Extending multivariate time series forecasting to resource-constrained edge devices is essential for enabling intelligent and sustainable IoT services. A common scenario involves vertically partitioned data across devices, where each device must forecast its own variables while benefiting from others’ information. This paper studies a resource-efficient solution for this scenario based on vertical federated learning (VFL). Prior VFL frameworks are designed for situations where only one party holds the labels and would struggle to meet the demand of the targeted scenario, as storage resources usage would increase dramatically with the number of devices. Going beyond VFL, we design multivariate vertical federated learning (MVFL) as a novel federated learning framework, where we separate communication features and local features in an embedded feature space. This design enables MVFL to utilize storage and communication resources more efficiently by eliminating redundant models. On four real-world benchmarks, MVFL outperforms the VFL approach in both efficiency and accuracy, demonstrating its practical value for distributed IoT systems.

In this paper, we propose LUVI, Lightweight UWB-VIO relative positioning method for indoor localization. Recent designs of handheld and embedded devices feature various technologies which have the means to enhance localization performance in indoor environments. These include visual odometry based on cameras and augmented reality, and communication hardware such as UWB. Integration of such technologies to exploit their advantages allows us to compensate for each other's errors in measurement. This improves the overall function of future services, such as visual representation of sensing information from sensors in areas that are not physically visible. However, existing work cannot fully exploit these technologies to high extent, often inducing high errors or wasted resources. LUVI is a novel localization method which estimates the location of a target object using relative coordinates of estimator devices without the aid of definitive coordinates. LUVI focuses on utilization of lightweight management of virtual anchors for localization, with functions that reduce the computing and communication complexity while maintaining the accuracy and improving energy efficiency of the localization. Our work has been fully implemented and tested in several indoor environments, showing robustness to NLOS while significantly reducing computational complexity, and up to 30% lower average error.

In this poster, we propose a sensor localization system assisted by wireless communication and augmented reality (AR) suitable for disaster relief applications. Generally, disaster environments are considered extremely hazardous and deteriorated, with unpredictable effects to human mobility and digital devices. To maximize the safety and efficiency of first responders, deployment of wireless sensors are of utmost importance, as sensor nodes can provide sensing information as well as location information. We analyze the issues and challenges that need to be tackled for high accuracy localization of sensor nodes in such environments, and then propose a system that we plan to develop in the near future.

Conjunction of USB applications with wireless technologies leads us to USB over Wi-Fi through the means of Wireless Serial Extension (WSE). However, USB and Wi-Fi has distinct data prioritization methods, which causes problems in existing QoS protocols when the two technologies are used together. To solve this problem, this poster proposes and evaluates the dynamic QoS adaptation scheme, which intelligently maps the data of USB with the weighted transmission queues of enhanced distributed channel access (EDCA) mechanism.

Topology control offers many advantages for wireless networks such as reduced energy cost, simplified communication graph and network-wide connectivity. There are mainly two methods for managing the network topology. In clustering, a hierarchy of backbone nodes is selected to improve on the systems scalability and network lifetime whereas in power control each node adjusts its transmission power to achieve certain desirable properties of the resulting topology. While the focus of the former approach is to find an ideal number of the backbone nodes, the main objective of the later approach is to estimate a minimal power level which can solve multi-objective design problem. There have been several topology control protocols proposed however they miss the potential of combining both approaches. In this paper we propose Two-Tiered Topology Control (TTTC) protocol, a generic framework which combines the clustering and power control approach towards topology control and evaluate different conditions under which it performs well. TTTC operation is divided into two phases. During the first phase, a parameterized clustering algorithm is executed to obtain clusters of varying properties. At the end of first phase, the network is organized into two tiers. The backhaul-tier consists of cluster-head nodes while the connectivity-tier contains the cluster-members. In the second phase, each cluster-head runs a local MST-based power control algorithm. Simulation results show that proposed framework achieves efficient trade-off in terms of energy cost, neighbor count and hop count while maintaining fully connected network. Moreover, the framework relies on local available information with lower communication overhead.

This poster proposes an adaptive flow control scheme for reliable packet delivery of Universal Serial Bus (USB) services when they are transmitted through wireless medium. In the future, USB can be utilized with the Wi-Fi technology to provide Wi-Fi USB (WSB). However, the data transmission of Wi-Fi and USB are asynchronous due to their different Medium Access Control (MAC) characteristics, which may result in buffer overflow. To prevent buffer overflow, we propose an adaptive flow control scheme that can select between the various transmission speeds of USB to match with the bandwidth provided by the Wi-Fi layer. Preliminary results of the proposed scheme evaluated via the NS-3 shows that the packet delivery ratio of USB data can be substantially increased, thus providing better USB reliability.

The capability to interconnect directly with neighboring wireless devices coupled with improvements in high-speed wireless connections, and the wide distribution of high-quality multimedia has led to the design of standards such as the WiFi-based Miracast [1], which allows handheld mobile devices to share their screen contents with larger-sized display devices (e.g., smart TVs). Such screen sharing capabilities allow various multimedia files to be easily accessed through mobile platforms, and played (in real-time) through larger screens; thus, has the potential to enable a variety of attractive entertainment applications. While widely available on all Android-based mobile devices (4.2 or recent), the Miracast standards introduce a significant level of inefficiency as it deals with different types of multimedia contents.

Research presented in this paper enhances the power management schemes of Wi-Fi Direct to provide better services for a wider range of Wi-Fi Peer-to-peer (P2P) applications. Although the Wi-Fi Direct defines two power management schemes to provide energy efficiency in P2P devices, neither is enough to meet power efficiency and reliability requirements for various services. To alleviate this problem, the proposed scheme recognizes the properties of applications and dynamically adjusts the duty cycle of P2P devices. The proposed scheme is compared with the schemes in Wi-Fi Direct to analyse its effectiveness.

This paper focuses on enhancing IEEE 802.11e EDCA for improving reliability in Smart grid networks. 802.11e can support prioritized QoS to differentiate traffic and guarantee fairness for delay-stringent applications. However, this may pose serious problems when applied to Smart Grid environments which support applications that have strict reliability requirements. To solve this problem, we suggest a contention mitigated EDCA algorithm (CM-EDCA), which utilizes packet collision and MAC retransmission parameters to change the packet priority and reduce packet collisions. The proposed scheme is compatible with IEEE 802.11e and easy to implement for Smart grid devices. Simulation studies show improved reliability compared to the existing methods in IEEE 802.11e.

We propose and demonstrate an adaptive wireless mesh network architecture for public surveillance and monitoring. The proposed architecture is based on IEEE 802.11s WLAN mesh network, which is enhanced to provide adaptation depending on the environment of the network. We first introduce the implementation architecture of the adaptive mesh, and then describe the demonstration of the network on the wireless mesh camera sensor testbed.

In video surveillance and monitoring systems based on wireless mesh networks, large volumes of multimedia data with different priorities and time demands are generated by video cameras and wireless sensors. Hence, one of the key issues is how to alleviate network congestion for highly reliable transfer of multimedia data towards gateways. This paper proposes a solution for utilizing multiple gateways to divert data streams and alleviate traffic congestion near a specific gateway. With a new routing protocol, named “Multi-Gateway Routing with Congestion Avoidance (MGR-CA)”, network congestion can be predicted in a distributed manner and amounts of data traffic transmitted to the congested path are redirected to an alternative gateway based on the severity of congestion. Preliminary testbed experiments are made to show the performance of our scheme.

Vehicular sensor networks (VSN) has recently become a hot research topic in the field of wireless networking, as this promising technology has potentials to provide variety of applications and safety to drivers on the road. As promising it is, VSN also introduces many new challenges to the table. For example, the vehicle mobility creates a very unique topology with frequent network disconnections, which may pose problems when existing network protocols for general ad hoc networks are applied. Majority of the research related with VSN limit their data harvesting schemes to only single-hop or short distance transmissions. Therefore these works cannot meet real-time requirements nor support applications where data must be transmitted in multi-hop. The authors believe such problems can be alleviated by our proposed scheme, named ‘multi-hop data harvesting’ (MDH). The simulation studies conducted using QualNet 4.0 simulator shows the effectiveness of the proposed scheme.

In a wireless sensor network application for tracking multiple mobile targets, large amounts of sensing data can be generated by a number of sensors. These data must be controlled with efficient data aggregation techniques to reduce data transmission to the sink node. Several clustering methods were used previously to aggregate the large amounts of data produced from sensors in target tracking applications. However, such clustering based data aggregation algorithms show effectiveness only in restricted type of sensing scenarios, while posing great problems when trying to adapt to various environment changes. To alleviate the problems of existing clustering algorithms, we propose a hybrid clustering based data aggregation scheme. The proposed scheme can adaptively choose a suitable clustering technique depending on the status of the network, increasing the data aggregation efficiency as well as energy consumption and successful data transmission ratio. Performance evaluation via simulation has been made to show the effectiveness of the proposed scheme.

This paper proposes a novel multi-hop data dissemination method using data replication technique to collect sensor data in vehicular sensor networks (VSN). Currently, existing network protocols for general sensor networks pose problems when used in VSN applications. Even the majority of research based on VSN network protocols has some problems, as they limit their data harvesting schemes to only single-hop or short distance transmissions, or use high-cost relay stations to support multi-hop cases. As a result, these works cannot completely solve some critical issues in VSN, such as data transmission delay, packet overhead, and network connectivity. We believe such problems can be alleviated by our scheme, which uses replicas in a multi-hop data dissemination scheme. Our simulation study conducted using the QualNet 4.0 simulator shows the effectiveness of the proposed scheme.