intelligent reflecting surfaces (IRSs) to user-base station (BS) pairs in a multi-IRS-assisted wireless communication network. Our objective is to optimize the allocatio...
详细信息
In this paper, we investigate the assignment of intelligent reflecting surfaces (IRSs) to user-base station (BS) pairs in a multi-IRS-assisted wireless communication network. Our objective is to optimize the allocation of one or more IRSs to each user-BS pair to maximize the overall sum-rate of the network. Using passive beamforming, the transmitted signal is directed towards each user through single or multiple IRSs. To achieve optimal user-IRS allocation, we employ the modified Hungarian algorithm. Additionally, we propose a greedy algorithm for user-IRS association, which offers lower complexity than the modified Hungarian algorithm while still aiming to maximize the sum-rate of the network. Simulation results demonstrate the superiority of modified Hungarian algorithm over other techniques. It has been observed that assigning two IRSs per user-BS pair significantly improves the sum-rate compared to assigning a single IRS per user-BS pair. In particular, with the transmission power of 10 dB and 128 reflection coefficients, the modified Hungarian algorithm achieves a sum-rate improvement of approximately 11.6% compared to the greedy algorithm and 20.6% compared to the random IRS assignment.
In this work, we study the compression of multichannel signals with irregular sampling rates and data gaps. We consider state-of-the-art algorithms, which were originally designed to compress gapless signals with regu...
详细信息
In this work, we study the compression of multichannel signals with irregular sampling rates and data gaps. We consider state-of-the-art algorithms, which were originally designed to compress gapless signals with regular sampling, adapt them to operate with signals with irregular sampling rates and data gaps, and then evaluate their performance experimentally, through the compression of signals obtained from real-world datasets. Both the original algorithms and our schemes compress signals by exploiting their temporal, and, in some cases, spatial correlation. They work in a near-lossless fashion, guaranteeing a bounded absolute error between each decompressed sample and its original value. This includes the important lossless compression case, which corresponds to an error bound of zero. Our schemes first encode the position of the gaps, using arithmetic coding combined with a Krichevsky-Trofimov probability assignment on a Markov model, and then encode the data values separately. Our experimental analysis consists of comparing the compression performance of our schemes with each other, and with representative special-purpose and general-purpose lossless compression algorithms. We also measure and compare the schemes' running times, to assess their practicality. From the results we extract some general conclusions: in the lossless case, TS2Diff and LZMA attain the best compression performance, whereas our adaptation of algorithm APCA is preferred for positive error bounds. At the same time, our adaptation of APCA, and TS2Diff, attain some of the best running times.
In this letter, we address the weighted sum-rate maximization problem in a cell-free massive multi-input multi-output (CF-mMIMO) system, subject to constraints on the minimum individual quality of service (QoS), maxim...
详细信息
In this letter, we address the weighted sum-rate maximization problem in a cell-free massive multi-input multi-output (CF-mMIMO) system, subject to constraints on the minimum individual quality of service (QoS), maximum power consumption at each access point (AP), and maximum fronthaul capacity. By harnessing the low computational complexity weighted minimum mean square error (WMMSE) framework, two algorithms are proposed to solve the formulated mixed integer nonlinear programming (MINLP) problems with instantaneous/statistical channel state information (CSI). Our instantaneous CSI-based approach can be applied to jointly optimize the power control, precoding, and user association, while the statistical CSI-based approach can be utilized to jointly optimize the power control and user association. Simulation results demonstrate that the proposed instantaneous CSI-based algorithm can provide approximately 66.72% sum-rate gain compared to the scheme with random user association, equal power allocation, and fixed local MMSE-based precoding design. Additionally, the statistical CSI-based algorithm offers competitive performance compared with the instantaneous CSI-based algorithm.
Precision ranging technology has become indispensable for ensuring efficient, reliable, and low-latency fifth-generation (5G) networks. In this paper, we propose a novel ranging method which is multipath component (MP...
详细信息
Precision ranging technology has become indispensable for ensuring efficient, reliable, and low-latency fifth-generation (5G) networks. In this paper, we propose a novel ranging method which is multipath component (MPC) power delay profile (PDP) based ranging. Whereas the Received Signal Strength (RSS) only summarizes the PDP into a single characteristic, we aim to furthermore exploit the range dependent curvature of the PDP envelope over its delay spread. However, the multipath propagation only allows to sample the PDP envelope at the path delays and suffers from (slow) fading. Hence our approach involves constructing a statistical fading model of the PDP and establishing a relationship between the distribution parameters and the propagation distance. To theoretically validate the feasibility of our proposed method, we adopt the widely accepted Nakagami-m fading model, which renders traditional estimation methods difficult to apply. Therefore we introduce the Expectation Maximization (EM)-Revisited Vector Approximate Message Passing (ReVAMP) algorithm. This algorithm is specifically designed to handle difficulties in parameter estimation for Gaussian linear models (GLMs) with hidden random variables and intractable posterior distributions. Extensive numerical simulation results have been conducted which exhibit preliminary evidence of the effectiveness of our MPCPDP-based ranging method compared to the received signal strength (RSS)-based method. Moreover, the versatility of the EM-ReVAMP algorithm allows for its extension to other statistical fading models beyond the Nakagami-m model with minor modifications, which opens the door to potential improvements based on more accurate statistical fading models. Nevertheless, the applicability of our MPCPDP-based ranging method should be validated in real-world scenarios in future studies.
A wide range of collaborative human-robot applications across diverse domains, encompassing manufacturing, industry, and social interactions, necessitate an autonomous robot to follow its human companion. However, the...
详细信息
A wide range of collaborative human-robot applications across diverse domains, encompassing manufacturing, industry, and social interactions, necessitate an autonomous robot to follow its human companion. However, the great degree of flexibility and unpredictability of human mobility makes the integration of multi-sensor measurements for human-following difficult. This paper addresses these challenges by demonstrating the design and implementation of a multi-sensor person-following system. On the one hand, a target tracking control method based on detection of center boxes is proposed. On the other hand, the positioning, navigation and tracking modules are integrated into the robot system. Exhaustive functional verification is conducted on an Ackermann-steering mobile robot, in which an embedded device is the core algorithm board. Notably, we discuss the degradation phenomenon of Light Detection and Ranging (LiDAR) odometry. The positioning error of our system is 2.42%, which is much smaller than other advanced methods. In indoor experiments, our system enables the robot to stably track the target at a distance of approximately 1.5 m. The effectiveness of the entire system is verified, demonstrating its advantage in real-time performance.
作者:
Zhang, LiChen, XiaoboHainan Normal Univ
Key Lab Data Sci & Intelligence Educ Minist Educ Hainan 571158 Peoples R China Jiangsu Univ Technol
Sch Comp Engn Changzhou 213001 Jiangsu Peoples R China Peoples Bank China
Changzhou City Ctr Branch Changzhou 213001 Jiangsu Peoples R China Jilin Univ
Key Lab Symbol Computat & Knowledge Engn Minist Educ Changchun 130012 Peoples R China
The rapid growth of data quantity directly leads to the increasing feature dimension, which challenges machine learning and data mining. Wrapper-based intelligent swarm algorithms are effective solution techniques. Th...
详细信息
The rapid growth of data quantity directly leads to the increasing feature dimension, which challenges machine learning and data mining. Wrapper-based intelligent swarm algorithms are effective solution techniques. The Grey Wolf Optimization (GWO) algorithm is a novel intelligent population algorithm. Simple principles and few parameters characterize it. However, the basic GWO has disadvantages, such as difficulty coordinating exploration and exploitation capabilities and premature convergence. As a result, GWO fails to identify many irrelevant and redundant features. To improve the performance of the basic GWO algorithm, this paper proposes a velocity-guided grey wolf optimization algorithm with adaptive weights and Laplace operators (VGWO-AWLO). Firstly, by introducing a uniformly distributed dynamic adaptive weighting mechanism, the control parameters $a$ are guided to undergo nonlinear dynamic changes to achieve a good transition from the exploratory phase to the development phase. Second, a velocity-based position update formula is designed with an individual memory function to enhance the local search capability of individual grey wolves and drive them to converge to the optimal solution. Thirdly, a Laplace cross-operator strategy is applied to increase the population diversity and help the grey wolf population escape from the local optimal solution. Finally, the VGWO-AWLO algorithm is evaluated for its comprehensive performance in terms of classification accuracy, dimensionality approximation, convergence, and stability in 18 classified datasets. The experimental results show that the classification accuracy and convergence speed of VGWO-AWLO are better than the basic GWO, GWO variants, and other state-of-the-art meta-heuristic algorithms.
暂无评论