Skip to main content

Wei Koong Chai

Followers

0

Following

1

Public Views

Interests

Uploads

Papers by Wei Koong Chai

Matching Supply and Demand in Online Parking Reservation Platforms

IEEE Transactions on Intelligent Transportation Systems, Mar 1, 2023

Our work concerns online parking reservation platforms proposed in the last decade to cope with t... more Our work concerns online parking reservation platforms proposed in the last decade to cope with the parking challenge in cities worldwide. Enlisting parking resources from commercial operators (e.g., lots) and individuals (e.g., doorways) and letting drivers make online reservations through mobile apps, the goal of those platforms is to ease transactions between the two sides and best match parking supply with parking demand. This way they maximize their value for drivers and parking space providers but also their revenue out of charged commissions. We distinguish between two types of commissions these platforms typically charge, fixed per transaction and proportional to its value, and formulate the respective optimization problems for the platform revenue maximization. We show that the two problems are NP-hard and design a novel algorithm that can treat both, combining greedy and dynamic programming principles. We study its optimality properties both analytically and experimentally, showing that the algorithm closely tracks optimal solutions for small and moderate problem sizes at run times that are several orders of size smaller than off-the-shelf ILP solvers. We then analyze real parking data we collected for the period 2018-2020 from the Bournemouth city in UK to realistically model the rich spatiotemporal dynamics of parking demand such as the location, start times and duration of parking events. These data drive the experimental evaluation of the proposed algorithm, which reports gains of up to 35% compared to the de facto reservation policy in use in such platforms. Notably, these gains are higher when the platform operates under constrained supply conditions.

The Sequence-to-Sequence Architecture with An Embedded Module for Long-Term Traffic Speed Forecasting with Missing Data

Traffic forecasting plays a crucial role in Intelligent Transportation Systems (ITSs), which is p... more Traffic forecasting plays a crucial role in Intelligent Transportation Systems (ITSs), which is proposed to provide traffic status in advance for road users to avoid traffic congestion or other traffic incidents and for authorities to optimise the strategies of traffic management. In this paper, we develop a novel deep learning framework, based on the Sequence-to-Sequence architecture with an embedded module, for long-term traffic speed forecasting with missing data and providing high forecasting accuracy. The embedded module uses Graph Convolution Neural Network for the local spatial dependency analysis by conducting convolutional operation on the k -hop neighbourhood matrix, while utilises Transformer for the global spatial dependency analysis by implementing the attention mechanism that assigns individual weights to neighbour detectors for contributing to the targeted detector. The sequence-to-sequence architecture is built to analyse temporal dependencies of the spatially-fused time series from the embedded module. To evaluate the proposed model against existing well-known ones, the real traffic speed dataset with missing data and frequent traffic incidents is used to train and test the models. The experimental results indicate that our proposed framework achieves the most accuracy forecasting, even obtaining more than 80% accuracy for forecasting two hours in advance.

A joint temporal-spatial ensemble model for short-term traffic prediction

Neurocomputing, Oct 1, 2021

A dynamic spatial–temporal deep learning framework for traffic speed prediction on large-scale road networks

Expert Systems With Applications, Jun 1, 2022

Traffic prediction plays a crucial role in an intelligent transportation system (ITS) for enablin... more Traffic prediction plays a crucial role in an intelligent transportation system (ITS) for enabling advanced transportation management and services. In this paper, we address the problem of multi-step traffic speed prediction, including both short-and long-term predictions. We assert that it is important to consider not just the fixed spatial dependency of the road network (i.e., the connections between road segments) but also the dynamic spatial dependency of traffic within the static topology that intertwines with the temporal evolution of traffic condition across the entire network. We propose a novel deep learning model, named Self-Attention Graph Convolutional Network with Spatial, Sub-spatial and Temporal blocks (SAGCN-SST) model, that specifically capture such complex dynamic spatial-temporal processes. In SAGCN-SST, we integrate self-attention mechanism into graph convolutional networks in a novel framework design while using a sequence-to-sequence model in an encoderdecoder architecture for extracting long-temporal dependency of traffic speed. Two real-world datasets with frequent traffic congestion and accidents from large-scale road networks (i.e., Seattle and Los Angeles) are used to train and test our model. Our experiment results indicate that the proposed deep learning model consistently achieves the most accurate predictions (higher than 98% accuracy on both datasets for the short-and long-term predictions) when compared against well-known existing models in recent literature. The results also indicate that SAGCN-SST is robust against emergent traffic situations.

Path-Based Epidemic Spreading in Networks

IEEE ACM Transactions on Networking, Feb 1, 2017

Conventional epidemic models assume omnidirectional contact-based infection. This strongly associ... more Conventional epidemic models assume omnidirectional contact-based infection. This strongly associates the epidemic spreading process with node degrees. The role of the infection transmission medium is often neglected. In realworld networks, however, the infectious agent as the physical contagion medium usually flows from one node to another via specific directed routes (i.e., path-based infection). Here, we use continuous-time Markov chain analysis to model the influence of the infectious agent and routing paths on the spreading behavior by taking into account the state transitions of each node individually, rather than the mean aggregated behavior of all nodes. By applying a mean field approximation, the analysis complexity of the path-based infection mechanics is reduced from exponential to polynomial. We show that the structure of the topology plays a secondary role in determining the size of the epidemic. Instead, it is the routing algorithm and traffic intensity that determine the survivability and the steady-state of the epidemic. We define an infection characterization matrix that encodes both the routing and traffic information. Based on this, we derive the critical pathbased epidemic threshold below which the epidemic will die off, as well as conditional bounds of this threshold which network operators may use to promote/suppress path-based spreading in their networks. Finally, besides artificially generated random and scale-free graphs, we also use real-world networks and traffic, as case studies, in order to compare the behaviors of contactand path-based epidemics. Our results further corroborate the recent empirical observations that epidemics in communication networks are highly persistent.

Cache “Less for More” in Information-Centric Networks

Lecture Notes in Computer Science, 2012

Ubiquitous in-network caching is one of the key aspects of information-centric networking (ICN) w... more Ubiquitous in-network caching is one of the key aspects of information-centric networking (ICN) which has recently received widespread research interest. In one of the key relevant proposals known as Networking Named Content (NNC), the premise is that leveraging in-network caching to store content in every node it traverses along the delivery path can enhance content delivery. We question such indiscriminate universal caching strategy and investigate whether caching less can actually achieve more. Specifically, we investigate if caching only in a subset of node(s) along the content delivery path can achieve better performance in terms of cache and server hit rates. In this paper, we first study the behavior of NNC's ubiquitous caching and observe that even naïve random caching at one intermediate node within the delivery path can achieve similar and, under certain conditions, even better caching gain. We propose a centrality-based caching algorithm by exploiting the concept of (ego network) betweenness centrality to improve the caching gain and eliminate the uncertainty in the performance of the simplistic random caching strategy. Our results suggest that our solution can consistently achieve better gain across both synthetic and real network topologies that have different structural properties.

A dynamic spatial–temporal deep learning framework for traffic speed prediction on large-scale road networks

Expert Systems with Applications, 2022

Traffic prediction plays a crucial role in an intelligent transportation system (ITS) for enablin... more Traffic prediction plays a crucial role in an intelligent transportation system (ITS) for enabling advanced transportation management and services. In this paper, we address the problem of multi-step traffic speed prediction, including both short-and long-term predictions. We assert that it is important to consider not just the fixed spatial dependency of the road network (i.e., the connections between road segments) but also the dynamic spatial dependency of traffic within the static topology that intertwines with the temporal evolution of traffic condition across the entire network. We propose a novel deep learning model, named Self-Attention Graph Convolutional Network with Spatial, Sub-spatial and Temporal blocks (SAGCN-SST) model, that specifically capture such complex dynamic spatial-temporal processes. In SAGCN-SST, we integrate self-attention mechanism into graph convolutional networks in a novel framework design while using a sequence-to-sequence model in an encoderdecoder architecture for extracting long-temporal dependency of traffic speed. Two real-world datasets with frequent traffic congestion and accidents from large-scale road networks (i.e., Seattle and Los Angeles) are used to train and test our model. Our experiment results indicate that the proposed deep learning model consistently achieves the most accurate predictions (higher than 98% accuracy on both datasets for the short-and long-term predictions) when compared against well-known existing models in recent literature. The results also indicate that SAGCN-SST is robust against emergent traffic situations.

Year 1 Project Dissemination Report

Access Schemes and Packet Scheduling Techniques

Springer eBooks, Apr 19, 2007

The dual objectives of achieving efficient satellite resource utilization and acceptable user QoS... more The dual objectives of achieving efficient satellite resource utilization and acceptable user QoS levels require a consistent, controllable and flexible Radio Resource Management (RRM) scheme. The interest is here in managing packet data traffic of multimedia nature in mobile satellite systems. Complexity is added by the presence of multimedia traffic classes with differentiated QoS requirements and for the dynamically-varying channel conditions with (possible) consequent adaptations at the physical layer. The MAC layer is the &amp;amp;amp;amp;amp;amp;#x27;place&amp;amp;amp;amp;amp;amp;#x27; ...

Design and Experimental Validation of an LTE-Based Synchrophasor Network in a Medium Voltage Distribution Grid

We present and experimentally validate in a realscale medium voltage (MV) grid a synchrophasor ne... more We present and experimentally validate in a realscale medium voltage (MV) grid a synchrophasor network that exploits the availability of a public 4G LTE communication infrastructure. An 18 buses, 10 kV feeder located in Huissen, The Netherlands, has been equipped with 10 Phasor Measurement Units (PMUs) connected to the MV grid by means of dedicated voltage and current sensors. The PMUs stream synchrophasor data through a public 4G LTE network via an informationcentric networking-based middleware, named C-DAX. The measurements are received and time-aligned at a phasor data concentrator and fed to a real-time state estimation application. The paper presents the various field-trial components and validates the feasibility of exploiting the 4G LTE technology for PMUbased applications. Specifically we assess the performance of the adopted wireless telecommunication infrastructure with and without the C-DAX middleware, as well as the accuracy of the real-time state estimation process.

Crowdcloud: a crowdsourced system for cloud infrastructure

Cluster Computing, Aug 30, 2018

The widespread adoption of truly portable, smart devices and Do-It-Yourself computing platforms b... more The widespread adoption of truly portable, smart devices and Do-It-Yourself computing platforms by the general public has enabled the rise of new network and system paradigms. This abundance of well-connected, well-equipped, affordable devices, when combined with crowdsourcing methods, enables the development of systems with the aid of the crowd. In this work, we introduce the paradigm of Crowdsourced Systems, systems whose constituent infrastructure, or a significant part of it, is pooled from the general public by following crowdsourcing methodologies. We discuss the particular distinctive characteristics they carry and also provide their ''canonical'' architecture. We exemplify the paradigm by also introducing Crowdcloud, a crowdsourced cloud infrastructure where crowd members can act both as cloud service providers and cloud service clients. We discuss its characteristic properties and also provide its functional architecture. The concepts introduced in this work underpin recent advances in the areas of mobile edge/fog computing and co-designed/cocreated systems.

Redundant codes in hybrid networks

Hybrid networks consisting of satellite links and mobile ad hoc networks present a series of chal... more Hybrid networks consisting of satellite links and mobile ad hoc networks present a series of challenges due to different packet-loss patterns, delay, and, usually, scarce available bandwidth. In this scenario, redundant encoding, in the form of Forward Erasure Correction (FZC) codes [1],[2], can provide an effective protection against losses in multicast videoconferencing and video streaming applications. The use of efficient encoding techniques can provide further reduction on bandwidth requirements. A real test-bed based on a remote video streaming server interconnected via a GEO-satellite pipe to a local wireless LAN is presented in [3], by adopting the multicast network protocol. The satellite pipe is based on the commercial Skyplex network [4] that operates in the Ka-band with the Hot Bird 6 transponder. The developed platform, described in [5], is shown in Figure 8.15. The purpose is to provide users with a low-cost, high-availability platform for performing experiments with I...

Algorithms for Fault-Tolerant Placement of Stateful Virtualized Network Functions

2018 IEEE International Conference on Communications (ICC), 2018

Traditional network functions (NFs) such as firewalls are implemented in costly dedicated hardwar... more Traditional network functions (NFs) such as firewalls are implemented in costly dedicated hardware. By decoupling NFs from physical devices, network function virtualization enables virtual network functions (VNF) to run in virtual machines (VMs). However, VNFs are vulnerable to various faults such as software and hardware failures. To enhance VNF fault tolerance, the deployment of backup VNFs in stand-by VM instances is necessary. In case of stateful VNFs, stand-by instances require constant state updates from active instances during its operation. This will guarantee a correct and seamless handover from failed instances to stand-by instances after failures. Nevertheless, such state updates to stand-by instances could consume significant network bandwidth resources and lead to potential admission failures for VNF requests. In this paper, we study the fault-tolerant VNF placement problem with the optimization objective of admitting as many requests as possible. In particular, the VNF placement of active/stand-by instances, the request routing paths to active instances, and state transfer paths to stand-by instances are jointly considered. We devise an efficient heuristic algorithm to solve this problem, and propose a bicriteria approximation algorithm with performance guarantees for a special case of the problem. Simulations with realistic settings show that our algorithms can significantly improve the request admission rate compared to conventional approaches.

Fault tolerant placement of stateful VNFs and dynamic fault recovery in cloud networks

Computer Networks, 2019

Traditional network functions such as firewalls and Intrusion Detection Systems (IDS) are impleme... more Traditional network functions such as firewalls and Intrusion Detection Systems (IDS) are implemented in costly dedicated hardware, making the networks expensive to manage and inflexible to changes. Network function virtualization enables flexible and inexpensive operation of network functions, by implementing virtual network functions (VNFs) as software in virtual machines (VMs) that run in commodity servers. However, VNFs are vulnerable to various faults such as software and hardware failures. Without efficient and effective fault tolerant mechanisms, the benefits of deploying VNFs in networks can be traded-off. In this paper, we investigate the problem of fault tolerant VNF placement in cloud networks, by proactively deploying VNFs in stand-by VM instances when necessary. It is challenging because VNFs are usually stateful. This means that stand-by instances require continuous state updates from active instances during their operation, and the fault tolerant methods need to carefully handle such states. Specifically, the placement of active/stand-by VNF instances, the request routing paths to active instances, and state transfer paths to stand-by instances need to be jointly considered. To tackle this challenge, we devise an efficient heuristic algorithm for the fault tolerant VNF placement. We also propose two bicriteria approximation algorithms with provable approximation ratios for the problem without compute or bandwidth constraints. We then consider the dynamic fault recovery problem given that some placed active instances of VNFs may go faulty, for which we propose an approximation algorithm that dynamically switches traffic processing from faulty VNFs to stand-by instances. Simulations with realistic settings show that our algorithms can significantly improve the request admission rate compared to conventional approaches. We finally evaluate the performance of the proposed algorithm for the dynamic fault recovery problem in a real test-bed consisting of both physical and virtual switches, and results demonstrate that our algorithms have potentials of being applied in real scenarios.

Crowdcloud: a crowdsourced system for cloud infrastructure

Cluster Computing, 2018

The widespread adoption of truly portable, smart devices and Do-It-Yourself computing platforms b... more The widespread adoption of truly portable, smart devices and Do-It-Yourself computing platforms by the general public has enabled the rise of new network and system paradigms. This abundance of well-connected, well-equipped, affordable devices, when combined with crowdsourcing methods, enables the development of systems with the aid of the crowd. In this work, we introduce the paradigm of Crowdsourced Systems, systems whose constituent infrastructure, or a significant part of it, is pooled from the general public by following crowdsourcing methodologies. We discuss the particular distinctive characteristics they carry and also provide their ''canonical'' architecture. We exemplify the paradigm by also introducing Crowdcloud, a crowdsourced cloud infrastructure where crowd members can act both as cloud service providers and cloud service clients. We discuss its characteristic properties and also provide its functional architecture. The concepts introduced in this work underpin recent advances in the areas of mobile edge/fog computing and co-designed/cocreated systems.

Cross-Layer Enhancement to TCP Slow-Start over Geostationary Bandwidth on Demand Satellite Networks

Lecture Notes in Computer Science

It is well-known that the transmission control protocol (TCP) does not perform well in wireless a... more It is well-known that the transmission control protocol (TCP) does not perform well in wireless and satellite environments. We investigate the use of cross-layer design involving the transport and medium access control (MAC) layers in the context of a geostationary bandwidth-on-demand satellite network to simultaneously enhance TCP performance and to improve bandwidth utilization. In this paper, we focus on the slow-start phase of the connection. In essence, we create a bandwidth pipe between the two layers so that through cross-layer interactions, the TCP connections are aware of the satellite resources available to them, thus adjusting their congestion window accordingly. Our proposal includes minimal changes to the original protocol, allowing easier integration and inter-working with existing infrastructure. Our evaluation results show a shorter slow-start duration with better bandwidth utilization. Although the performance gain is higher in a lossy satellite link, we also found that it is dependent on the network load.

A Policy-Driven Network Management System for the Dynamic Configuration of Military Networks

Lecture Notes in Computer Science, 2009

Military networks constantly evolve to accommodate state-of-the-art technological developments ac... more Military networks constantly evolve to accommodate state-of-the-art technological developments across both military and commercial systems. Operating and maintaining such complex networks is no longer a trivial task. This paper presents a policy-based network management system for military networks which allows non-technical personnel (e.g. the military commander) to guide the network to behave towards specific objectives. Through policies, the system can optimize an IP-based multi-class military network with different or combinations of objectives, as requested by the decision makers. We show how the system can dynamically produce the required network configurations given specific requirements and illustrate its practicality using case studies.

Providing Relative Service Differentiation to TCP Flows over Split-TCP Geostationary Bandwidth on Demand Satellite Networks

Lecture Notes in Computer Science

We propose a combined transport-medium access control (MAC) layer scheme to provide relative serv... more We propose a combined transport-medium access control (MAC) layer scheme to provide relative service differentiation to Transmission Control Protocol (TCP) flows over a geostationary (GEO) bandwidth on demand (BoD) satellite networks. Our approach involves the joint configuration of TCP-Performance Enhancing Proxy (TCP-PEP) agents at the transport layer and the scheduling algorithm controlling the resource allocation at the MAC layer. The scheme is independent of the TCP variant used in the network. Extensive simulation results show that the two mechanisms exhibit complementary behavior in achieving the desired differentiation throughout the traffic load space: the TCP-PEP agents control differentiation at low system utilization, whereas the MAC scheduler becomes the dominant differentiation factor at high load.

Satellite Emulator for IP Networking Based on Linux

21st International Communications Satellite Systems Conference and Exhibit, 2003

Satellite environment provides a whole new spectrum of challenges to communications and networkin... more Satellite environment provides a whole new spectrum of challenges to communications and networking. The ability to study satellite network conditions in a reproducible and controlled manner has relied mainly on simulations and experimental testbed. Investigations on network protocols and multimedia applications over satellite networks via operational hardware-based testbed are costly and inflexible while results from simulators may be inaccurate due to assumptions made during the modeling process. Hence, a satellite network emulator is proposed to overcome these obstacles as it has complementing properties of both hardware-based testbed and pure softwarebased simulators. The proposed emulator is to imitate the performance dynamics in IP networks over satellite links. Operating at the Medium Access Control (MAC) of Data Link Layer, the emulator is able to emulate the propagation delay and bit errors of satellite channels. This is a very useful and cost-effective approach to show the impact of satellite link delay and transmission errors on Internet services and applications. Architecturally, the paper describes how the emulator is built upon LINUX IEEE 802.3 bridging facility (brctl) and LINUX firewall (netfilter). As packet level emulation paradigm is used, the emulator is capable of producing precise satellite channel model and thus, put forward itself as a reliable, adaptable and economical alternative for research on satellite communication networking and applications. This paper presents a project on a LINUX-based satellite link emulator, including the design and algorithm of the emulator and performance evaluation of Internet applications (data, voice and video) based on the emulator. It also explains the design and setup of a flexible and expandable testbed to supplement the performance evaluation process. The effects of the satellite emulator on voice, video and data are also studied through the use of Internet applications such as Microsoft Netmeeting and TCP/IP based data transfer.

Scheduling for proportional differentiated service provision in geostationary bandwidth on demand satellite networks

GLOBECOM '05. IEEE Global Telecommunications Conference, 2005., 2005

The rapid emergence of multimedia applications in the Internet has highlighted the need for servi... more The rapid emergence of multimedia applications in the Internet has highlighted the need for service differentiation in broadband satellite networks, which aim at being an integral part of the broadband network infrastructure. This paper presents a novel scheduler, called SWTP (Satellite Waiting Time Priority), to provide relative service differentiation in a DVB-RCS geostationary (GEO) satellite system where the network is structured to support a finite number of ordered service classes. We advocate the adoption of the proportional service differentiation model in the satellite domain to provide proportional delay differentiation to different traffic classes. The lightweight nature of the model makes it especially suitable for satellite systems as it minimizes computational cost by doing away with mechanisms such as admission control and resource reservation. Simulation results suggest that the SWTP scheduler can effectively and consistently provide proportional delay differentiation in satellite networks.