Xiangyang Gong

BalanceFlow: Controller load balancing for OpenFlow networks

In the discussion about Future Internet, Software-Defined Networking (SDN), enabled by OpenFlow, is currently seen as one of the most promising paradigm. While the availability and scalability concerns rises as a single controller could be alleviated by using replicate or distributed controllers, there lacks a flexible mechanism to allow controller load balancing. This paper proposes BalanceFlow, a controller load balancing architecture for OpenFlow networks. By utilizing CONTROLLER X action extension for OpenFlow switches and cross-controller communication, one of the controllers, called “super controller”, can flexibly tune the flow-requests handled by each controller, without introducing unacceptable propagation latencies. Experiments based on real topology show that BalanceFlow can adjust the load of each controller dynamically.

Incentive mechanism for participatory sensing under budget constraints

Incentive strategy is important in participatory sensing, especially when the budget is limited, to decide how much and where the samples should be collected. Current auction-based incentive strategies purchase sensing data with lowest price requirements to maximize the amount of samples. However, such methods may lead to inaccurate sensing result after data interpolation, particularly for participants that are massing in certain subregions where the low-price sensing data are usually aggregated. In this paper, we introduce weighted entropy as a quantitative metric to evaluate the distribution of samples and find that the distribution of data samples is another important factor to the accuracy of sensing result. We further propose a greedy-based incentive strategy which considers both the amount and distribution of samples in data collection. Simulations with real datasets confirmed the impact of samples distribution to data accuracy and demonstrated the efficacy of our proposed incentive strategy.

An enhanced personal photo recommendation system by fusing contextual and textual features on mobile device

As a main means to record scene in personal daily life, personal photos convey high-level semantic information (e.g., who, what, when, where) of an activity user engaged in. Different from other information retrieval tasks, personal photo recommendation depends on the measure of activity relevancy which is implicitly embedded in photos. Spurred by this observation, an enhanced recommendation approach by fusing both contextual and textual features is proposed. First, contextual relevancy is incrementally refined with an enhanced temporal and spatial clustering method respectively. Second, textual similarity of photo annotations is calculated using WordNet to augment the activity relevancy. Third, a fuzzy decision based multi-criteria ranking algorithm i.e., Preference Ranking Organization Method of Enrichment Evaluation (PROMETHEE) is adopted to make recommendations when giving an entry photo. A prototype has been developed on mobile device to illustrate this concept. Experiment results on a real dataset which contains 10,827 photos collected from 50 volunteers during 12 months demonstrate that our approach is more accurate than traditional schemes.

A Graph Indexing Approach for Content-Based Recommendation System

Conventional content-based recommendation systems use different classifying algorithms to group items into several groups and for each group generate a ranking list of items. An important characteristic of conventional content-based recommendation systems is that they use the same ranking list to make recommendations for items in each group, ignoring differences among items inside of a group. The paper proposes a content-based recommendation system built on top of a weighted un-directional graph. The graph describes the content similarity between items based on the semantic relations of their metadata. Neighbors of a node in the graph construct a ranking list of items to be recommended and there is a ranking list for each item. So it is able to emphasize differences among related items. We developed a prototype of the proposed system in Kaleido Photo project, and it proves to be sufficient to recommend most similar photos according to what the user is viewing.

Context-Aware Image Annotation and Retrieval on Mobile Device

In this paper, we present a novel method for image annotation and retrieval on mobile device by using contextual information. Image annotation is an effective way for content based image retrieval. However, manual annotation is an expensive and time consuming work, especially for mobile device. Here, we propose a semi-automatic way to give image annotation on mobile device. There are rich contexts for a mobile device, such as photo captured context, personal context and social network context. We synthesize these contexts and get useful semantic content of the photo. The synthesized results are treated as annotation suggestions. Our annotations include time, location, event, persons etc. We have implemented the annotation method in a prototype. The results show the method is simple and efficient.

Energy-Efficient Collaborative Localization for Participatory Sensing System

Location based services are getting increasingly popular in participatory sensing systems. They make use of location information on the mobile devices to support applications that improve personal health, object search, and entertainment. However, GPS positioning consumes a lot of energy, which can drain a mobile devices battery. Although WiFi localization and cell tower localization have been suggested as alternatives, they have lower localization accuracy and limited coverage. In this paper, we suggest a novel solution for multiple mobile devices to perform collaborative localization to reduce energy consumption and provide accurate localization. We divide the mobile devices into two groups, the aggregator group and the collector group. The aggregator group turns on their GPS periodically, while the collector group uses the locations of the aggregators to estimate their own locations. We formulate the aggregator set selection problem and propose two novel algorithms to minimize the energy consumption in collaborative localization. Simulations with real traces showed that our proposed solution can save up to 88% of the energy of the entire network.

Empirical analysis of different hierarchical addressing deployments

Currently the increasing prevalence of multi-homing and traffic engineering leads to an explosive growth of the global routing table. It is well known that hierarchical addressing could improve the routing scalability. Hence, some proposals exploring the routing architecture for future Internet reuse the hierarchical addressing for the locator assignment. However, this may result in some ASes assigned too many prefixes which would consequently make hosts, routers, Internet Service Providers and Domain Name System faced with big challenges. By modeling the Internet AS-level topology using a hierarchical graph, we define the processes of prefix assignment and routing advertisement in different hierarchical addressing deployment ways. Then, we quantify the impact of these deployments on the prefix assignment and the routing scalability based on the real routing data. we find that when the deploying position gets lower, the prefix amount of arbitrary AS is getting smaller, while the size of the global Forwarding Information Base is monotonically increasing. Comparing with the actual Internets data, suitable deployment ways for hierarchical addressing are obtained. With these deployment ways, the excessive prefix problem is solved and the size of the global Forwarding Information Base could be reduced into 56% or even 32% of the one in current Internet.

VNMC for network virtualization in OpenFlow network

Network virtualization provides an efficient way to share physical network infrastructure so that multiple virtual slices can run concurrently. OpenFlow is a promising and successful future networking technology which enables network innovation efficiently. However, the existing network virtualization methods in OpenFlow network usually have limitations such as introducing a proxy, flow entry conflicts, lacking of extensibility. In this paper, we present VNMC, a Virtual Network Management Component for rapidly creating, configuring and managing virtual networks in OpenFlow network. It has the following features: VNMC is implemented in a single controller without a middle layer. VNMC provides bandwidth guarantee, topology isolation, flow isolation and control isolation. It resolves flow entry conflicts from different controllers using the multiple flow tables mechanism defined in the latest OpenFlow standard. It manages the virtual networks in a modular way, which brings great extensibility to add new functionalities to virtual networks.

A hybrid model based on Kalman Filter and neutral network for traffic prediction

In this paper, a hybrid model based on Kalman Filter and Neural Network is introduced for traffic prediction to make our travel more convenient. The proposed model, taking both the real-time data and the historical data, can predict the link travel time in near future more accurately and thus increase the user service quality of APTS. The performance of evaluation is demonstrated on the real link travel time from Wenhui Bridge to Mingguang Bridge collected by mobile phone supporting GPS. Finally MAPE is used to calculate the prediction error and the result shows that the hybrid model performs well than both the two separate models. Based on our proposed model for traffic prediction, the APTS, which is one of the most important applications of ITS, would attract much more people to use the public transportation system and greatly reliever the burden of the urban traffic pressure.

Phone-Radar: Infrastructure-Free Device-to-Device Localization

In some practical scenarios such as tour guiding and children babysitting, one mobile device held by tour guides or parents need to know the distance and direction of another nearby mobile device held by tourist or children. However, to date, most existing pedestrian localization methods rely on a fixed external infrastructure, such as a global positioning system(GPS) or pre-deployed wifi access points to provide Localization service for mobile devices. Such methods are constrained either by limited GPS coverage or by complicated set-up procedures. We observe that, when two devices are moving, the change of their positions leads to the change of distance between them. Given the same movements, different relative locations between devices lead to different distance changes. Besides, the distance between and relative movement of devices can be measured by two phone-embedded sensors respectively. This motivates us to exploit the relative localization method by merely two mobile devices. In this work, we present Phone-Radar, which is an infrastructure-free device-to-device localization system. According to the propagation model of wireless signals, the change of distance between devices are modeled by the change of wireless signal strength between them. The movements of devices are recorded by the inertial sensors using step-counting method. We further study the relationship among the initial relative locations between the two devices, their relative movements and the change of received signal strength measurements. Moreover, we implement the proposed method and measure its performance under real world conditions. The testbed experiments show the efficiency of our proposed method.