Edge AI inference is an important workload in 5G networks. Multiple classes of edge AI workloads often share edge infrastructure, yet each may require distinct latency and rate guarantees. 5G network slicing supports differentiated requirements on the network side, but shared GPU inference also needs compute-side guarantees after requests reach the Multi-access Edge Computing (MEC) host. We extend 3GPP network slicing with compute-side enforcement so that slice guarantees remain effective after traffic reaches the MEC host. To realize this extension, we design a GPU scheduler that combines Hierarchical Token Bucket (HTB)-based traffic conditioning with Earliest Deadline First (EDF) scheduling. Our scheduler enforces per-class assured goodput, defined as the committed rate of latency-compliant completions for each class. The GPU scheduler identifies request classes via tags, which are assigned during GTPU encapsulation at the 5G user plane. This integration preserves overall latency guarantees across both the network and compute domains of a 5G slice for AI.

To support the 5G Wireless and Wireline Conver-gence (5G-WWC) architecture, the Access Gateway Function (AGF) is defined to serve as an interworking function between the 5G core (5GC) and fixed network residential gateways (FN-RG) with a non-3GPP wireline access network. AGF acts as a proxy to communicate with 5GC on behalf of FN-RG. This paper presents a design for AGF that realizes control and user plane separation. Three different implementations based on user-space, kernel-space, and kernel-bypass (DPDK) softwares are reported and tested. The results confirm the superiority of the kernel-bypass approach in terms of latency and throughput of user-plane traffic.

Traffic splitting is to distribute traffic over multiple paths so as to better utilize link bandwidth and avoid potential link congestion. Many traffic splitting schemes have been proposed for datacenter networks, which provide different levels of splitting granularity. This paper proposes a packet-level traffic splitting scheme, which provides the most fine-grained link utilization as it lays out a forwarding path for each individual packet. The proposed scheme mitigates possible packet reordering problem by equalizing the build-ups of packet queues in all switches as the path layout utilizes links in proportional to the associated exploitable link bandwidth. The result is a packet dispatching rule designated to each switch. We implemented the rules using P4 (Programming Protocol-Independent Packet Processors) switches and conducted experiments to measure the performance. Experimental results show that the proposed scheme provides higher goodput compared with another packet-level traffic splitting scheme which dispatches packets randomly among all links.

Network slicing is an essential technology for 5G mobile networks. It partitions network resource logically into multiple isolated slices, each of which can satisfy a suite of network requirements for one specific service. However, it cannot be fulfilled by the current SDN (Software-Defined Networks), since the conventional SDN data-plane technology, OpenFlow, is not flexible enough to offer fine-grained network resource control or queue/packet scheduling. It leads to many research studies developing corresponding solutions on programmable switches. In this work, we focus on the support of the bandwidth guarantee and management for network slices. Although several studies with the similar goal have been proposed, they do not consider interference among different flow types or use the built-in meter for easy deployment on COTS (Commercial Off-The-Shelf) P4 switches. To this end, we first conduct a case study to examine the interference cases. We then propose a solution, designated as P4-TINS (P4-driven Traffic Isolation for Network Slicing), to resolve the interference by isolating different types of traffic flows in priority queues and set the P4 switch’s bucket size based on the time granularity of its bandwidth management operation. It cannot only ensure the guaranteed bandwidth for each slice but also enable coexistent slices to fairly share residual bandwidth. We have confirmed its effectiveness experimentally based on our prototype over an ONOS (Open Network Operating System) controller and a COTS P4 switch.

Software-defined network (SDN) allows traffic engineering and flow management by decoupling the control and data planes. SDN has attracted a lot of attention recently because of its centralized control, flow programmability and flexible resource management. However, to support centralized control, flow programmability and flexible resource management, most of the advanced SDN designs require per-flow management, which would easily make the flow table of a software switch overflow and incur undesired processing overhead when a network scales up. To overcome such huge traffic demands, we propose acluster-based macro-flow managementframework to balance the trade-off between customized services and management costs. The key idea of our design is to group similar flows into a cluster and replace per-flow management with per-cluster management. To achieve this goal, we cluster flows with consideration of the similarity of their traffic patterns. To adapt to varying traffic load, we further propose a hierarchical cluster merging scheme to dynamically merge clusters and make the best tradeoff between flow entry saving and routing performance guarantee. Our evaluation demonstrates that the proposed cluster-based flow management framework can significantly reduce the flow table usage, while producing minimum negative impact on routing performance.

The goal of P4MT is to leverage the role-based control mechanism in P4Runtime Specification to further enable multi-tenancy on a single P4 switch. The evaluation results show that P4MT consumes only a small percentage of ASIC space and causes a negligible increment in data plane/control plane latency. International P4 Experimental Networks (i-P4EN) has been created to realize the potential of P4 experimental networks. To perform multiple experiments for research groups, multitenancy is required to increase the efficiency of the testbed. By modifying the P4Runtime design and P4 pipeline, we designed and implemented multi-tenancy on a P4 switch. With P4MT, each tenant can select its own P4 pipeline and control the packet processing without intervening with the packet processing of other tenants. Future work includes supporting P4 Externs for programmable data plane networking, multi-tenant networking and automatic pipeline migration on P4MT.

To realize P4 language benefits: “Protocol Independent, Target Independent, Field Reconfigurable” [1] in a network research and development environment, a few P4 research institutions around the world formed a partnership to design and implement the international P4 Experimental Networks (Figure 1). The participating institutions can share distributed P4 resources over international research and education networks, as well as initiate international P4 research collaboration projects. Initial network scenarios in this research testbed are Software-Defined Network Exchanges (SDXs), Network and Cloud Testbed Networks, Science Networks, and Campus Research Networks. These four network scenarios require multi-tenant support. Current P4Runtime specifications include a Multi-Controller design for twofold: Control Plane Partition and Redundancy [2]. However, the current P4Runtime does not implement Control Plane partition to support multi-tenant service on a switch. This project modified the current design to support a provider and multiple tenant model for the four international P4 Experimental Networks (i-P4EN)scenarios and other applications.

As the next generation network is supposed to support diverse service requirements, managing Quality of Service (QoS) is a crucial part of it. QoS guarantees have long been deemed too complicated until the emergence of software-defined networking (SDN) and widely adopted standard OpenFlow. Recently, Programming Protocol-independent Packet Processors (P4) has gained much attention because of its features like programmable data plane and independent protocol and platform. It is anticipated that the high flexibility of P4 can enhance the QoS control for production networks. In this paper, we show a design of bandwidth management for QoS with SDN and P4-programmable switch. The design classifies packets into different categories based on their QoS demands and usages, which are then disaggregated by a two-level priority queue. Experiments with P4 switch shows that the proposed design not only effectively limits the maximum allowed rate but also guarantees the minimum bandwidth of each traffic flow. As such, the design can maximize bandwidth utilization and serves a building block for network slicing.

Due to the advance in wireless communication technologies, smartphones and tablets are becoming part of our life. People can use mobile devices to take pictures to share their life with others. In consideration of privacy and usability, this paper focuses on developing a private platform for content sharing in home. To facilitate content management and fast content retrieval, we design a Content-Centric Platform (CCP) that allows users to manage and search contents according to the tags (or attributes) of contents. With CCP, family members can access shared contents by the tags of contents without knowing the exact physical locations of the contents. The CCP platform has an auto-tagging subsystem that can add annotated tags automatically for the shared contents. It then applies a data mining technique to analyze the tags of the contents and creates a Tag Tree to represent the association of contents based on the tags. Furthermore, CCP also adopts a Tag Cloud mechanism to guide users to retrieve contents of interest. The functionalities provided by CCP include device joining and leaving, content information collection and distribution, auto-tagging and tag analysis. Finally, we have implemented a prototype system to verify the effectiveness of the CCP for content sharing in home networks.

Software-defined networking (SDN) has been a promising solution to multicast streaming data due to its scalability and manageability. However, offering a multicast service that spans a large geographical area is still challenging because we still lack a unified multicast scheme that interconnects independent SDN-managed networks and bridges the service between SDN- based networks and the rest of the Internet. This paper proposes a solution that integrates SDN, CORD, and PIM technologies. We provides preliminary performance evaluation results. This work serves as a stepping stone to the ultimate goal of multicast as a service.

All active sessions of an ordinary host will be broken if the host changes its IP address as a result of migrating to a new subnet. Traditional solutions toward this problem either need modifying mobile hosts or create tunnels that cause inefficient triangle routing. SDN-based mobility schemes, on the other hand, focus on handover latency reduction or fast packet redirection after handover. There is no SDN-based approach that handles autonomous IP address changes by mobile hosts. As a remedy, we propose SDN-enabled Session Continuity (SDN-SC), as an SDN-based mobility management mechanism that retains session for hosts roaming across subnets in an SDN network. Particularly, SDN-SC suppresses possible address re-configurations with techniques such as gateway spoofing and DHCP lease renewal for mobile hosts away from home. We studied the performance of SDN-SC and compared it with that of MIPv4 and PMIPv6. The results show that SDN-SC outperforms both MIPv4 and PMIPv6 in terms of handover latency.

As the aging population grows, the elderly care service has become an important part of the service industry in the aging population era. Activity monitoring is one of the most important services in the field of the elderly care service. In this paper, we proposed a wearable solution to provide an activity monitoring service on elders for caregivers. This service monitors restroom activities, such as washing hands, urinating and defecation. In the proposed solution, wireless motion sensors are wore on elder's wrist and waist to measure their body movement. The measured motion data are processed to statistical features and aggregated to cloud servers through gateways. A two-layer hierarchical framework is used for the activity recognition. In the first layer, a preliminary recognition is performed by a supervised Reduced Error Pruning (REP) Tree classifier to detect the transition of the activity. In the second layer, a Variable Order Hidden Markov Model (VOHMM) is proposed to determine the sequence of the activities. The experiment results show that the recognition accuracy is 70 percent. We developed a prototype service App to provide a life log for the recording of the activity sequence. The caregivers can make use of this information to take necessary actions accordingly.

The Internet has evolved greatly in this decade. For economic benefits, the management of Internet networks needs to adapt to mass changes without huge modifications in hardware and software. Especially in campus networks, there are a variety of users and devices with different Quality of Service (QoS) requirements and management policies. Software-defined networking (SDN) decouples the data and control planes. The separated control plan resides on a centralized controller to make networks easier to manage. In this paper, a role-based SDN campus network slicing is proposed by utilizing an authentication controller and the virtualization technology of FlowVisor to divide the campus network into several virtual networks according to the types of users. However, the mapping between the devices' MAC addresses and the users' roles results in a heavy loading problem on FlowVisor. Therefore, a VLAN-based slicing is introduced in this framework to offload the workload of FlowVisor for classifying packets into corresponding slices. Packets are appended with VLAN tags for recognizing types of users to lower the overhead of FlowVisor. Analyses of experimental results show that compared with MAC-based slicing, VLAN-based slicing can reduce the flow setup latency by 14% to 60% depending on the number of devices.

Mobile devices equipped with multiple network interfaces have the potential to increase transmission throughput by exploiting concurrent multipath transmission (CMT). However, packet scheduling for CMT is challenging since the diversity of latencies among transmission paths can easily yield out-of-order packet receptions and cause receiver buffer blocking if the receiver buffer is not large enough. Previous studies proposed several solutions attempting to eliminate out-of-order receptions and receiver buffer blocking. Delay-Aware Packet scheduling (DAPS) is the only one among them that considers delay diversity at the sender side when scheduling packets. However, DAPS assumes quasi-static path delay and thus does not perform well if path delay changes dynamically. In this paper, we analyze how jitter affects the performance of DAPS. Furthermore, we propose and implement a jitter-aware packet scheduler named JAPS. Performance evaluations show that JAPS outperforms existing algorithms in terms of throughput under different settings of data volume, receiver buffer size, network jitter and bandwidth ratio.

Low-latency handover is crucial to real-time traffic in wireless networks. This paper considers an enterprise network managed by emerging Software Defined Network (SDN) technology. For this networking environment, we propose a mobility management scheme named Mobility SDN (M-SDN) that reduces the traffic pause time caused by a host-initiated layer-2 handover. M-SDN performs handover preparation in parallel with the layer-2 handover that involves N-casting of active flows to every potential handover target. Handover preparation is enabled by efficient address resolution and location tracking. We have implemented a prototype of M-SDN and conducted several experiments to evaluate the performance of M-SDN. Experimental results show that M-SDN effectively reduces the impact of layer-2 handovers without any modification on mobile devices.

In most developed and developing countries, the maintenance of roadway pavement consumes considerable resources and represents a major item in government budget. However, due to the lack of sustainable and systematic approaches to implement long-term road pavement monitoring and assessment, the cost-effectiveness of performing such tasks is often a concern. In this paper, a crowdsourcing framework is proposed as a solution based on a smart probe car (SPC) system that utilizes mobile sensing technologies to collect environmental roadway data. Built upon the probe-vehicle system, the proposed method not only has the potential to save a significant amount of resources but also makes it possible to acquire real-time pavement information in a large scale.

Sniffer tools capture protocol data. Kernel-profiling tools track function calls and events occurring in the kernel. These two types of tools help us observe external and internal behaviors of networking protocols, respectively. We need both types of data for a comprehensive view of protocol behavior. However, none of existing tools performs these two tasks in an integrated way. We developed Kernel and Protocol Analysis Tool (KPAT). KPAT injects software probes into Linux kernel to track interested function calls and event occurrences in the kernel. Probe injection is done systematically and does not require recompiling the kernel. A module in KPAT finds the association between the tracked functions and protocol data captured by an independent sniffer. The result as an integrated log allows users to identify two-way relationship between protocol data and the execution sequence of network functions in the kernel. We successfully used KPAT to identity accurate latency of each handover phase in IEEE 802.11 wireless networks. Experimental results show that KPAT causes light overhead to the patched kernel.

The increasing demand for mobile data usage is overwhelming our wireless network. However, most of these devices are equipped with multiple network interfaces (multi-homed) which when properly exploited have the potential to increase bandwidth capacity and ensure reliable connectivity. But some technical challenges are restricting our multimode devices to using one single interface at a time. In this paper, we advocate for the use of multiple interfaces simultaneously, which can achieve reliable network connectivity, allow applications to distribute their traffic over multiple interfaces, and aggregate the capacity of different paths. We proposed an OpenFlow-based Multiple WLAN Interface (OpenMWF) prototype, designed by embedding OpenFlow network resource controller and a data forwarding open virtual switch (OVS) in the network stack of the device operating system. A proof of concept implementation and evaluation was carried out using Web-based Real-Time Communication (WebRTC) interactive video conferencing and multi-source downloading utility (Aria2c) applications use cases. The result showed the approach yielded benefits for applications performance, over single interface usage.

As wireless technologies continue to advance, mobile users have developed a need for ubiquitous accessing to the Internet while roaming among heterogeneous wireless networks. Guaranteeing service continuity requires an effective mechanism to optimally use all available access networks. In order to achieve this goal, applications need to handle transient network outage and react accordingly to retain ongoing services. Applications with such capability require expert programmers to make huge efforts on many details that are totally irrelevant to application logic. In this paper, we propose Heterogeneous Network Communication System (HNCS), which is an Android platform that provides always-best-connection and service continuity for applications. Furthermore, HNCS allows programmers to easily implement applications that are invulnerable to transient network changes. The experiment results show that HNCS meets the goal of maintaining service continuity when roaming among heterogeneous networks.

Connecting TV sets to the Internet gives developers opportunities to build more intelligent smart TV systems. With Internet capability, developers can utilize existing network services to build fancier features. For example, a TV program recommender service can help users find their favorite TV programs. This study proposes a service composition platform for building a smart TV system, which provides an easy way to configure how these various network services and functions of physical devices are integrated and orchestrated. The design considerations, system architecture, and interfaces are also given in detail. This study aims to facilitate developers and users in creating and customizing their composite services without the needing of coding. The prototype demonstration also validates the usability and feasibility of the proposed platform.

This study is motivated by a plan to install a software-defined wireless local area network (SDWLAN) on campus, which possesses a desired property that both data flow and device behaviors can be software-definable. Because the installation involves hundreds of access points, we must conduct simulations beforehand to verify the design and scalability of the target system. However, existing SDN simulator like Mininet does not support modeling of wireless channel and mobility. On the other hand, common network simulator like ns-3 only has limited support for software-defined controllers and does not fully implement handover process. We thus develop OpenNet, which connects Mininet to ns3 to enjoy both Mininet's advantage of controller compatibility and ns-3's ability in the wireless/mobility modeling. OpenNet also complements ns-3 by adding probe mechanism, which is missing in the current ns-3 implementation. Our simulation result demonstrates the effectiveness of OpenNet.

Crossover nodes have been utilized to achieve smooth handoffs for micro mobility management schemes. IEEE 802.11s supports proxy-based frame delivery services for mobile hosts roaming within a wireless mesh network (WMN). Discovering crossover node for mobile hosts in 802.11s WMNs, however, faces several challenges due to intrinsic properties of mesh networks. This paper identifies these challenges and proposes a scheme suitable for crossover node discovery under 802.11s WMNs. This scheme is characterized by source-oriented, MAP-centric, and per-source. It can be performed off-line with the derived results cached for on-line retrievals during handoffs. Simulation results show that the proposed scheme can reduce packet losses due to handoffs, and confirm the necessity of the source-oriented principle.

Churn refers to a large number of arriving and departing participants within a short time in peer-to-peer (P2P) networks. This paper studies the impact of churn on real-world unstructured P2P networks. To this end, we need collecting snapshots of P2P network topology and associated performance metrics. Because P2P topology changes dynamically, the time to take a snapshot must be sufficiently short for the snapshots to be accurate. We propose Third-party-to-servent Crawling with Servent-to-servent Sampling (TCSS) system. TCSS uses a third-party crawling technique to collect network topology information without disturbing the original P2P network under investigation. Furthermore, TCSS adopts distributed and parallel techniques to speed up the crawling process. TCSS also employs a servent-to-servent sampling technique to gather the corresponding performance metrics of the P2P network simultaneously. Empirical results show that TCSS takes around 7 minutes to take a topology snapshot of the P2P network. Besides, we found that churn is indeed a combined effect of peer arrivals/departures and neighbor replacements. As the number of peers increases, the number of very long-lived peers remains nearly constant and the P2P network possesses a small-world property. Moreover, as churn aggravates, the average booting time of peers increases and the variation is proportional to the degree of churn. The response time of the top-rank keyword searches is not affected by the degree of churn.

Mobile devices equipped with two or more interfaces can roam across heterogeneous networks. This feature enables the devices to maintain service continuity by performing vertical handovers and to increase transmission capability by exploiting multiple physical transmission paths. However, this type of service continuity is not yet commonly provided. Previous approaches toward the provision of service continuity are limited either in their functionalities or in the scope of cross-layer information. In this paper, we analyze how cross-layer architecture provides service continuity from a broader perspective and propose Cooperative Layered Architecture (CoLA) implemented on Android system for service continuity during handover process. CoLA helps comprehensive information gathering from all layers for handover decisions, provides simple interfaces to make applications mobility-sensitive, and achieves efficient multipath transmissions across heterogeneous wireless links. Experimental results show that CoLA achieves its design goal in providing service continuity.

In mobile opportunistic networks, security is a major concern in consequence of store-carry-and-forward data dissemination. To guarantee that messages can be only accessed by authorized users, access control is a necessity. Fuzzy-IBE that has been proposed can provide group access control. In addition, if a sender wants to expose his/her message only to some members of his/her group, Fuzzy-IBE can provide a whitelist mechanism to further indicate which users in the group can access the data. Because overhead linearly increases with respect to the length of the whitelist, this mechanism is not sufficiently scalable as the whitelist becomes larger. In this work, we propose a Fuzzy-IBE based negative access control scheme (NAC) that allows users to selectively exclude specific members from accessing data dynamically. NAC, which does not require intensive calculation in mobile devices, is suitable for mobile opportunistic networks.

A NEtwork that is MObile (NEMO) usually consists of at least one Mobile Router (MR) attached to the infrastructure to manage all external communication for this moving network. Users inside a NEMO might also require quality of services (QoS) guarantee while accessing the Internet via a mobile device. In this paper, we propose a Bandwidth-sharing Reservation (BSR) scheme so that a NEMO can support QoS services. The proposed BSR scheme establishes a bandwidth sharing reservation tunnel for all NEMO nodes to efficiently utilize external bandwidth of MR, and it makes advance reservations at neighboring networks to solve the mobility problem. Three reservation adjustment policies are designed to further improve bandwidth utilization. Both mathematical analysis and simulation results show that the proposed BSR scheme can significantly increase bandwidth utilization and decrease handoff blocking rate while providing QoS services to NEMO nodes in movement.

In this paper, we propose a QoS-aware power-strength classified packet scheduling (PCPS) scheme that can classify the transmission power-strength of a VBR connection into several classes and determine the admitted service rate of each class at connection setup time. During the transmission time of each connection, PCPS scheme can dynamically adjust the downlink service rate of each connection in accordance with the admitted service rate of the current transmission power-strength class while retaining all connection's frame error rate requirements. With this power-strength classified scheduling, PCPS scheme can decrease the average packet delay and increase the bandwidth utilization. Furthermore, a two-stage scheduling architecture is also employed to reduce the scheduling complexity resulted from simultaneous adjustment of both service rate and transmission power-strength in WCDMA. Performance results show that the delay bound and the packet loss ratio of a VBR connection can be guaranteed.

Due to its flexibility, mobile ad hoc network (MANET) has attracted a lot of attention recently. Most existing works, however, limit a MANET as a stand-alone network. In this paper, we propose a two-tier MANET by extending the connectivity of the MANET to the Internet. Some hosts in the MANET are equipped with cellular interfaces and are called gateways, which can provide Internet connections. Such extension would greatly improve the connectivity of MANET. To maintain seamless connectivity, we modify the mobile IP to make traversing private networks (i.e., NAT) of cellular networks possible. We also propose a load-balancing routing protocol to utilize the cellular interfaces of gateways, which are likely to be bottlenecks to the Internet. A prototype is implemented and performance results obtained from the testbed is reported.

We present a traffic-scheduling framework that can dynamically allocate radio resources to a general packet radio service (GPRS) mobile station (MS) based on the interference levels of the radio links and the quality of service (QoS) specification of the MS. The underlying idea of this scheduling scheme is to preserve more bandwidth for use by those MSs that are within a low interference region so that the limited radio resources can be used more effectively. In this scheme, an MS uses a low transmission rate for data transfer when the MS is within a high interference region to avoid wasting bandwidth by transmitting data in a condition with high interference. In order to compensate for the service loss of the MS, we allocate more bandwidth to the MS when it is within a low interference region. In addition, we also propose an analytical model that can be used to derive the transmission rate for an MS in a low interference region based on the delay-bound requirement of the MS. The performance results show that our dynamic scheme can utilize the bandwidth more effectively to satisfy various QoS requirements of the MSs in the GPRS system without changing the convolution-coding rate.

Abstract Owing to the increasing population of mobile subscribers, the rapidly expanding signaling traffic has become a challenge to the mobility management in PCS networks. Multiple database schemes to reduce signal traffic and to solve the bottleneck problem of the singlehome location register(HLR) architecture have been proposed by many researchers. However, in most of the multiple location databases or HLR systems, extra signaling is required for the multiple database updates. We propose a batch‐update strategy, instead of the immediate update method, for the location‐tracking schemes with replication to reduce the signaling overhead. In this paper, we first introduce a distributed HLRs architecture in which each HLR is associated with a localized set of VLRs and the location registrations and queries are processed locally. Then we propose our batch‐update strategy and present two pointing schemes for inter‐HLR call deliveries. The numerical result shows that our approach can effectively decrease the signaling cost of location registration and call delivery compared with the IS‐41 standard. Copyright © 2002 John Wiley & Sons, Ltd.

We first compare the performance of the direction-based location update (DBLU) schemes with that of the distance-based location update scheme under various mobility behaviors of the mobile hosts (MTs). The comparison results show that under the random walk model, the base DBLU scheme outperforms the distance-based location update scheme when the mobility to call ratios (MCRs) of the MTs are low. However, it has been shown that the MTs are likely to possess forward-prone movement characteristics in microcellular architectures. If the MTs are indeed prone to make forward movements, the base DBLU scheme will be better than the distance-based location update scheme in all the MCRs according to our analysis. Therefore we further propose two enhanced DBLU schemes, namely the forward direction-based location update (FDBLU) and the restricted forward direction-based location update (RFDBLU) schemes, for the location tracking of the MTs in micro-cellular architectures. Analytical results also show that both FDBLU and RFDBLU schemes outperform the base DBLU scheme, and RFDBLU scheme is the best among the direction-based location update schemes.

A mobile computing environment allows hosts to roam while retaining access to the Internet. Multicasting is one of the most important facilities for constructing reliable distributed systems and cooperative applications. Host mobility, however, challenges multicasting in this environment: the established multicast delivery paths may frequently restructure along with host migrations, incurring expensive overheads. This paper presents a framework for network-layer multicasting while keeping a low overhead in adapting multicast routes to mobile host locations. This is achieved by partitioning the mobile environment into non-overlapping regions, so that changes in the multicast routes due to host intra-region movements are hidden from other regions. An analytical model was developed for performance evaluation. It shows that, compared with the best known proposal, our scheme reduces the average multicast latency by more than 66%, while causing less than 7% overhead.

This paper presents performance analysis of two general approaches to IP multicasting in mobile environments. In the first approach, copies of each multicast datagram were forwarded to Mobility Agents in all wireless LANs on a campus. The other approach designated the Mobility Agent as a proxy (fixed host) for its local mobile hosts to participate in communication group, thus allowing current multicast protocols on the wired backbone to be easily applied. We develop an analytical model for quantifying their performance metrics. Performance is measured in terms of the latency required for group members, fixed and mobile hosts, to receive a datagram addressed to them, network bandwidth required to accomplish a single multicast. Numerical results show that each approach can outperform the other in certain circumstances.

One of the main challenges in personal communication service (PCS) is to locate many mobiles that may move frequently from place to place. Such a system operation is called location tracking. Many network signals flow, and database queries are required to achieve such a task. In addition to the two-level hierarchical strategy in IS-41 and GSM, several strategies have been proposed to improve the efficiency of location tracking. Pointer forwarding was used to reduce the expensive home location register (HLR) accesses. Previously, the distributed HLR scheme was proposed to prevent the HLR from becoming a bottleneck in the signaling network. However, the length of a forwarding pointer chain may be lengthened in a distributed HLR environment. We propose a more efficient strategy to overcome this potential problem. This strategy attempts to migrate the locating chains in a distributed HLR system when a mobile issues a registration operation. As a consequence, the length of any forwarding pointer chain does not exceed one in our strategy. Simulation results indicate that our strategy significantly decreases the locating cost. In fact, this strategy provides an upper bound of location tracking time owing to the fact that the length of any locating path does not exceed one. Furthermore, obsolete entries in local databases visitor location registers can be reclaimed in this strategy.

This paper presents an integrated compiler, runtime control, and hardware solution to improve inter-thread communication and synchronization in a multithreaded processor architecture. Multithreading improves processor utilization by exploiting more parallelism. The improvement in utilization, however, is hindered by interthread communication and synchronization problems, which incur extra communication overhead and thus degrade the performance of the system. In this paper, we propose efficient inter-thread communication and synchronization schemes based on a superscalar DLX processor with multithreading functionality. Compiler, runtime control, and hardware support used in the schemes are discussed. Simulations are presented to show the effectiveness of the proposed schemes.