Research

Research Interests

Long story short, I’m interested in applied research and technology to build intelligent and inclusive smart cities (as like everyone else).¬†ūüôā

Short story longer, I’m currently interested in the use of¬†applied research and technology to build intelligent,¬†inclusive and integrative societies, to improve the lives of vulnerable groups of people – such¬†as the elderly and persons with disabilities. I play in the arena of networked embedded sensor¬†systems and Internet of Things (IoT), and dabble in data analytics and crowdsourcing. I believe that technology alone cannot solve all human problems, but that technology must¬†be co-designed and co-developed in tandem with other stakeholders in the ecosystem.

R&D Experience

SAMBA @ SMU, Singapore :: expected commencement date: Jan 2018; Role: Co-Investigator
Smart Accessibility & Mobility for Barrier-free Access
SAMBA aims to design a scalable, self-sustaining system that can collect, classify and determine accessible point-to-point routes that are suitable for barrier-free access. This system relies primarily and heavily on passive crowdsourced data collection in order to overcome the problem of volunteer and user fatigue. The project is approved for funding by the Tote Board РEnabling Lives Initiative (TB-ELI) Grant Call 4. This is an 18-month project that will officially commence in Jan 2018.

NUHS @ SMU, Singapore :: Oct 2016 – present
In-Home Sensors for Assessment of Cognitive and Psychological Health of Older Adults: A Pilot Study
This project aims to detect early cognitive decline in elderly, through continuous activity measurements from sensors that are instrumented in each elderly home over a period of two months.

SHINESeniors @ SMU, Singapore :: May 2015 – present
Smart Homes and Intelligent Neighbors to Enable Seniors
SHINESeniors aims to enhance healthcare and community caregiver support for the elderly, through the use of unobtrusive in-home monitoring sensor devices. One of the key focus of the project is the use of heterogeneous data sources (such as surveys, observations and sensor data) to provide pre-emptive care, and thereby allowing the elderly to age-in-place. The key stakeholders of the project include Eastern Health Alliance (EHA), voluntary welfare organizations such as GoodLife! and Thye Hua Kwan Moral Charities, and Ministry of Health (MOH).

IDA-JLD @ I2R, Singapore :: Dec 2013 – Jun 2015
Supply and Delivery of Shared Sensor Network for the Smart & Connected JLD Test-bed
This project aims to provide a smart testbed in the Jurong Lake District (JLD) vicinity in Singapore, through the inter-connection of about two hundred nodes with multi-modal sensing capabilities. My primary role is to provide the architectural design and development of networking protocols to enable bi-directional, multihop data transfer between the gateways and sensor nodes.

SenseSURF @ I2R, A*STAR, Singapore :: Dec 2014 РMar 2015
Wireless Sensor Mesh Networks Test-bed (NCS -I2R Joint Lab)
As the technical lead for SenseSURF, I work closely with a team of researchers to develop a proof-of-concept for the design, development and deployment of: (i) a small-scale wireless sensor mesh network, for both indoor and outdoor monitoring; and (ii) a sensor network management platform to evaluate technology capabilities for future large-scale rollout of a Smart Nation Platform.

UNI-SENSE @ I2R, Singapore :: Aug 2012 – Jun 2015
Unified and Sustainable Sensing and Transport Architecture for Large Scale and Heterogeneous Sensor Networks
As part of the Sensor Network Design group, I am involved in designing and developing intelligent networking protocols for large scale sensor networks. I also look into auxiliary issues, such as adaptive duty cycling and smart site survey tools. In addition, I am the main designer of UniNet – a unified networking architecture for scalable, sustainable and robust wireless sensor networks. UniNet has been deployed and tested in indoor and outdoor deployments, such as the WSNoise and IDA-JLD projects.

WSNoise @ I2R, Singapore :: Aug 2012 – Jun 2014
Wireless Sensor Networks for Real-time and Continuous Ambient Noise Mapping
WSNoise focuses on the real-time acquisition noise sensor readings in outdoor urban environments. My task is to provide wireless networking capabilities for the 150 sensor nodes that have been deployed across three physical sites – Clementi, Ang Mo Kio and Jurong Lake District – in Singapore.

DTN-AN @ I2R, Singapore :: Aug 2012 – May 2014
Disruption Tolerant Networking (DTN) for Airborne Networks
Airborne networks generally suffer from frequent disruptions due to high node mobility, ad hoc connectivity and line-of-sight blockages. This project explores the feasibility of utilizing disruptiontolerant networking techniques to alleviate these challenges. I am the co-designer of GTA-m, a multi-copy greedy trajectory-aware routing protocol for airborne networks that exploits the use of flight information to forwarded bundles to intended destination(s).

FUM @ School of Computing, NUS :: Jul 2010 РAug 2012
Improving Future Urban Mobility through ICT
Postdoctoral research as part of the Future Urban Mobility (FUM) project, which is a collaboration with the Singapore-MIT Alliance for Research & Technology (SMART) center. The key objective of FUM is to improve the urban transportation system while enhancing its environmental sustainability. Some topics of interest include efficient communication and resource allocation in vehicular disruption tolerant networks, which are essential in applications such as traffic information dissemination as well as taxi booking and dispatch systems.

Doctoral Research @ School of Computing, NUS :: Nov 2006 – Dec 2010
Communication Protocols for Energy-Constrained Networks
We identify the caveats of existing networking protocols for energy constrained networks and propose three novel algorithms that provide better energy efficiency: (i) A2 -MAC – an adaptive, anycast MAC protocol that effectively reduces energy expenditure in generic low-powered wireless sensor networks by allowing nodes to operate with different duty cycles and forwarding sets; (ii) IQAR – an information quality (IQ) aware routing protocol that finds the least-cost routing tree that satisfies a given IQ constraint when a phenomenon of interest (PoI) occurs in the network; and (iii) IQDEA – an IQ-aware delay efficient aggregation scheme that minimizes PoI detection delays and transmission costs in duty cycled networks while satisfying application-level IQ requirements.

USCAM-CQ @ School of Computing, NUS and I2R, Singapore :: Dec 2006 – Jul 2009
UWB-enabled Sentient Computing Architecture and Middleware with Coordinated QoS (USCAM-CQ)
The USCAM-CQ project is part of the Ultra Wideband-enabled Sentient Computing (UWB-SC) Research Programme funded by A*STAR. My focus in this project is on communication protocols for UWB networks. The characteristics of the UWB physical layer are studied and implemented in the Qualnet (network) simulator. In addition, we design and implement: (i) a slotted-Aloha MAC protocol over the Time-Hopping (TH) UWB PHY; and (ii) a fair resource allocation over a TH-UWB PHY.

TRITON @ I2R, Singapore :: Mar 2005 – Oct 2006
TRI-media Telematic Oceanographic Network (TRITON)
Graduate student attached to Work Package 3 (WP3) of the project, which specializes in underwater acoustic sensor network communications. We study the differences between the underwater acoustic environment and terrestrial RF environment, and implement the acoustic environment in the Qualnet (network) simulator. In addition, we design and implement: (i) a MAC protocol that is feasible for use in underwater environments that experience high latencies; and (ii) a distributed CDMA code assignment algorithm for wireless sensor networks.

Bachelors Research @ I2R and School of Computing, NUS :: Jan 2004 РDec 2004
Dynamic Adaptation of MANET Routing Protocols
We first provide an overview of protocol characteristics and a detailed study of network characteristics that can affect the performance of mobile ad hoc network routing protocols. We then propose three adaptive schemes that can be implemented on top of any reactive routing protocol: (i) dynamic adaptation of frequency of broadcasts; (ii) reducing network overhead using link stability; and (iii) dynamic topology control to reduce interference.