Yanwei Zhao

Bio-inspired rate control scheme for IEEE 802.11e WLANs

The uncontrolled use of limited resources in conjunction with unpredictable nature of traffic load injection in Wireless Local Area Networks (WLANs) may lead to congestion, and cannot guarantee strict Quality of Service (QoS) required by real-time service. Rate control is an important mechanism for the provisioning of QoS in the IEEE 802.11e WLANs. In this paper, a bio-inspired rate control scheme is proposed based on the extended Lotka-Volterra model, which considers the effects of arrival traffic on the system stability according to the limited network resources and competitions with others traffic flows, and ensures that the network works in unsaturated case and rapidly converge to a global stable equilibrium point (EP). Moreover, all traffic flows are of peaceful coexistence and QoS differentiation. Extensive simulations are conducted to illustrate the performance of the proposed rate control scheme.

Multimodal region-consistent saliency based on foreground and background priors for indoor scene

Visual saliency is a very important feature for object detection in a complex scene. However, image-based saliency is influenced by clutter background and similar objects in indoor scenes, and pixel-based saliency cannot provide consistent saliency to a whole object. Therefore, in this paper, we propose a novel method that computes visual saliency maps from multimodal data obtained from indoor scenes, whilst keeping region consistency. Multimodal data from a scene are first obtained by an RGB+D camera. This scene is then segmented into over-segments by a self-adapting approach to combine its colour image and depth map. Based on these over-segments, we develop two cues as domain knowledge to improve the final saliency map, including focus regions obtained from colour images, and planar background structures obtained from point cloud data. Thus, our saliency map is generated by compounding the information of the colour data, the depth data and the point cloud data in a scene. In the experiments, we extensively compare the proposed method with state-of-the-art methods, and we also apply the proposed method to a real robot system to detect objects of interest. The experimental results show that the proposed method outperforms other methods in terms of precisions and recall rates.

A Passive Network Measurement-Based Traffic Control Algorithm in Gateway of P2P Systems

Abstract With the progress of peer-to-peer technology, P2P applications have evolved and established themselves as the leading ‘growth app’ of Internet traffic workload. In order to reserve enough bandwidth for other applications, gateway of P2P systems must control P2Ps traffic. In this paper, passive network measurement of P2P traffic is introduced, and network behaviors of some P2P applications are discussed. Then, one model of receiving and forwarding P2P packets is constructed. One algorithm of traffic control in P2P is designed, which includes two important factors: rejection probability of P2P request and disconnection probability of P2P channel. The results of two parallel experiments show that control algorithm of P2P traffic based on network measurement can reserve enough network resource without disabling all P2P packets.