Tsang-Ling Sheu

An efficient routing scheme with optimal power control in wireless multi-hop sensor networks

Power consumption is a critical problem in providing multimedia communications among wireless sensor nodes (WSNs). To reduce power consumption and satisfying QoS requirements, in this paper, we propose an efficient routing scheme with optimal power management and on-demand quality control for WSNs. Two cost functions are developed to minimize the transmitting power and maximize the link quality under the constraint that an end-to-end frame error probability should be met. The heuristic problem of minimizing power consumption under frame error constraints is formulated and resolved with a closed-form expression. With this closed-form expression, we can determine an optimal route rapidly by calculating the power requirement for each sensor node. Finally, our analytical results indicate that the proposed scheme is superior to a previous work with the same constraint and is also comparable to the results obtained from a heuristic simulation.

Adaptive bandwidth allocation model for multiple traffic classes in IEEE 802.16 worldwide interoperability for microwave access networks

This study presents an adaptive bandwidth allocation model for multiple classes in wireless broadband network, worldwide interoperability for microwave access (WiMAX) (IEEE 802.16). Data to and from Internet are all conveyed through WiMAX links to its final destination. In order to promise the quality of real-time traffic and allow more transmission opportunity for other traffic types, the proposed adaptive bandwidth allocation (ABA) algorithm would first reserve the unsolicited bandwidth for constant-bit-rate traffic (unsolicited grant services). Then, polling bandwidth is allocated for real-time traffic (rtPS) to meet their end-to-end delay constraints and for non-real-time traffic (nrtPS) to meet their minimum throughput requirements. One of the novelties presented by this study is right in that ABA does not greedily overtake too much bandwidth from the lowest-priority class, best effort (BE). Instead, it is very intelligent to only meet the delay constraint of rtPS and the minimum throughput requirement of nrtPS, while it endeavours to avoid any possible starvation of BE traffic. For the purpose of performance evaluation, a four-dimensional Markov chains is built to analyse the proposed ABA. The analytical results are validated by a simulation. Finally, from the comparison with a previous work, the authors observe the performance superiority of the ABA in satisfying the delay constraints (for rtPS), meeting the minimum throughput requirements (for nrtPS) and reducing the average packet drop ratio (for BE), when traffic environments are varied.

A buffer allocation mechanism for VBR video playback

This paper presents a buffer allocation mechanism for VBR (variable bit rate) video playback. VoD (video on demand) is one of the multimedia applications which attract more attention. However, it remains a big challenge for real-time VBR-based VoD to achieve a satisfactory QoS, since the size of VBR video frames can be largely varied and the packets of each frame may encounter different delay variations. The proposed buffer allocation scheme allocates a suitable number of pre-buffers before video streams are pushed out from the server. We first build a mathematical model to estimate accurately the pre-buffer size based on the maximum and the minimum network delay. Then, we determine a better partition interval to minimize the percentages of average buffer size in empty. For the purpose of demonstration, we perform an experiment with a video server pushing video streams to clients through two routers. From the experiment, we have shown that the calculation results of the pre-buffer size before video playback and the dynamically allocated buffer sizes during video presentation are very close to those values from the actual measurements.

A Generalized Channel Preemption Model for Multiclass Traffic in Mobile Wireless Networks

This paper presents a generalized channel preemption model (GCPM) for multiclass traffic in mobile wireless networks. In the proposed model, a mobile call is identified by four parameters - call type, traffic class, channel requirement, and preemption ratio - in which different classes of traffic can have different priority levels. To effectively reduce dropping probability, high-priority handoff calls are allowed to fully or partially preempt low-priority ongoing calls when the mobile network becomes congested. A fully preempted call can be completely disconnected, whereas a partially preempted call can still communicate with certain degree of quality-of-service degradation. We introduce an analytical model with multidimensional Markov chains. In the model, a cost function, which is expressed as a weighted sum of the ongoing calls to be preempted and the bandwidth reduction rate in the preempted calls, is defined to select candidate calls from a certain traffic class. The statistical result from the model allows us to tune the parameters of the GCPM to support high-priority handoff calls and therefore offers new insights in understanding the tradeoffs in supporting multiclass traffic.

A channel preemption model for vertical handoff in a WLAN-embedded cellular network

In this paper, a channel preemption model for vertical handoff in a WLAN-embedded cellular network is presented. In a heterogeneous networking environment, since many wireless LANs may be deployed within the coverage of a cellular network, horizontal handoffs among neighboring WLANs and vertical handoffs between a WLAN and the cellular network could occur frequently. Performance in terms of blocking probability of the cellular network can be seriously degraded if the channels are not appropriately allocated. The novelty of this paper is right in that a newly initiated mobile node (MN) outside the WLAN coverage can preempt the channels occupied by an MN inside the WLAN coverage when the cellular channels are completely used up. The channel preempted MN is forced to switch its network access to a WLAN. This proposed channel preemption scheme can effectively reduce the blocking probability while not disrupting any of the existing connections within WLANs. For the purpose of performance evaluation, we build a three-dimension Markov chains to analyze the proposed channel preemption mechanism. We derive the equations of move-in and move-out mobility rates based on the node speed and residence times, respectively. The network performance in terms of the number of active WLAN users, the channel utilization and the channel blocking probability of a cellular network, the preemption probability, and the preempted probability of an MN are calculated. From the analytical results, we observe the performance improvements by varying the node speed and the ratio of WLAN coverage.

A preemptive channel allocation scheme for multimedia traffic in mobile wireless networks

This paper presents a channel allocation model with both partial and full preemptive capabilities for multimedia traffic in mobile wireless networks. The different types of multimedia traffic in transmitting through next-generation networks possess different characteristics and demand an adequate channel allocation scheme to fulfill individual Quality of Service (QoS) requirements. In the proposed model, multimedia traffic is classified into three classes corresponding to different priority levels in a decreasing order. To effectively reduce the dropping probability, a higher-priority handoff call is allowed to fully or partially preempt any lower-priority ongoing calls when the channel capacity becomes insufficient. In addition, to prevent from possible starvation of the lowest-priority traffic, a portion of system channels are purposely reserved. Performance measures, including the dropping probability of handoff calls, the call-interruption probability that an ongoing call is forced to termination due to full preemption, and the bandwidth reduction ratio due to partial preemption, are investigated through an analytical model built with 4-D Markov chains. The numerical results demonstrate the effectiveness of our model.

An analytical model of two-tier handoff mechanisms for a hierarchical NEMO system

In this paper, we present an analytical model of two-tier handoff mechanisms for a hierarchical NEMO system composed of access routers (ARs), mobile routers (MRs), and visiting mobile nodes (MNs). In the proposed two-tier handoff mechanism, service areas of MR and AR are respectively divided into handoff and non-handoff regions. Consequently, concurrent voice/data sessions in a hierarchical AR-MR model can be classified into four types according to the locations of MN and MR within the MRs and the ARs service areas, respectively. We build a mathematical model with 4-D Markov chains to analyze the performance in terms of the system utilization and the blocking probabilities of new sessions and handoff sessions. For the purpose of validation, we conduct experiments through simulation. Simulation results reveal that both the blocking probabilities of new sessions and handoff sessions are very close to the analytical results, particularly when the system utilization is high. The impacts of session residence time and session arrival rates on the blocking probabilities are also investigated.

Asynchronous Power Management Protocols With Minimum Duty Cycle and Maximum Adaptiveness for Multihop Ad Hoc Networks

IEEE 802.11 is currently the de facto medium access control (MAC) standard for mobile ad hoc networks (MANETs). However, in a multihop MANET, 802.11 power management may completely fail if stations are out of synchronization. To fix this problem, several papers have proposed various cyclic quorum-based power management (CQPM) protocols, which, however, may also fail if stations have different schedule repetition intervals (SRIs). Hence, adaptive CQPM protocols, namely, adaptive quorum-based energy conserving (AQEC) and hyper quorum system (HQS), were proposed to overcome this drawback. However, the duty cycles of AQEC and HQS are far from optimal. In this paper, we propose the optimal fully adaptive and asynchronous (OFAA) power management protocol for a multihop MANET, which has the following attractive features: 1) By means of factor-hereditary quorum space, the OFAA protocol guarantees that two neighboring stations can discover each other in bounded time, regardless of their clock difference and individual SRIs; 2) given the length of SRI, the duty cycle of a station reaches the theoretical minimum; 3) the number of tunable SRIs of every station reaches the theoretical maximum; 4) the time complexity of OFAA neighbor maintenance is O(1) ; 5) a cross-layer SRI adjustment scheme is proposed such that a station can adaptively tune the values of SRI to maximize energy conservation according to flow timeliness requirements. Both theoretical analysis and simulation results demonstrate that the OFAA protocol is much more energy efficient than AQEC and HQS protocols.

A novel hierarchical cache architecture for WWW servers

This paper presents a novel hierarchical cache architecture to achieve efficient Web access. The cache servers being used today have encountered the problems of lack of efficient collaboration among different levels in the hierarchy. In other words, the highest level of the hierarchical cache servers may easily become the performance bottleneck, which in turn slows down the entire cache system. In the proposed new cache design, we apply an inclusive and exclusive relationship to modify the ICP (Internet cache protocol) and use the recursive concept to build a hierarchical cache architecture to avoid the unnecessary information query existing in current ICP. This implies that the lowest level can fetch the data from the original Web server directly and perform the update procedures recursively to its upper levels. The up-side-down traffic flow, referred to as reverse traffic flow in this paper, can substantially release the server load of the upper levels in the hierarchy. For the purpose of evaluation and comparison, we derive mathematical equations by analyzing the operation procedures step by step. The analytical results have shown that the server load of the highest level in the hierarchy can be reduced almost 50% under the worst-case assumption. Although the lower levels may slightly increase their workload, it does significantly decrease the average object searching time by balancing the loads among different levels.

Power minimization with end-to-end frame error constraints in wireless multi-hop sensor networks

Battery life is a very precious resource in sensor networks to provide high-quality multimedia communications among wireless nodes. For reducing power consumption and satisfying QoS requirements, we propose a routing scheme with efficient power management and on-demand quality control for wireless sensor networks. Two cost functions are developed to minimize the transmitting power and maximize the link quality under the constraint that an end-to-end frame error probability should be met. The heuristic problem on minimizing power consumption under frame error constraints can be derived as a closed-form solution in this paper. Therefore, an optimal route can be rapidly determined by accurately calculating the power requirement for each sensor node. Finally, our analytical results indicate that the proposed scheme is superior to a previous work with the same constraint and is also comparable to the results obtained from a heuristic simulation.