Institution
Huawei
Company•Shenzhen, China•
About: Huawei is a company organization based out in Shenzhen, China. It is known for research contribution in the topics: Terminal (electronics) & Node (networking). The organization has 41417 authors who have published 44698 publications receiving 343496 citations. The organization is also known as: Huawei Technologies & Huawei Technologies Co., Ltd..
Papers published on a yearly basis
Papers
More filters
•
14 Jan 2013TL;DR: In this article, the authors present a system in which a central control point learns that a virtual machine of a tenant is deployed on a first server, and the central controller point delivers, to the first server a first local virtual local area network identifier VLAN ID pre-configured for the tenant, so that a Virtual Switch vSwitch of the first virtual switch vSwitch stores a correspondence between the tenant and the first local VLANID and can control, according to the correspondence, communication between virtual machines of the tenant in a top of rack switch ToR to which the first
Abstract: The present invention belongs to the field of communications technologies, and discloses a communications method and system. When a central control point learns that a virtual machine of a tenant is deployed on a first server, the central control point delivers, to the first server, a first local virtual local area network identifier VLAN ID pre-configured for the tenant, so that a virtual switch vSwitch of the first server stores a correspondence between the tenant and the first local VLAN ID and can control, according to the correspondence, communication between virtual machines of the tenant in a top of rack switch ToR to which the first server belongs; the central control point delivers, to a first ToR, a first mapping relationship that is between a global VLAN ID and the first local VLAN ID and is pre-configured for the tenant, so that the first ToR stores the first mapping relationship between the global VLAN ID and the first local VLAN ID and can control communication between virtual machines of the tenant in different servers according to the first mapping relationship.
72 citations
••
TL;DR: It is shown that perfect self-interference cancelation plays a crucial role when the SINR threshold is small, since it may result to worse performance in throughput and delay comparing with the half-duplex case.
Abstract: The effect of full-duplex cooperative relaying in a random access multiuser network is investigated here. First, we model the self-interference incurred due to full-duplex operation, assuming multi-packet reception capabilities for both the relay and the destination node. Traffic at the source nodes is considered saturated and the cooperative relay, which does not have packets of its own, stores a source packet that it receives successfully in its queue when the transmission to the destination has failed. We obtain analytical expressions for key performance metrics at the relay, such as arrival and service rates, stability conditions, and average queue length, as functions of the transmission probabilities, the self interference coefficient, and the links' outage probabilities. Furthermore, we study the impact of the relay node and the self-interference coefficient on the per-user and aggregate throughput, and the average delay per packet. We show that perfect self-interference cancelation plays a crucial role when the SINR threshold is small, since it may result to worse performance in throughput and delay comparing with the half-duplex case. This is because perfect self-interference cancelation can cause an unstable queue at the relay under some conditions.
72 citations
•
21 Oct 2020
TL;DR: It is theoretically suggested that the knockoff condition can be approximately preserved given the information propagation of network layers, and can reduce 57.8% parameters and 60.2% FLOPs of ResNet-101 with only 0.01% top-1 accuracy loss on ImageNet.
Abstract: This paper proposes a reliable neural network pruning algorithm by setting up a scientific control. Existing pruning methods have developed various hypotheses to approximate the importance of filters to the network and then execute filter pruning accordingly. To increase the reliability of the results, we prefer to have a more rigorous research design by including a scientific control group as an essential part to minimize the effect of all factors except the association between the filter and expected network output. Acting as a control group, knockoff feature is generated to mimic the feature map produced by the network filter, but they are conditionally independent of the example label given the real feature map. We theoretically suggest that the knockoff condition can be approximately preserved given the information propagation of network layers. Besides the real feature map on an intermediate layer, the corresponding knockoff feature is brought in as another auxiliary input signal for the subsequent layers. Redundant filters can be discovered in the adversarial process of different features. Through experiments, we demonstrate the superiority of the proposed algorithm over state-of-the-art methods. For example, our method can reduce 57.8% parameters and 60.2% FLOPs of ResNet-101 with only 0.01% top-1 accuracy loss on ImageNet. The code is available at this https URL
72 citations
•
10 Aug 2012TL;DR: In this paper, the uplink control channel resources for providing HARQ feedback for data transmitted on a downlink data channel associated with the downlink control channel are assigned. But the uplINK channel resources are configurable over a subset of available time frequency resources.
Abstract: Methods and nodes, for assignment of uplink control channel resources for providing HARQ feedback for data transmitted on a downlink data channel associated with a downlink control channel. The downlink control channel comprises at least a set of downlink control channel resources, which is configurable over a subset of available time frequency resources. The method comprises dividing the set of downlink control channel resources into at least two subsets of references, comprising indices to the respective downlink control channel resources. Also, the method comprises defining, for each such subset of references, an associated subset of references comprising indices to unique uplink control channel resources. Furthermore, the method comprises associating each element in the subset of references with a subframe and an element in the subset of references. Additionally, the method further comprises transmitting data on the downlink data channel associated with the downlink control channel.
72 citations
•
12 May 2015TL;DR: In this paper, a method for reporting channel state information, a user equipment, and a base station is presented, where a UE determines a channel state measurement resource by a UE according to an aperiodic CSI trigger signaling, so that the UE transmits a CSI-RS only when the UE needs to report the CSI.
Abstract: The present disclosure relates to a method for reporting channel state information, a user equipment, and a base station. A channel state information measurement resource is determined by a UE according to an aperiodic CSI trigger signaling, so that a base station transmits a CSI-RS only when the UE needs to report the aperiodic CSI. Compared with transmitting a periodic CSI-RS by an existing base station, unnecessary transmitting of the CSI-RS is reduced, and energy of the base station such as an eNB is saved; and reducing the CSI-RS transmission may reduce resources occupied by the CSI-RS, thereby improving the spectrum efficiency of the system. In addition, reducing unnecessary CSI-RS transmission allows the interference in other cells by the base station being reduced; finally, CSI measurement resources are reserved only when the CSI needs to be fed back, the cost occupied by the CSI measurement resources is reduced.
72 citations
Authors
Showing all 41483 results
Name | H-index | Papers | Citations |
---|---|---|---|
Yu Huang | 136 | 1492 | 89209 |
Xiaoou Tang | 132 | 553 | 94555 |
Xiaogang Wang | 128 | 452 | 73740 |
Shaobin Wang | 126 | 872 | 52463 |
Qiang Yang | 112 | 1117 | 71540 |
Wei Lu | 111 | 1973 | 61911 |
Xuemin Shen | 106 | 1221 | 44959 |
Li Chen | 105 | 1732 | 55996 |
Lajos Hanzo | 101 | 2040 | 54380 |
Luca Benini | 101 | 1453 | 47862 |
Lei Liu | 98 | 2041 | 51163 |
Tao Wang | 97 | 2720 | 55280 |
Mohamed-Slim Alouini | 96 | 1788 | 62290 |
Qi Tian | 96 | 1030 | 41010 |
Merouane Debbah | 96 | 652 | 41140 |