Shan Guo Quan

Fast Data Aggregation Algorithm for Minimum Delay in Clustered Ubiquitous Sensor Networks

We propose a time-efficient local data aggregation algorithm that aggregates delay-constrained data within a given time deadline in clustered wireless sensor networks. Our approach consists of two phases. First, we design a zone-based fast data aggregation tree (ZFDAT) to eliminate unnecessary packets being forwarded to many receiver nodes and avoid long detour paths until cluster-head, where cluster-head is local control center to coordinate the data transmissions in the cluster. Next, we propose optimal link scheduling algorithm to minimize aggregation time by given variable length of time slots for all links in the ZFDAT. Our simulation results show that our approach can minimize the aggregation time under energy constraint and outperforms the similar existing algorithms. In particular, our approach distributes the total aggregation time, then it can also reduce the heavy processing load at the cluster-head.

Utility-based Rate Allocation Scheme for Mobile Video Streaming over Femtocell Networks

Abstract: This paper proposes a utility-based data rate allocation algorithm to provide high-quality mobile video streaming over femtocell networks. We first derive a utility function to calculate the optimal data rates for maximizing the aggregate utilities of all mobile users in the femtocell. The total sum of optimal data rates is limited by the link capacity of the backhaul connections. Furthermore, electromagnetic cross-talk poses a serious problem for the backhaul connections, and its influence passes on to mobile users, as well as causing data rate degradation in the femtocell networks. We also have studied a fixed margin iterative water-filling algorithm to achieve the target data rate of each backhaul connection as a counter-measure to the cross-talk problem. The results of our simulation show that the algorithm is capable of minimizing the transmission power of backhaul connections while guaranteeing a high overall quality of service for all users of the same binder. In particular, it can provide the target data rate required to maximize user satisfaction with the mobile video streaming service over the femtocell networks.

Adaptive File Distribution in P2P Network Using Ant Colony Optimization for Smart Home Environment

In this paper, we consider the file distribution problem in peer-to-peer networks. We propose a novel algorithm, called AntReplica, for an efficient and adaptive replication of large files in the Internet environment. AntReplica is inspired by the stigmergy model of communication observed in ant colonies, and allows for peers to experience the short distribution delay. In order to replicate an original file among group member nodes, the file is partitioned into small chips of equal size, and each chip is transferred to a member node in proportion to the probability of selecting a link. Then, each node propagates it to the remaining nodes in the group. Proposed method achieves significantly reduced file replication time as we demonstrate through experiment with an eventdriven simulator.

Multicast Partition Problem for Smart Home Applications

In smart home, multimedia services can be provided to users, who want to receive multimedia data through mobile phones. Most of multimedia applications are implemented using the streaming technique from contents provider to subscribers. It is important to manage the buffering amount remained in the user side in order to support the adequate quality of video contents and minimize complaints of subscribers due to buffer starvation. In the wireless environment, it is difficult to provide several subscribers with multicast traffic stably, since the data rate which users can receive varies according to the channel state of each user. Therefore, we propose and discuss an efficient usage together with MIU (Multicast In Unicast portion) as well as MIM (Multicast In Multicast portion) in order to maximize the aggregate throughput of users and minimize the starvation duration degrading QoS supported to users. We consider the multicast partitioning problem applying list partitioning algorithm and provide the analytical foundation that the proposed scheme can be evaluated. Simulation results show that proposed model can provide enhanced performance and allowed streaming quality.

A design for distortion-aware cooperative relaying over wireless video sensor networks

This paper describes a distortion-aware cooperative relaying (DACR)scheme for distortion minimized video transmission over wireless video sensor networks. We first introduce a simple performance metric considering the different spatial and temporal error propagation effects existing in a video packet. Then, we propose the DACR algorithm which achieves gradual video quality degradation due to packet loss under conditions of limited channel resources. Experimental results show that the proposed scheme outperforms other existing scheme, and can also support optimal video quality for variable channel status.

Maximizing network lifetime under reliability constraints using a cross-layer design in dense wireless sensor networks

Recent experimental studies have shown that radio links between low-power devices are extremely unreliable. In particular, the instability and unpredictability of low-power radio links in dense wireless sensor networks(WSNs) makes it extremely challenging to improve energy efficiency for reliable communication. In this paper, we propose a novel energy efficient cross-layer based scheme for WSNs. We consider the joint design of the physical(PHY)layer and the medium access control( MAC) to maximize network lifetime under reliability constraints. Our approach consists of opportunistic transmission power control and cell-based efficient time slot assignment. Our approach can support more reliable and energy-efficient data transmission in dense cluster-based WSNs. Performance evaluation results show that our cross-layer based approach outperforms traditional layered protocols. Furthermore, our approach is more energy efficient and can maximize the network lifetime. The network lifetime in our approach has more than doubled compared with the current scheme.

Asymmetric link dynamic routing algorithm for ubiquitous multimedia services

In this paper, we propose an asymmetric link dynamic routing algorithm for ubiquitous multimedia services named DRAMA (Dynamic Routing Algorithm for asymmetric link in Mobile Ad hoc networks). The proposed algorithm allows a first sink node of the unidirectional link to detect the unidirectional link and its reverse route and provide information regarding the reverse route to other sink nodes, thereby minimizing unnecessary flows and effectively using the unidirectional link. DRAMA also incooperates redirection process that actively supports a unidirectional link generated during data transmission to deal with a topology change due to the movement of node. Further, we shows the superiority of DRAMA to the existing algorithms in terms of a delay in data transmission, throughput, and transmission overhead through simulation study.

Opportunistic cross-layer power control for high density clustered wireless sensor networks

In this paper, we propose a novel transmission power control scheme to improve energy efficiency in high density clustered wireless sensor networks(WSNs). Our scheme is based on cross-layer strategy that combines both medium access control (MAC) layer and physical (PHY) layer. The algorithm consists of an efficient time slot allocation and opportunistic transmission power control to support energy-efficient intra- cluster communication. Performance evaluation results reveal that our approach is more energy efficient and can maximize the network lifetime in high density clustered WSNs. Our approach can achieve a network lifetime of up to 2 times longer than that generated by the current schemes. Furthermore, our approach outperforms traditional layered protocols.