Pouring snacks and moving containers with beverages are challenging for a service robot. To obtain accurate content properties for planning robotic motion, tactile sensing can provide information about the pressure distribution of the contact surface, which is not obvious by visual observation. In this work, we focus on estimating the content properties of various content materials in distinct deformable containers through tactile interactions. We propose a learning-based model that can estimate content properties by using the tactile data collected by slightly squeezing a container with the content of interest. We analyzed an uncalibrated tactile sensor and collected a dataset consisting of 1125 sets of tactile sequences, which are combinations of five types of deformable containers and eleven types of content materials in different content heights. Experiments were conducted on content estimation with known contents and containers, unknown contents, and unknown containers. For unknown contents, our model can still achieve 8.5% height relative error and 79.7% state of matter accuracy. Furthermore, we analyzed that the tactile features of contents with similar content properties are close in the latent snace to show the effectiveness of our model.

In this article, we focus on rearranging an object by pushing it to any target planar pose. We identify the essential elements to guarantee that the target pose can be reached at an acceptable precision. We present a simple rearrangement algorithm that relies on only a few known straight-line pushes for some novel object and requires no analytical models, force sensors, or large training datasets. We derive the step upper bound, which relates the initial pose of the object, stopping criterion, and quality of the set of pushes, to facilitate the estimation of the maximum number of required steps without the need to perform a task. We experimentally verified the performance of our algorithm at different noise levels, stopping criteria, and task difficulties on datasets containing several types of objects. By applying combinations of only nine known pushes, our simple algorithm performed successfully in real-world experiments with challenging objects, including partially deformable objects that are difficult to model analytically; it achieved the precise stopping criterion (7.5 mm, 5$^\circ$) in various rearrangement tasks.

Studies on object rearrangement planning typically consider known objects. Some learning-based methods can predict the movement of an unknown object after single-step interaction, but require intermediate targets, which are generated manually, to achieve the rearrangement task. In this work, we propose a framework for unknown object rearrangement. Our system first models an object through a small-amount of identification actions and adjust the model parameters during task execution. We implement the proposed framework based on a low-cost tabletop robot (under 180 USD) to demonstrate the advantages of using a physics engine to assist action prediction. Experimental results reveal that after running our adaptive learning procedure, the robot can successfully arrange a novel object using an average of five discrete pushes on our tabletop environment and satisfy a precise 3.5 cm translation and 5° rotation criterion.

Structured-light and stereo cameras, which are widely used to construct point clouds for robotic applications, have different limitations on estimating depth values. Structured-light cameras fail in black, transparent, and reflective objects, which influence the light path; stereo cameras fail in texture-less objects. In this work, we propose a depth fusion model that complements these two types of methods to generate high-quality point clouds for short-range robotic applications. The model first determines the fusion weights from the two input depth images and then refines the fused depth using color features. We construct a dataset containing the aforementioned challenging objects and report the performance of our proposed model. The results reveal that our method reduces the average L1 distance on depth prediction by 75% and 52% compared with the original depth output of the structured-light camera and the stereo model, respectively. A noticeable improvement on the Iterative Closest Point (ICP) algorithm can be achieved by using the refined depth images output from our method.

The performance of an OpenCL program is strongly influenced by both hardware and software attributes. To achieve superior performance, developers may leverage automatic performance tuning techniques to determine the optimal parameters on the target device. Although existing approaches have shown promising tuning results in their target scenarios, other requirements such as efficiency, portability, and usability should also be considered because of the rapid growth of heterogeneous computing applications and platforms. In this paper, we re-examine the workgroup size tuning problem and propose a novel approach to meet the aforementioned requirements. We abstract the architectural details into a set of hardware parameters so that the proposed approach can be applied without the presence of target devices, which makes it more accessible to developers. The proposed approach is evaluated on 20 OpenCL kernels and six devices, including both CPUs and GPUs. Experimental results demonstrate that, with negligible overhead, our approach filters out 88.6 percent of the possible workgroup sizes on average. Among all the workgroup size candidates, the bestand worst-performing candidates can achieve average performance of 95.5 and 92.1 percent, respectively, compared with the optimal workgroup size.

This paper presents an object-centric method for efficiently performing two types of challenging pick-and-place tasks, namely sequential pick and place and object sorting. We propose multiclass dense object nets (MCDONs) for learning object-centric dense descriptors that maintain not only intra-class variations but also inter-class separation. Intra-class consistency is also inherently learned and is useful for our pick-and-place tasks. All the tasks only require a single demonstration from users, which can then be generalized to all class instances. A dataset containing eight classes and a total of 52 objects was provided in this study. We obtained a task success rate of 93.33% on a five-block stacking task and 97.41% on a three-class object sorting task.

Previous studies have used on-chip thermal sensors (diodes) to estimate the leakage power of a CPU. However, an embedded CPU equips only a few thermal sensors and may suffer from considerable spatial temperature variances across the CPU core, and leakage power estimation based on insufficient temperature information introduces errors. According to our experiments, the conventional leakage power models may have up to 22.9% estimation error for a 70-nm embedded CPU. In this study, we first evaluated the accuracy of leakage power estimates based on thermal sensors on different locations of a CPU and suggested locations that can reduce the error to 0.9%. Then, we proposed temperature-referred and counter-tracked estimation (TRACE) that relies on temperature sensors and hardware activity counters to estimate leakage power. The simulation results demonstrated that employing TRACE could reduce the error to 3.4%. Experiments were also conducted on a real platform to verify our findings.

Heterogeneous CPU-GPU systems have recently emerged as an energy-efficient computing platform. A robust integrated CPU-GPU simulator is essential to facilitate researches in this direction. While few integrated CPU-GPU simulators are available, similar tools that support OpenCL 2.0, a widely used new standard with promising heterogeneous computing features, are currently missing. In this paper, we extend the existing integrated CPU-GPU simulator, gem5-gpu, to support OpenCL 2.0. In addition, we conduct experiments on the extended simulator to see the impact of new features introduced by OpenCL 2.0. Our OpenCL 2.0 compatible simulator is successfully validated against a state-of-the-art commercial product, and is expected to help boost future studies in heterogeneous CPU-GPU systems.

Distributed renewable energy has emerged as a promising resource because of its environmental friendliness and economic considerations. However, most renewable energy sources are unreliable and may require considerable effort to be efficiently utilized in a computing center for providing services. In this paper, we exploit distributed renewable energy (e.g., solar energy) and peer-to-peer (P2P) technologies to aggregate distributed computing resources to provide an infrastructure called solar-based energy-efficient distributed server (SEEDS) farm for distributed computing and distributed storage. Energy-efficient devices (e.g., embedded devices) powered by solar energy form a P2P computing system to provide their computing resources to end-users. Specifically, this paper uses a Web-based service as a case study. A group of solar-powered embedded devices acts as a front-end cooperative caching system for Web servers. Web objects may be accessed through the distributed caching system without going through servers, and thus we can reduce brown energy consumption of the servers. This paper also develops an analytical model to evaluate the total energy consumption of Web-based services with and without SEEDS. Theoretical and simulation results show that the SEEDS system can support services and achieve significant improvements in energy efficiency by aggregating distributed energy resources.

In recent years, thermal management, which improves the reliability, performance, power leakage, etc. of modern microprocessors, has been the subject of numerous computer architecture and system software studies. To determine the detailed thermal distribution of a microprocessor is among the critical tasks for thermal management. However, because thermal modeling tools require considerable computation time and memory to simulate fine-grain thermal information, they may be unsuitable for dynamic thermal management and hardware implementation. This study proposes a novel model based on reduced resistance-capacitance (RC) networks for efficiently calculating the temperature of a microprocessor. The proposed model is compared with two existing thermal simulation tools, namely, HotSpot [1] and Temptor [2]. The experiment studies show that the results generated using the proposed model differ from those of the existing tools by only 0.5 to 1.5%. However, the suggested model can increase computation speeds by 5 to 9 times and 98 to 161 times that of Temptor and HotSpot, respectively. For the memory usage, the proposed model consumes merely 0.45% of the space used by the existing tools.

Multimedia playback programs must satisfy certain hard real-time constraints to give their user a pleasant viewing experience. Having a tight estimation on the worst case execution time (WCET) is essential for the resource scheduler to ensure all the tasks in the system can satisfy their timing requirements while maintaining good system resource utilization. In this paper we present a detailed hybrid WCET analysis of the H.264/AVC reference decoder and demonstrate that the coding information can be used to tighten the pure static WCET estimation up to 70%.

Current heterogeneous wireless networks often overlap because of their complementary characteristics and the large deployment of various wireless access technologies. Mobile devices, which are equipped with multiple wireless interfaces, called multiple radio access technologies (multi-RATs) mobile stations (MSs), are also becoming increasingly popular. Therefore, common radio resource management (CRRM) has been proposed to coordinate heterogeneous radio resource allocations and improve the resource utilization of heterogeneous wireless networks. However, CRRM is an NP-hard problem, and low-complexity approaches for dynamic resource management are in high demand. In this paper, the resource request and allocation between multi-RAT MSs and heterogeneous wireless networks are modeled as a directed graph. Thus, the problem of searching for feasible radio resource allocations is simplified to finding trees in a directed graph. Based on the proposed model, a heuristic scheme can be used to find a feasible solution efficiently, and dynamically adjust the workload of heterogeneous BSs to accommodate new requests. Experimental results demonstrate that the heuristic scheme can reduce the request reject rate by 10%-55% compared with conventional approaches.

Random access channels (RACHs) in cellular networks are normally designed for Poisson-distributed arrivals with a constant rate. Unexpected bursty arrivals may result in severely collisions in RACHs and thus, degrade users' service qualities. This paper presents an analytical model to investigate the transient behavior of the RACHs with bursty arrivals generated in a specific time interval in OFDMA wireless networks. The proposed model considers the effect of new arrivals and the real-world implementation constraints of the OFDMA random access procedure such as periodic access characteristic, uniform random backoff policy, and power-ramping effect. The average number of contending and success mobile stations (MSs) in each random-access slot are then derived using the proposed model. The accuracy of the proposed model was verified through computer simulations and the results show the accuracy of the analysis.

Intelligent Transportation Systems (ITSs) have attracted much attention recently. One of the most important services for ITSs is to provide real-time traffic information service. To reduce the deployment cost and achieve high service availability, decentralized approaches such as vehicular ad hoc network (VANET)-based or infrastructure-based peer-to-peer (P2P) traffic information systems were proposed. To further improve the success rate and lookup latency of traffic information lookups, a new two-tier traffic information system which integrates low-tier VANET and high-tier P2P overlay has been developed. However, conventional two-tier traffic information system may introduce extra routing overheads since the information lookups are broadcasted over two tier networks and introduce redundant lookup messages. This paper proposes an adaptive routing mechanism in the two-tier traffic information system to improve the efficiency of traffic information lookup. Simulation results demonstrate that the proposed adaptive scheme reduces the lookup latency and overhead while achieving the same lookup success rate compared with the conventional approach.

This paper summarizes the two definitions of the collision probability from the perspective of an MTC device and an RAO were presented in 3GPP TR 37.868. We use the results presented in TR 37.868 to show the inconsistency between the two definitions. It worth to note that the researcher needs to clearly specify the definition they used in presenting the results of their proposed radio access network overload control schemes. This paper further presents an analytical model to derive the collision probability, the success probability, and the idle probability based on the two definitions.

Code repositioning is a well-known method of reducing inefficient off-chip memory accesses by streamlining cache behavior. Embedded systems with predetermined applications can achieve further improvement with the addition of fast and energy efficient scratchpad memory (SPM) on chip and moving frequent accesses code and/or data from main memory to SPM. While many researchers have attempted to either streamline cache accesses or improve the effectiveness of SPM, few studies focus on exploring their joint synergy. This study proposes integer linear programming (ILP) models that include both code repositioning and SPM code selection to identify the optimal code layout and reduce energy consumption in embedded systems with a cache and SPM. This study also proposes a two-stage metaheuristic algorithm. Experimental results reveal that 1) allocating a dedicated portion of the on-chip SRAM to the SPM is not always better than using a cache-only configuration and 2) it is not trivial to select code objects for the SPM. As much as 55 percent additional energy can be saved by applying both code repositioning and SPM code selection techniques.

In Mobile WiMAX, a mobile station (MS) scans neighboring base stations (BSs) before conducting handover. The MS may perform the association further, i.e., initial ranging, with neighboring BSs during scanning to obtain ranging parameters and/or service availability information for a potential handover to a BS and/or the selection of target BSs. However, the scanning with association scheme may introduce additional latency for handover. This study first evaluates the conventional scanning with association process, and then proposes a mechanism to reduce the association latency. The performance models of the proposed mechanism are presented and the accuracies of analytical models are verified by simulations. Results demonstrate that the proposed mechanism reduces the association latency by 61.9%-78.0% for audio services and 41.3%-65.4% for video services.

To minimize the power consumption of a mobile device, mobile WiMAX based on IEEE 802.16/802.16e defines three power saving classes for transmitting different types of data traffic. However, the standard specification does not suggest a power saving class for transmitting background TCP traffic, such as a long FTP session. This paper investigates the energy efficiency of TCP transmission over a WiMAX network under different MAC configurations, wireless and wired traffic conditions. We conduct a number of experiments in a real WiMAX test-bed and obtain power consumption parameters of a WiMAX network interface card (NIC). These parameters are further fed into the network simulator 2 (NS-2) to evaluate the energy consumption of TCP transmission in a large-scale WiMAX network. Experimental and simulation results demonstrate that the IEEE 802.16 bandwidth request mechanism introduces considerable power consumption overhead in TCP transmission. Moreover, the energy consumption of a long TCP transmission session in a WiMAX network can be significantly reduced by applying the power saving class type II and base station-proxy mechanism.

One of the most important issues in offering real-time communication services in a mobile environment is support for seamless handover between base stations during a communication session. In Mobile WiMAX, a mobile station may have to perform a scan and association process before handover, but unfortunately, this scan and association process introduces service disruption. In this letter, we investigate the scan and association latency in Mobile WiMAX, and evaluate its performance through analytic models and simulations.

Wireless relay network is a low-cost solution for extending the service coverage and increasing the throughput of the conventional single-hop wireless network. By deploying a relay network, a base station (BS) can provide services to subscriber stations (SSs) through relay links. The functionality of a relay station (RS) is simpler than a BS, and RSs only forward packets between SSs, RSs and the BS without too much packet processing. However, current solutions require a RS to store and process the packet before it can determine the destination, and thus introduce extra processing and buffering overheads on the RS. This study proposes a simple and efficient forwarding scheme which multiplexes bursts to the SSs attached to the same access RS. By applying the proposed end-to-end burst switching mechanism, the RS merely checks the frame header and can decide the final destination without packet processing. The experiment results demonstrate the superior performance of the proposed scheme than other approaches.

Abstract Voice over IP (VoIP) over WLAN (VoWLAN) is an important application for public and private WLANs. However, VoWLAN systems suffer from several technical challenges such as power consumption of a WLAN station (STA) and service capacity of an access point (AP), making the commercial deployment of a large‐scale VoWLAN service problematic. This study presents a cross‐layer and energy‐efficient mechanism for transmitting VoIP packets over IEEE 802.11 WLAN. The proposed mechanism considers the characteristics of voice packets that can tolerate certain loss, and dynamically disables the medium access control (MAC) layer acknowledgement for voice packets. In doing so, the time and energy consumed to transmit and receive voice packets for an STA can be reduced. Simulation results demonstrate that the mechanism improves the energy efficiency of a VoWLAN STA and WLAN utilization without sacrificing voice qualities. Copyright © 2009 John Wiley & Sons, Ltd.

The embedded software engineers are highly demanded in recent several years in order to support fast development of SoCs and embedded systems. These engineers need both strong hardware/software knowledge and hands-on experiences of system-level software. Unfortunately, the practical training of the system software development such as the OS design and implementation is often insufficient for computer science students in Taiwan recently. To minimize the gap between theory and practical implementation, a practical implementation course of OSs is thus developed in National Chiao Tung University. The course provides students trainings on the design details of modern OSs and comprehensive hands-on practices on OS implementation. In this paper, the curriculum and hands-on lab design, the teaching experiences and student feedbacks for the trial run are presented.

Conventional traffic information systems based on client-server approaches allow car drivers to navigate real-time road traffics, but the cost for maintaining the centralized traffic centers is high. On the other hand, decentralized traffic information systems that traffic conditions are exchanged between vehicles via inter-vehicle communication (IVC) significantly reduce the construction cost and efforts. Unfortunately, while the penetration of vehicles with such the system is insufficient, the systems suffer from problems to disseminate traffic conditions in disconnected vehicular ad-hoc networks. Therefore, we propose a novel road traffic information system which utilizes both IVC and infrastructure-based peer-to-peer communications. The system is organized as a two-tier hierarchical architecture where all vehicles broadcast traffic conditions over lower-tier vehicular ad-hoc networks. Moreover, parts of vehicles with infrastructure communication facilities further form a high-tier peer-to-peer overlay and exchange traffic information and lookups to redeem the dis-connectivity of the vehicular ad-hoc networks. Simulation results demonstrate that the proposed system performs well in terms of both delay for traffic information lookup and message exchanging overheads.

Embedded systems such as mobile phones widely adopt heterogeneous dual-core processors to improve the performance and reduce the cost and power consumption. Considering mobile devices that are usually battery-operated, one of the critical issues for such an embedded system is software design which must fully utilize the computation power and minimizes the power consumption. A number of power consumption models were proposed to facilitate designers to evaluate and diagnose the power consumption of the software. Unfortunately, previous studies only investigated single-core processors, but have not yet developed power consumption models for heterogeneous dual-core processors. In this paper, a power consumption model for the dual-core processor based on TI OMAP5912 is proposed for evaluating the power consumption of embedded software. This tool not only estimates the energy consumed by embedded software but also analyzes the energy distributions of the software on each processor core. With the aid of this tool, designers could diagnose the power consumption of dual-core embedded software efficiently and easily.

People who travel to a place may not know well about the information of the sites. Fortunately, people may have a personal device such as a mobile phone which has a digital camera and broadband wireless accesses, and the device can serve as a convenient equipment to access the information of geographic locations when people visit the place. This paper presents a system, called the GLOBE (GLobe On moBilE), consisting of a mobile, a client and a server. The GLOBE mobile is a mobile phone equipped with GPS, height, direction, angle sensors which are used to estimate the three-dimensional (3D) geographic location on the earth of the objects in the photo. The GLOBE client and server can manage these photos with 3D location information, annotate the objects on the photo, associate the information and travel notes with the photos, and share their photos with people on the Internet. Therefore, people just use the GLOBE mobile to take a picture on the object that they feel interested during a trip, and can immediately obtain the information of the object by searching on the GLOBE server. The proposed system helps people to learn information behind sites and objects they visited, and helps users to manage and share the information during a trip.

Presents the introductory welcome message from the conference proceedings. May include the conference officers' congratulations to all involved with the conference event and publication of the proceedings record.

Session Initiation Protocol (SIP) has been widely used as the call setup protocol for multimedia communications such as Voice over IP (VoIP). The client-server architecture, however, exposes SIP servers to the risks of performance bottleneck and the single point of failure. Peer-to-Peer (P2P) SIP which is implemented by using the Distributed Hash Table (DHT)-based Chord is consequently proposed with the advantages of scalability and reliability by replacing centralized servers with the self-organized P2P network. Unfortunately, previous studies on P2P SIP have not yet considered the mobility issue which is an essential property of mobile Internet. While users roam in wireless networks, to ensure a stable DHT-based Chord introduces serious problems. In this paper, we propose a new unstructured P2P SIP system for mobile Internet telephony. An unstructured P2P network is less influenced by the churn property than a structured P2P network, and it is suitable for mobile environment. Moreover, the proposed unstructured P2P SIP network is constructed according to the relationships between SIP users and their friends. Therefore, SIP users can find their friends in the proposed P2P SIP network very easily. Also, the social network exhibits the small-world property, and an SIP user can locate non-friends in the proposed network within a small number of hops. Simulations results demonstrate that the proposed method introduces less maintenance overhead and achieves better performance than the conventional Chord-based approach.

In a heterogeneous overlay wireless network, a mobile node (MN) with multiple wireless interfaces can attach to various networks and obtains multiple IP addresses simultaneously. To allow correspondent nodes (CNs) to connect to an MN at anytime through any MN's IP address, an MN must keep its interfaces awake to receive incoming packets. Unfortunately, the previous studies indicate that power consumption of an MN with multiple interfaces poses a serious problem, even when interfaces are idle. This work presents two novel energy-conserving always-on schemes for an MN with multiple interfaces. These schemes only maintain one awake interface of an MN and turn off completely other interfaces of the MN to minimize its power consumption. By updating proper binding information on home agents (HAs), packets sent to the IP addresses associated with these turned-off interfaces can be received by the MN without packet loss.

The design of packet transmission mechanisms of video and voice over wireless networks generally utilizes the characteristics of real-time data to improve the network performance, the energy efficiency of mobile devices, and etc. Unfortunately, WLAN medium access control (MAC) layer does not distinguish real-time and data packets during MAC frame transmission, and this design introduces network overheads and extra power consumption of video and voice over WLAN (V 2 oWLAN) stations (STAs). In this paper, we propose a transmission scheme to eliminate WLAN MAC acknowledgment frames of real-time packets dynamically in order to reduce the power consumption of V 2 oWLAN STAs and to improve the network utilization without sacrificing video and voice qualities. The proposed transmission scheme utilizes the WLAN multicasting method that does not require MAC acknowledgment, and the method thus can be implemented on the existing WLAN infrastructures by only software upgrades. Simulation results demonstrate that the proposed scheme improves the network utilization and the energy efficiency of V 2 oWLAN applications.

Voice over WLAN (VoWLAN) service is regarded as one of the killer applications for public and private WLANs, but it suffers from a number of technical challenges, especially the power consumption problem. In order to conserve the energy of a VoWLAN device during idle, the power saving mode (PSM) defined in the IEEE 802.11 is widely employed. Experimental results indicate that 300ms to 1000ms listen intervals of PSM for a VoWLAN device could achieve better energy efficiencies. Unfortunately, low-end WLAN access points (APs) typically have limited memory spaces, and can buffer incoming packets for a small time window, say 100ms. For a VoWLAN device requesting long listen intervals, packets queued on the APs might be dropped or overwritten when the buffer overflows. This phenomenon introduces considerable call setup delays for session initiation protocol (SIP) that implements the exponential retransmission mechanism of lost messages. This study investigates the problem and proposes three approaches to reduce call setup delays for VoWLAN applications. Simulation results demonstrate that the proposed mechanisms reduce call setup delays significantly for low-power VoWLAN devices.

Broadband wireless access systems usually provide flexible sleep-mode operations for mobile stations to conserve energy during idle or active modes. For example, the IEEE 802.16e offers several sleep-mode classes that can be applied to mobile subscriber stations (MSSs) to reduce their power consumptions for different applications. Unfortunately, previous studies did not fully utilize sleep-mode features, and the power consumption of an MSS with multiple network connections is not optimized. In this work, two scheduling algorithms for sleep-mode operations based on the IEEE 802.16e are proposed. These schemes guarantee the quality of services (QoSs) of the network connections for an MSS and also minimize the power consumption of the MSS. Simulation results demonstrate that the proposed schemes outperform the traditional approaches.

The integration of cellular and voice over WLAN (VoWLAN) systems recently has attracted considerable interest from both academia and industry. A cellular/VoWLAN dual mode system helps mobile users to access a low cost voice over IP (VoIP) service in a WLAN area and switch to a wide-coverage cellular system without WLANs. However, the system suffers the power consumption problem and the service continuity issue between cellular and WLAN systems. This study presents a network architecture for an enterprise or a dual mode service provider to permit service continuity between cellular and VoWLAN systems. This architecture is used as a basis for further elaborating a novel power saving mechanism, called PIANO (paging via another radio). By applying the proposed mechanisms to a dual mode mobile, the mobile can completely switch off its WLAN interface, only leaving the cellular interface awake to listen to paging signals. When a dual mode mobile receives a paging message from its cellular interface, it wakes up its WLAN interface and responds to the call via WLAN networks. Therefore, a dual mode mobile can reduce the idle mode power consumption, and can also obtain VoWLAN services. The simulation results illustrate that the proposed methods significantly extend the standby hours of a dual mode mobile.

Mobile telecommunication systems such as General Packet Radio Service (GPRS) and Universal Mobile Telecommunications System (UMTS) provide packet-switched service for mobile terminals. 3GPP Release 4 and Release 5 further offer packet voice over a GPRS/UMTS infrastructure so that IP based communications between mobile to mobile is expected to be one of the major traffic types in the future 3G all-IP network. Considering packet communications between mobile to mobile in GPRS/UMTS networks, every packet has to go to the Gateway GPRS Support Node (GGSN) and then can be delivered to the destinations according to the current GPRS/UMTS specifications. However, packet delivery through GGSN is not efficient if two mobile nodes are in the same base station or the same routing area. This design degrades overall system performance and introduces the extra delay of mobile-to-mobile communications. This paper investigates packet routing/forwarding procedures and proposes a novel approach to improve mobile-to-mobile packet delivery in GPRS/UMTS networks. Simulation results demonstrate that end-to-end latency can be reduced significantly in GPRS/UMTS networks.

High tier wireless systems such as GPRS and UMTS provide users high mobility but low bandwidth. On the other hand, low tier wireless systems such as wireless LAN offer higher bandwidth with less mobility. To support seamless roaming between. heterogeneous wireless networks is regarded as one of the key issues in the future mobile communication system. The paper presents method and system for seamless roaming between WLAN and UMTS networks. The two networks are assumed to be constructed and to operate independently, but wish to offer roaming services to their customers. A new logical node, called virtual GPRS support node (VGSN), is designed for this purpose to connect UMTS and WLAN core networks. VGSN is a normal GPRS support node (GSN) in a UMTS system and also acts as an access route in a WLAN network. The node converts control signals and routes data packets between two networks for roaming users. With the presented approach, Mobile IP is not required and users can have seamless roaming between networks based on one IP address.

Differentiation service (Diffserv) is regarded as one of the practical architectures to realize quality of service (QoS) on the Internet. Relative differentiated service, which achieves relative QoS differentiation between traffic classes, is a simple and easily-deployed service model. Based on the concept of relative differentiated service, Dovrolis et al. (ACM SIGMETRICS Performance Evaluation Review vol.27, no.1, pp.204-5, 1999; IEEE Network, September 1999; ACM SIGCOMM-99, September 1999) proposed a proportional differentiation service model which guarantees the ratios of service differences between classes. They claimed that weighted fair queuing (WFQ) is not suitable for implementing relative differentiation service and employed priority-based scheduling algorithms in their model. In this paper, we extend WFQ and apply it to proportional delay differentiation service. The extended WFQ algorithm adjusts the weighting of each class dynamically so that the delay differences between classes can be well controlled. Simulations show that the proposed methods can realize proportional delay differentiation service effectively and efficiently.