http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
OFEX Controller to Improve Queueing and User Performance in Multi-bottleneck Networks
Jungang Liu,Oliver W.W. Yang 한국전자통신연구원 2014 ETRI Journal Vol.36 No.3
We have designed and investigated a new congestioncontrol scheme, called optimal and fully explicit (OFEX)controller. Different from existing relatively explicitcontrollers, this new scheme can provide not only optimalbandwidth allocation but also a fully explicit congestionsignal to sources. It uses the congestion signal from themost congested link instead of the cumulative signal fromthe flow path. In this way, it overcomes the drawback ofrelatively explicit controllers exhibiting bias toward multibottleneckedusers and significantly improves theirconvergence speed and source throughput performance. Furthermore, our OFEX-controller design considers adynamic model by proposing a remedial measure againstthe unpredictable bandwidth changes in contention-basedmulti-access networks. Compared with formerworks/controllers, this remedy also effectively reduces theinstantaneous queue size in a router and thus significantlyimproves queuing delay and packet loss performance. Wehave evaluated the effectiveness of the OFEX controller inOPNET. The experimental comparison with the existingrelatively explicit controllers verifies the superiority of ournew scheme.
Multi-Task FaceBoxes: A Lightweight Face Detector Based on Channel Attention and Context Information
( Shuaihui Qi ),( Jungang Yang ),( Xiaofeng Song ),( Chen Jiang ) 한국인터넷정보학회 2020 KSII Transactions on Internet and Information Syst Vol.14 No.10
In recent years, convolutional neural network (CNN) has become the primary method for face detection. But its shortcomings are obvious, such as expensive calculation, heavy model, etc. This makes CNN difficult to use on the mobile devices which have limited computing and storage capabilities. Therefore, the design of lightweight CNN for face detection is becoming more and more important with the popularity of smartphones and mobile Internet. Based on the CPU real-time face detector FaceBoxes, we propose a multi-task lightweight face detector, which has low computing cost and higher detection precision. First, to improve the detection capability, the squeeze and excitation modules are used to extract attention between channels. Then, the textual and semantic information are extracted by shallow networks and deep networks respectively to get rich features. Finally, the landmark detection module is used to improve the detection performance for small faces and provide landmark data for face alignment. Experiments on AFW, FDDB, PASCAL, and WIDER FACE datasets show that our algorithm has achieved significant improvement in the mean average precision. Especially, on the WIDER FACE hard validation set, our algorithm outperforms the mean average precision of FaceBoxes by 7.2%. For VGA-resolution images, the running speed of our algorithm can reach 23FPS on a CPU device.