Hwee-Pink Tan

Bio: Hwee-Pink Tan is an academic researcher from Singapore Management University. The author has contributed to research in topics: Wireless sensor network & Key distribution in wireless sensor networks. The author has an hindex of 35, co-authored 140 publications receiving 4866 citations. Previous affiliations of Hwee-Pink Tan include University College Dublin & École Polytechnique Fédérale de Lausanne.

Machine Learning in Wireless Sensor Networks: Algorithms, Strategies, and Applications

Abstract: Wireless sensor networks (WSNs) monitor dynamic environments that change rapidly over time. This dynamic behavior is either caused by external factors or initiated by the system designers themselves. To adapt to such conditions, sensor networks often adopt machine learning techniques to eliminate the need for unnecessary redesign. Machine learning also inspires many practical solutions that maximize resource utilization and prolong the lifespan of the network. In this paper, we present an extensive literature review over the period 2002–2013 of machine learning methods that were used to address common issues in WSNs. The advantages and disadvantages of each proposed algorithm are evaluated against the corresponding problem. We also provide a comparative guide to aid WSN designers in developing suitable machine learning solutions for their specific application challenges.

Sensor OpenFlow: Enabling Software-Defined Wireless Sensor Networks

Abstract: While it has been a belief for over a decade that wireless sensor networks (WSN) are application-specific, we argue that it can lead to resource underutilization and counter-productivity. We also identify two other main problems with WSN: rigidity to policy changes and difficulty to manage. In this paper, we take a radical, yet backward and peer compatible, approach to tackle these problems inherent to WSN. We propose a Software-Defined WSN architecture and address key technical challenges for its core component, Sensor OpenFlow. This work represents the first effort that synergizes software-defined networking and WSN.

Wireless sensor networks powered by ambient energy harvesting (WSN-HEAP) - Survey and challenges

Abstract: Wireless sensor networks (WSNs) research has pre-dominantly assumed the use of a portable and limited energy source, viz. batteries, to power sensors. Without energy, a sensor is essentially useless and cannot contribute to the utility of the network as a whole. Consequently, substantial research efforts have been spent on designing energy-efficient networking protocols to maximize the lifetime of WSNs. However, there are emerging WSN applications where sensors are required to operate for much longer durations (like years or even decades) after they are deployed. Examples include in-situ environmental/habitat monitoring and structural health monitoring of critical infrastructures and buildings, where batteries are hard (or impossible) to replace/recharge. Lately, an alternative to powering WSNs is being actively studied, which is to convert the ambient energy from the environment into electricity to power the sensor nodes. While renewable energy technology is not new (e.g., solar and wind) the systems in use are far too large for WSNs. Those small enough for use in wireless sensors are most likely able to provide only enough energy to power sensors sporadically and not continuously. Sensor nodes need to exploit the sporadic availability of energy to quickly sense and transmit the data. This paper surveys related research and discusses the challenges of designing networking protocols for such WSNs powered by ambient energy harvesting.

Mobile big data analytics using deep learning and apache spark

Abstract: The proliferation of mobile devices, such as smartphones and Internet of Things gadgets, has resulted in the recent mobile big data era. Collecting mobile big data is unprofitable unless suitable analytics and learning methods are utilized to extract meaningful information and hidden patterns from data. This article presents an overview and brief tutorial on deep learning in mobile big data analytics and discusses a scalable learning framework over Apache Spark. Specifically, distributed deep learning is executed as an iterative MapReduce computing on many Spark workers. Each Spark worker learns a partial deep model on a partition of the overall mobile, and a master deep model is then built by averaging the parameters of all partial models. This Spark-based framework speeds up the learning of deep models consisting of many hidden layers and millions of parameters. We use a context-aware activity recognition application with a real-world dataset containing millions of samples to validate our framework and assess its speedup effectiveness.

The Self-Organizing Map

Abstract: An overview of the self-organizing map algorithm, on which the papers in this issue are based, is presented in this article.

Wireless Networks With RF Energy Harvesting: A Contemporary Survey

Abstract: Radio frequency (RF) energy transfer and harvesting techniques have recently become alternative methods to power the next-generation wireless networks As this emerging technology enables proactive energy replenishment of wireless devices, it is advantageous in supporting applications with quality-of-service requirements In this paper, we present a comprehensive literature review on the research progresses in wireless networks with RF energy harvesting capability, which is referred to as RF energy harvesting networks (RF-EHNs) First, we present an overview of the RF-EHNs including system architecture, RF energy harvesting techniques, and existing applications Then, we present the background in circuit design as well as the state-of-the-art circuitry implementations and review the communication protocols specially designed for RF-EHNs We also explore various key design issues in the development of RF-EHNs according to the network types, ie, single-hop networks, multiantenna networks, relay networks, and cognitive radio networks Finally, we envision some open research directions

When is nearest neighbor meaningful

Abstract: We explore the effect of dimensionality on the nearest neighbor problem. We show that under a broad set of conditions (much broader than independent and identically distributed dimensions), as dimensionality increases, the distance to the nearest data point approaches the distance to the farthest data point. To provide a practical perspective, we present empirical results on both real and synthetic data sets that demonstrate that this effect can occur for as few as 10-15 dimensions. These results should not be interpreted to mean that high-dimensional indexing is never meaningful; we illustrate this point by identifying some high-dimensional workloads for which this effect does not occur. However, our results do emphasize that the methodology used almost universally in the database literature to evaluate high-dimensional indexing techniques is flawed, and should be modified. In particular, most such techniques proposed in the literature are not evaluated versus simple linear scan, and are evaluated over workloads for which nearest neighbor is not meaningful. Often, even the reported experiments, when analyzed carefully, show that linear scan would outperform the techniques being proposed on the workloads studied in high (10-15) dimensionality!.

Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition

Abstract: Human activity recognition (HAR) tasks have traditionally been solved using engineered features obtained by heuristic processes. Current research suggests that deep convolutional neural networks are suited to automate feature extraction from raw sensor inputs. However, human activities are made of complex sequences of motor movements, and capturing this temporal dynamics is fundamental for successful HAR. Based on the recent success of recurrent neural networks for time series domains, we propose a generic deep framework for activity recognition based on convolutional and LSTM recurrent units, which: (i) is suitable for multimodal wearable sensors; (ii) can perform sensor fusion naturally; (iii) does not require expert knowledge in designing features; and (iv) explicitly models the temporal dynamics of feature activations. We evaluate our framework on two datasets, one of which has been used in a public activity recognition challenge. Our results show that our framework outperforms competing deep non-recurrent networks on the challenge dataset by 4% on average; outperforming some of the previous reported results by up to 9%. Our results show that the framework can be applied to homogeneous sensor modalities, but can also fuse multimodal sensors to improve performance. We characterise key architectural hyperparameters’ influence on performance to provide insights about their optimisation.

Mobile Edge Computing: A Survey

Abstract: Mobile edge computing (MEC) is an emergent architecture where cloud computing services are extended to the edge of networks leveraging mobile base stations. As a promising edge technology, it can be applied to mobile, wireless, and wireline scenarios, using software and hardware platforms, located at the network edge in the vicinity of end-users. MEC provides seamless integration of multiple application service providers and vendors toward mobile subscribers, enterprises, and other vertical segments. It is an important component in the 5G architecture which supports variety of innovative applications and services where ultralow latency is required. This paper is aimed to present a comprehensive survey of relevant research and technological developments in the area of MEC. It provides the definition of MEC, its advantages, architectures, and application areas; where we in particular highlight related research and future directions. Finally, security and privacy issues and related existing solutions are also discussed.