Dezhong Yao

Energy efficient indoor tracking on smartphones

Abstract Continuously identifying a user’s location context provides new opportunities to understand daily life and human behavior. Indoor location systems have been mainly based on WiFi infrastructures which consume a great deal of energy mostly due to keeping the user’s WiFi device connected to the infrastructure and network communication, limiting the overall time when a user can be tracked. Particularly such tracking systems on battery-limited mobile devices must be energy-efficient to limit the impact on the experience of using a phone. Recently, there have been a lot of studies of energy-efficient positioning systems, but these have focused on outdoor positioning technologies. In this paper, we propose a novel indoor tracking framework that intelligently determines the location sampling rate and the frequency of network communication, to optimize the accuracy of the location data while being energy-efficient at the same time. This framework leverages an accelerometer, widely available on everyday smartphones, to reduce the duty cycle and the network communication frequency when a tracked user is moving slowly or not at all. Our framework can work for 14 h without charging, supporting applications that require this location information without affecting user experience.

Energy Efficient Task Scheduling in Mobile Cloud Computing

Cloud computing can enhance the computing capability of mobile systems by offloading. However, the communication between the mobile device and the cloud is not free. Transmitting large data to cloud consumes much more energy than processing data in mobile device, especially in a low bandwidth condition. Further, some processing tasks can avoid transmitting large data between mobile device and server. Those processing tasks encoding, rendering are as the compress algorithm, which can reduce the size of raw data before it is sent to server. In this paper, we present an energy efficient task scheduling strategy EETS to determine what kind of task with certain amount of data should be chosen to be offloaded under different environment. We have evaluated the scheduler by using an Android smartphone. The results show that our strategy can achieve 99% of accuracy to choose the right action in order to minimize the system energy usage.

Location-aware private service discovery in pervasive computing environment

Service discovery is an important and challenging issue in pervasive environments. Recent studies on service discovery mainly adopt Distributed Hash Tables (DHTs) based approaches without the consideration of private information protection. The main disadvantage of such approaches is that P2P overlay network does not reflect the physical topology and consequently generate substantial traffic overhead. In this paper, we propose a new service discovery scheme, considering both discovery efficiency and privacy protection. In the proposed scheme, we will present polar coordinate description and semantic service description to build location-aware overlay network. The polar coordinate description is able to efficiently build overlay networks, provide location-based searching, and alleviate traffic congestion. The physical hop counts and message overhead are significantly reduced due to the essentially distributed topology and ignorance of flooding operations. The usage of semantic description can find similar services more quickly and efficiently. We also consider the protection of mobile nodes private information. Results show that our proposed service discovery scheme is able to achieve much higher discovery success ratio and lower cost, compared to existing approaches.

Probabilistic routing algorithm based on contact duration and message redundancy in delay tolerant network

Summary The Delay Tolerant Network (DTN) is a novel Wireless Sensor Network architecture for an opportunistic network environment, in which environment end-to-end connection cannot be set up constantly between source and destination nodes pairs. In this paper, we have proposed a novel routing algorithm based on a hybrid of message delivery probability and message redundancy to reduce the communication overhead while keeping the high message delivery ratio. In this algorithm, the message delivery probability is calculated by the combined impact of meeting frequency and length of contact duration. Further, the maximum number of copies of the message is designated in the source node, and the forwarding task of message copies is assigned to relay nodes based on the pattern of a binary tree, so that multi-path parallel transmission can be implemented on message forwarding. Simulated results showed that the proposed routing algorithm can achieve a higher efficiency of message delivery than the related existing routing algorithms and it can also reduce the communication overhead significantly in general DTNs. Copyright

Human mobility synthesis using matrix and tensor factorizations

Human mobility prediction is of great advantage in route planning and schedule management. However, mobility data is a high-dimensional dataset in which multi-context prediction is difficult in a single model. Mobility data can usually be expressed as a home event, a work event, a shopping event and a traveling event. Previous works have only been able to learn and predict one type of mobility event and then integrate them. As the tensor model has a strong ability to describe high-dimensional information, we propose an algorithm to predict human mobility in tensors of location context data. Using the tensor decomposition method, we extract human mobility patterns with multiple expressions and then synthesize the future mobility event based on mobility patterns. The experiment is based on real-world location data and the results show that the tensor decomposition method has the highest accuracy in terms of prediction error among the three methods. The results also prove the feasibility of our multi-context prediction model.

Modeling User Activity Patterns for Next-Place Prediction

Location has played a very important role in pervasive computing systems. Beyond the current location, knowing an individuals next location in advance can also enable many novel mobile applications and services such as targeted advertising and the smooth handover between two separate networks. Although extensive studies about location prediction have been carried out, the existing prediction methods either encounter “cold start” problems when an individuals trajectory data are sparse or erratically perform when an individual performs activities in a new region. In this paper, we propose a novel approach based on the activity pattern for location prediction. Instead of directly predicting an individuals next location, we first infer the individuals next activity by modeling user activity patterns, and then, we predict his/her next location on the basis of the inferred next activity. Using real-life trajectory data, we demonstrate that the proposed approach can realize the smooth upgrade of the prediction performance and perform robustly.

Energy Conservation in Progressive Decentralized Single-Hop Wireless Sensor Networks for Pervasive Computing Environment

In the pervasive computing environment, energy efficiency is very important in terms of prolonging the lifetime of the communication. As a practical application in pervasive computing environment, wireless sensor networks consist of many sensors and several access points that make them work cooperatively to monitor/measure certain areas. Since the deployment of sensors in unknown sites impedes their recharging, thus exhausting their energy quite quickly, energy conservation becomes a critical issue. Unavoidably, the lavishness of both single- and multihop modes on energy conservation declines the systems life span severely. Therefore, a progressive decentralized single-hop method is conceived. This mechanism works with several phases, in each of which sensors may act in single- or multihop mode. The well-balanced energy consumption rate results in the extension of the systems life span. The method also adapts in general cases and has been proven by mathematical demonstration to totally balance the energy consumption. In addition, the method is very easy to implement. According to the simulation results, it attains most of the original aims of energy conservation.

The Best of BothWorlds: Combining Data-Independent and Data-Driven Approaches for Action Recognition

Motivated by the success of CNNs in object recognition on images, researchers are striving to develop CNN equivalents for learning video features. However, learning video features globally has proven to be quite a challenge due to the difficulty of getting enough labels, processing large-scale video data, and representing motion information. Therefore, we propose to leverage effective techniques from both data-driven and data-independent approaches to improve action recognition system. Our contribution is three-fold. First, we explicitly show that local handcrafted features and CNNs share the same convolution-pooling network structure. Second, we propose to use independent subspace analysis (ISA) to learn descriptors for state-of-the-art handcrafted features. Third, we enhance ISA with two new improvements, which make our learned descriptors significantly outperform the handcrafted ones. Experimental results on standard action recognition benchmarks show competitive performance.

Temporal-Based Ranking in Heterogeneous Networks

Ranking is a fundamental task for network analysis, benefiting to filter and find valuable information. Time information impacts results in content that is sensitive to trends and events ranking. The current ranking either assumes that user’s interest and concerns remain static and never change over time or focuses on detecting recency information. Meanwhile most prevalent networks like social network are heterogeneous, that composed of multiple types of node and complex reliance structures. In this paper, we propose a general Temporal based Heterogeneous Ranking (TemporalHeteRank) method. We demonstrate that TemporalHeteRank is suitable for heterogeneous networks on the intuition that there is a mutually information balance relationship between different types of nodes that could be reflected on ranking results. We also explore the impact of node temporal feature in ranking, then we use the node life span by carefully investigating the issues of feasibility and generality. The experimental results on sina weibo ranking prove the effectiveness of our proposed approach.

Sparse Online Relative Similarity Learning

For many data mining and machine learning tasks, the quality of a similarity measure is the key for their performance. To automatically find a good similarity measure from datasets, metric learning and similarity learning are proposed and studied extensively. Metric learning will learn a Mahalanobis distance based on positive semi-definite (PSD) matrix, to measure the distances between objectives, while similarity learning aims to directly learn a similarity function without PSD constraint so that it is more attractive. Most of the existing similarity learning algorithms are online similarity learning method, since online learning is more scalable than offline learning. However, most existing online similarity learning algorithms learn a full matrix with d2 parameters, where d is the dimension of the instances. This is clearly inefficient for high dimensional tasks due to its high memory and computational complexity. To solve this issue, we introduce several Sparse Online Relative Similarity (SORS) learning algorithms, which learn a sparse model during the learning process, so that the memory and computational cost can be significantly reduced. We theoretically analyze the proposed algorithms, and evaluate them on some real-world high dimensional datasets. Encouraging empirical results demonstrate the advantages of our approach in terms of efficiency and efficacy.