Baihua Zheng

Data management in location-dependent information services

Location-dependent information services have great promise for mobile and pervasive computing environments. They can provide local and nonlocal news, weather, and traffic reports as well as directory services. Before they can be implemented on a large scale, however, several research issues must be addressed.

Cache invalidation and replacement strategies for location-dependent data in mobile environments

Mobile location-dependent information services (LDISs) have become increasingly popular in recent years. However, data caching strategies for LDISs have thus far received little attention. In this paper, we study the issues of cache invalidation and cache replacement for location-dependent data under a geometric location model. We introduce a new performance criterion, called caching efficiency, and propose a generic method for location-dependent cache invalidation strategies. In addition, two cache replacement policies, PA and PAID, are proposed. Unlike the conventional replacement policies, PA and PAID take into consideration the valid scope area of a data value. We conduct a series of simulation experiments to study the performance of the proposed caching schemes. The experimental results show that the proposed location-dependent invalidation scheme is very effective and the PA and PAID policies significantly outperform the conventional replacement policies.

IR-Tree: An Efficient Index for Geographic Document Search

Given a geographic query that is composed of query keywords and a location, a geographic search engine retrieves documents that are the most textually and spatially relevant to the query keywords and the location, respectively, and ranks the retrieved documents according to their joint textual and spatial relevances to the query. The lack of an efficient index that can simultaneously handle both the textual and spatial aspects of the documents makes existing geographic search engines inefficient in answering geographic queries. In this paper, we propose an efficient index, called IR-tree, that together with a top-k document search algorithm facilitates four major tasks in document searches, namely, 1) spatial filtering, 2) textual filtering, 3) relevance computation, and 4) document ranking in a fully integrated manner. In addition, IR-tree allows searches to adopt different weights on textual and spatial relevance of documents at the runtime and thus caters for a wide variety of applications. A set of comprehensive experiments over a wide range of scenarios has been conducted and the experiment results demonstrate that IR-tree outperforms the state-of-the-art approaches for geographic document searches.

Semantic Caching in Location-Dependent Query Processing

A method is presented in this paper for answering location-dependent queries in a mobile computing environment. We investigate a common scenario where data objects (e.g., restaurants and gas stations) are stationary while clients that issue queries about the data objects are mobile. Our proposed technique constructs a Voronoi Diagram (VD) on the data objects to serve as an index for them. A VD defines, for each data object d, the region within which d is the nearest point to any mobile client within that region. As such, the VD can be used to answer nearest-neighbor queries directly. Furthermore, the area within which the answer is valid can be computed. Based on the VD, we develop a semantic caching scheme that records a cached item as well as its valid range. A simulation is conducted to study the performance of the proposed semantic cache in comparison with the traditional cache and the baseline case where no cache is used. We show that the semantic cache has a much better performance than the other two methods.

Grid-partition index: a hybrid method for nearest-neighbor queries in wireless location-based services

Traditional nearest-neighbor (NN) search is based on two basic indexing approaches: object-based indexing and solution-based indexing. The former is constructed based on the locations of data objects: using some distance heuristics on object locations. The latter is built on a precomputed solution space. Thus, NN queries can be reduced to and processed as simple point queries in this solution space. Both approaches exhibit some disadvantages, especially when employed for wireless data broadcast in mobile computing environments.In this paper, we introduce a new index method, called the grid-partition index, to support NN search in both on-demand access and periodic broadcast modes of mobile computing. The grid-partition index is constructed based on the Voronoi diagram, i.e., the solution space of NN queries. However, it has two distinctive characteristics. First, it divides the solution space into grid cells such that a query point can be efficiently mapped into a grid cell around which the nearest object is located. This significantly reduces the search space. Second, the grid-partition index stores the objects that are potential NNs of any query falling within the cell. The storage of objects, instead of the Voronoi cells, makes the grid-partition index a hybrid of the solution-based and object-based approaches. As a result, it achieves a much more compact representation than the pure solution-based approach and avoids backtracked traversals required in the typical object-based approach, thus realizing the advantages of both approaches.We develop an incremental construction algorithm to address the issue of object update. In addition, we present a cost model to approximate the search cost of different grid partitioning schemes. The performances of the grid-partition index and existing indexes are evaluated using both synthetic and real data. The results show that, overall, the grid-partition index significantly outperforms object-based indexes and solution-based indexes. Furthermore, we extend the grid-partition index to support continuous-nearest-neighbor search. Both algorithms and experimental results are presented.

Spatial queries in wireless broadcast systems

Owing to the advent of wireless networking and personal digital devices, information systems in the era of mobile computing are expected to be able to handle a tremendous amount of traffic and service requests from the users. Wireless data broadcast, thanks to its high scalability, is particularly suitable for meeting such a challenge. Indexing techniques have been developed for wireless data broadcast systems in order to conserve the scarce power resources in mobile clients. However, most of the previous studies do not take into account the impact of location information of users. In this paper, we address the issues of supporting spatial queries (including window queries and kNN queries) of location-dependent information via wireless data broadcast. A linear index structure based on the Hilbert curve and corresponding search algorithms are proposed to answer spatial queries on air. Experiments are conducted to evaluate the performance of the proposed indexing technique. Results show that the proposed index and its enhancement outperform existing algorithms significantly.

DSI: A Fully Distributed Spatial Index for Location-Based Wireless Broadcast Services

Recent announcement of the MSN Direct Service has demonstrated the feasibility and industrial interest in utilizing wireless broadcast for pervasive information services. To support location-based services in wireless data broadcast systems, a distributed spatial index (called DSI) is proposed in this paper. DSI is highly efficient because it has a linear yet fully distributed structure that facilitates multiple search paths to be naturally mixed together by sharing links. Moreover, DSI is very resilient in error-prone wireless communication environments. Search algorithms for two classical location-based queries, window queries and kNN queries, based on DSI are presented. Performance evaluation of DSI shows that DSI significantly outperforms R-tree and Hilbert Curve Index, two state-of-the-art spatial indexing techniques for wireless data broadcast

Proactive caching for spatial queries in mobile environments

Semantic caching enables mobile clients to answer spatial queries locally by storing the query descriptions together with the results. However, it supports only a limited number of query types, and sharing results among these types is difficult. To address these issues, we propose a proactive caching model which caches the result objects as well as the index that supports these objects as the results. The cached index enables the objects to be reused for all common types of queries. We also propose an adaptive scheme to cache such an index, which further optimizes the query response time for the best user experience. Simulation results show that proactive caching achieves a significant performance gain over page caching and semantic caching in mobile environments where wireless bandwidth and battery are precious resources.

The D-tree: an index structure for planar point queries in location-based wireless services

Location-based services (LBSs), considered as a killer application in the wireless data market, provide information based on locations specified in the queries. In this paper, we examine the indexing issue for querying location-dependent data in wireless LBSs; in particular, we focus on an important class of queries, planar point queries. To address the issues of responsiveness, energy consumption, and bandwidth contention in wireless communications, an index has to minimize the search time and maintain a small storage overhead. It is shown that the traditional point-location algorithms and spatial index structures fail to achieve either objective or both. This paper proposes a new index structure, called D-tree, which indexes spatial regions based on the divisions that form the boundaries of the regions. We describe how to construct a binary D-tree index, how to process queries based on the D-tree, and how to page the binary D-tree. Moreover, two parameterized methods for partitioning the original space, called fixed grid assignment (FGA) and adaptive grid assignment (AGA), are proposed to enhance the D-tree. The performance of the D-tree is evaluated using both synthetic and real data sets. Experimental results show that the proposed D-tree outperforms the well-known indexes such as the R/sup */-tree, and that both the FGA and AGA approaches can achieve different performance trade-offs between the index search time and storage overhead by fine-tuning their algorithmic parameters.

PRESS: a novel framework of trajectory compression in road networks

Location data becomes more and more important. In this paper, we focus on the trajectory data, and propose a new framework, namely PRESS ( P aralleled R oad-Network-Based Trajectory Compr ess ion), to effectively compress trajectory data under road network constraints. Different from existing work, PRESS proposes a novel representation for trajectories to separate the spatial representation of a trajectory from the temporal representation, and proposes a Hybrid Spatial Compression (HSC) algorithm and error Bounded Temporal Compression (BTC) algorithm to compress the spatial and temporal information of trajectories respectively. PRESS also supports common spatial-temporal queries without fully decompressing the data. Through an extensive experimental study on real trajectory dataset, PRESS significantly outperforms existing approaches in terms of saving storage cost of trajectory data with bounded errors.