Connected car has become one of the fastest growing segments in the IoT market. A key feature of connected car solutions is personalization. An advanced solution shall be able to enable complete personalization for each seat automatically. This paper presents the design and implementation of BlueID - a framework for robust in-car localization and on-demand personalization using Bluetooth Low Energy (BLE) beacons. BlueID identifies a user based on the user's mobile device and characterizes each seat using signals received from a pre-installed BLE beacon array. BlueID features a machine learning engine that can robustly distinguish seats in the presence of various environment and system dynamics such as varied antenna sensitivities and multiple seat occupants. To enable on-demand use cases, BlueID employs a connected car cloud platform to store each vehicle's seat prediction model and each user's personalization profile. The prediction model and the user's personalization profile are retrieved to personalize the environment where the user sits when the user enters the vehicle. We have implemented a prototype of BlueID and evaluated it through extensive experiments. The results show that BlueID achieves satisfactory in-car localization performance with average position accuracy over 98%.

Based on damage modes of container cranes under earthquake and the characteristic of over-span flexible structures, the anti-seismic design of such machines was investigated. Taking the first crane equipped with isolating device as an example, through dynamic analysis and simulating to anti-seismic crane flexible system under the earthquake and validating analysis and results though the earthquake table tests, efforts was made to research for suitable anti-seismic device and design method for port cranes. The calculation and analysis in the present study could be used as a new way for earthquake resistant design of large-scale flexible portal machine.

Many research activities on wireless sensor networks (WSNs) need detailed performance statistics about protocols, systems, and applications; however, current simulation tools and testbeds lack mechanisms to report these statistics realistically and conveniently. To address this need, we have developed a WSN emulator, VMNet. VMNet emulates networked sensor nodes at the level of CPU clock cycles and executes the binary code of real applications directly. It emulates the radio channel with loss and noise as well as emulates the peripherals in sufficient detail. Moreover, VMNet takes parameter values from the real world and logs detailed runtime information of emulated nodes. Consequently, the application performance, both in response time and in power consumption, is reported realistically in VMNet, as demonstrated by our comparison studies with real sensor networks

Peer-to-peer networks are widely criticized for their inefficient flooding search mechanism. Distributed hash table (DHT) algorithms have been proposed to improve the search efficiency by mapping the index of a file to a unique peer based on predefined hash functions. However, the tight coupling between indices and hosting peers incurs high maintenance cost in a highly dynamic network. To properly balance the tradeoff between the costs of indexing and searching, we propose the distributed caching and adaptive search (DiCAS) algorithm, where indices are passively cached in a group of peers based on a predefined hash function. Guided by the same function, adaptive search selectively forwards queries to "matched" peers with a high probability of caching the desired indices. The search cost is reduced due to shrunk searching space. Different from the DHT solutions, distributed caching loosely maps the index of a file to a group of peers in a passive fashion, which saves the cost of updating indices. Our simulation study shows that the DiCAS protocol can significantly reduce the network search traffic with the help of small cache space contributed by each individual peer

Although the original intent of the peer-to-peer (P2P) concept is to treat each participant equally, heterogeneity widely exists in deployed P2P networks. In this paper, we suggest to improve the search efficiency of P2P network by utilizing the query answering heterogeneity. Our proposed differentiated search (DiffSearch) algorithm can evolve an unstructured P2P network to a two-tier hierarchical structure, where peers with high query answering capabilities are grouped in the first tier, which has higher priority to be queried than the second tier. Because the query answering capability is extremely unbalanced among peers, a high query success ratio can be achieved by querying only a small portion of the network. The search traffic is dramatically reduced due to the shrunken search space. Our trace analysis and simulation show that the DiffSearch algorithm can save up to 60% of search traffic.

Multi-source content distribution and file download systems such as BitTorrent and Edonkey emerge as a better scheme than the traditional ftp systems in distributing files to very large scale systems. However, a major drawback of those systems is that they largely rely on a tracker server to maintain state information of a horde. Furthermore, if the peer holding all the pieces of a file leaves the system, there is a high probability that the other peers cannot download the entire file. We propose a scalable and highly available P2P multi-source content distribution system called SODON to solve those problems. SODON distributes all the pieces of a file from a seed to a number of carefully selected sub-seeds in order to guarantee a complete set of pieces that are available for a long time. The sub-seeds also work as sub-trackers to maintain the horde state information. As a result, the original tracker server is not overwhelmed by a large population of download requests. Our traces on BitTorrent and simulation result show that SODON is feasible to build, and the sub-seeding and sub-tracking schemes effectively reduce the tracker load while offering a comparatively high availability of shared files

To improve the scalability of Gnutella-like unstructured peer-to-peer (P2P) networks, a uniform index caching (UIC) mechanism was suggested in some earlier work. In UIC, query results are cached in all peers along the inverse query path such that the same query of other peers can be replied from their nearby-cached results. However, our experiments show that the UIC method causes a large amount of duplicated and unnecessary caching of items among neighboring peers. Aiming at improving the search efficiency, we propose a distributed caching mechanism, which distributes the cache results among neighboring peers. Furthermore, based on the distributed caching mechanism, an adaptive search approach is built which selectively forwards the query to the peers with a high probability of providing the desired cache results. All the enhancements above are defined in a protocol called distributed caching and adaptive search (DiCAS). In the DiCAS enhanced Gnutella network, all the peers are logically divided into multiple layers, with the character that all the peers in the same layer have the same group ID. The query flooding is restricted in one layer with the matched group ID. Our simulation study shows that, with the help of the index caching and search space division, the DiCAS protocol can significantly reduce the network search traffic in unstructured P2P systems without degrading query success rate.

Network research generally requires a simulation or emulation environment to test protocol implementations, to evaluate the performance of a scheme or a system, and to study complicated and highly varying network operations. For large network simulation, simulators consume a large amount of time and memory; and its result is largely based on some modeling assumptions that may not hold in the real world. Emulators are difficult to scale for large network emulation because of the high cost of equipment if a one-to-one mapping scheme is employed. Otherwise, the target network has to be abstracted to a single router modeled with some performance metrics. We present a distributed IP network emulator cluster EMPOWER, which not only can be used to emulate a large network with a limited number of commodity computers, but also can generate user-defined arbitrary network conditions and traffic dynamics at packet level for specific test scenarios. EMPOWER is highly scalable in that each emulator node could be configured to emulate multiple network nodes, and the increment of the number of emulator nodes does not affect emulation validity. Some significant research issues such as network mapping and mobile wireless network emulation are discussed and addressed. Preliminary emulation results show that EMPOWER is capable of assisting the study of both wireline and wireless network protocols and applications.

Network emulation offers real-time simulation that enables test and evaluation of real network systems, protocols, and applications in a reconfigurable and controllable hardware and software environment. Real network traffic is processed by the protocol stacks of physical systems across the entire emulation environment. Standalone emulators can only be used to test end-to-end protocols and applications; simple emulator testbeds are non-scalable due to the one-to-one mapping scheme. We present a distributed emulator cluster that can faithfully emulate large-scale wide area networks with only a considerably smaller number of workstations. In the emulator cluster, a network topology is accurately mapped to a virtual topology in which certain number of virtual nodes are properly configured and emulated. We have developed an algorithm to partition a target network such that each partition can be emulated by a physical node. The emulator cluster can emulate a variety of network conditions and traffic dynamics, as well as incorporating new protocols and models. Preliminary test results show the high scalability of emulator cluster in terms of emulation validity and accuracy with a single physical node and in the entire environment.

The increasing need of protocol development environments and network performance evaluation tools gives rise to the research of flexible, scalable, and accurate network emulators. The desired network emulator should be able to facilitate the emulation of either wireline or wireless networks. In the case when network topology is critical to the underlying network protocol, the emulator should provide specific mechanisms to emulate network topology. In this paper, we present a distributed network emulation system EMPOWER, which not only can fulfill those requirements, but also can generate user-defined network conditions and traffic dynamics at packet level. EMPOWER is highly scalable in that each emulator node could be configured to emulate multiple network nodes. Some significant research issues such as topology mapping scheme and scalability of the emulator are discussed and addressed. Preliminary emulation results show that EMPOWER is capable of assisting the study of both wireless and wireline network protocols and applications.

Differentiated Services (DiffServ) architecture has been proposed as a scheme to provide scalable end-to-end QoS by employing a simple classification based on the DSCP in IP header. One of the major reasons that hinder ISPs to provide DiffServ is that the service level cannot be guaranteed when a specific traffic flow traverses a number of DS and non-DS domains. The issue of the impact of non-DS domains or non-DS routers in a hybrid multi-domain DiffServ network on the performance of Premium Services and Assured Services has been largely ignored by the research community. In this paper, we first present a preliminary queueing analysis of the hybrid multidomain DiffServ network based on a tandem queue model. Then we propose a performance metric called SPI (Service Provision Index) to measure the service level degradation of preferred services in a hybrid multi-domain DiffServ network with respect to an All-DS network. Network simulations and emulations are conducted on a variety of scenarios with different parameter settings to explore the performance of a hybrid multi-domain DiffServ network. Based on the quantitative analysis, the simulation and emulation results, we make some observations on the underlying problem, which could help ISPs or researchers determine the major factors of service degradation, as well as possible directions to improve the services.

Network emulation systems are widely used to explore the behavior of network protocols and to test and evaluate protocol implementations and applications. The major problem of a network emulation system is its scalability in terms of emulation capacity and emulation capability of specific network parameters such as maximum emulated bandwidth and emulated packet delay of the system. In particular, an emulator with a large number of workstations is generally too costly for researchers to afford. We present our experience in building a scalable distributed network emulation system named "EMPOWER". EMPOWER provides the unique feature of precisely emulating multiple nodes with a single workstation, making it possible to support large network emulation with a limited number of commodity computers. We describe some critical design issues such as the system resource competition within an emulator node and its impact on the emulation of network throughput and packet delay. We present our methods to determine the maximum number of virtual routers an emulator node can generate and some techniques to improve maximum throughput and packet delay accuracy of an emulator node. We believe such experiences are valuable for the study, design and implementation, performance tuning of a variety of systems such as network emulators and high performance host-based routers.

The development and implementation of new network protocols and applications need accurate, scalable, reconfigurable, and inexpensive tools for debugging, testing, performance tuning and evaluation purposes. Network emulation provides a fully controllable laboratory network environment in which protocols and applications can be evaluated against predefined network conditions and traffic dynamics. In this paper, we present a new framework of network emulation EMPOWER. EMPOWER is capable of generating a decent network model based on the information of an emulated network, and then mapping the model to an emulation configuration in the EMPOWER laboratory network environment. It is highly scalable not only because the number of emulator nodes may be increased without significantly increasing the emulation time or worrying about parallel simulation, but also because the network mapping scheme allows flexible ports aggregation and derivation. By dynamically configuring a virtual device, effects such as link bandwidth, packet delay, packet loss rate, and out-of-order delivery, can be emulated.