VRSense: Validity region sensitive query processing strategies for static and mobile point-of-interests in MANETs
Introduction
With increasingly popular in recent wireless, mobile, and location-aware devices, users are able to query any point-of-interest (POI) as a major part of location-based services (LBS) supported by ubiquitous communication infrastructure. For example, a user walking in a street queries a coffee shop, or a driver in a vehicle equipped with an on-board global positioning system (GPS) queries the nearest gas-station. They are all supported by 3G/4G or roadside unit (RSU) (e.g., an access point, an infostation [1], or a message relaybox [2]) located along the road that acts as a gateway to an infrastructure-based network, where a centralized server searches the queried POI and answers a set of query results back. A good deal of effort has been devoted into the designing of query processing strategies [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15] applied to diverse wireless and/or mobile networks to realize seamless services and mobile and pervasive computing.
Although 3G/4G high speed wireless networks are increasingly popular, many areas are still remained uncovered, such as an urban or remotely isolated area. The current infrastructure could even be collapsed and unavailable due to the disasters including recent hurricanes and earthquakes. For example, emergency communications and queries among rescue people can be conducted through single- and multi-hop relays without the help of infrastructure [3], [4], [5], [9]. In this paper, we investigate query processing strategies in mobile ad hoc networks (MANETs) facing with two key challenges. First, due to the lack of centralized coordination, limited computing and communication capabilities, and time-varying network topologies, it is admittedly challenging to design an efficient query processing scheme in MANETs. In query processing, when a user (later node) broadcasts a query to collect an object information what it is interested (e.g., POI), it may receive more than one query reply as a query result if multiple nodes can answer the query. Unlike an infrastructure-based network, where the query can directly be sent to and responded from a centralized server, the query is often flooded to the rest of nodes through multi-hop relays in the network. Thus, a blind query broadcast operation followed by a series of unconditional query forwarding operations is inefficient and even harmful because of redundant query retransmissions, causing query contentions and collisions [16].This could consume a significant amount of battery energy because wireless communication could be responsible for more than half of the total energy consumption [17]. In addition, when a node intends to reply a query result by unicasting it back to the query sender, unlike a wired network, all one hop neighbor nodes can still overhear the query result, as if it is a broadcast packet [16]. Thus, it is critical to reduce the number of broadcasted queries without degrading the validity of query result. Second, unlike traditional stationary POIs, it is also challenging to consider mobile POIs in designing query processing schemes in MANETs. This is because the mobility of POIs may invalidate prior query result. The query generating node in fact even does not aware whether the query result becomes obsolete. Thus, how to efficiently update the freshness of query results and when to judiciously re-generate a new query also become critical.
To address these challenges, we propose a set of query processing strategies based on a validity region, called VRSense, to efficiently update the freshness of queried data and reduce the query traffic in the presence of static and mobile POIs in MANETs. A validity region is defined as a virtual area where a query result remains the same as long as a query issuing node is located within the area. Several validity regions and variants based querying techniques [15], [18], [19], [20] have been proposed to improve query performance and reduce query traffic. However, most approaches still rely on an infrastructure-based network with a centralized server in forming a validity region. They may not directly be applied to an infrastructure-less network, where each node is required to build its region in a distributed manner. Our contributions are briefly summarized in three-fold:
- •
We first present both time- and location-sensitive query types in the presence of static and mobile POIs without deploying a validity region based approach in MANETs. Then we provide a set of query scenarios, identify its corresponding query processing operations, and analyze its query processing implications.
- •
Second, we investigate a validity region based query processing approach and propose basic rectangle (Rect) and convex hull (Conv) based validity regions and their corresponding query processing operations. We also extend them by adaptively combining both techniques (Adapt) and considering an opportunistic overhearing (Adapt:Ovhr). In addition, we modify the safe exit algorithm [20], deployed in an infrastructure-based query processing scheme (Greedy), to work in MANETs for the purpose of performance comparison.
- •
Third, we observe the mobility of POIs and its impact on the validity region in MANETs. Then we propose both reduced (Reduced) and probabilistic (Prob) techniques and their corresponding query processing operations (Reduced – Rect/Conv/Adapt/Adapt:Ovhr and Prob – Rect/Conv/Adapt/Adapt:Ovhr) to flexibly approximate the validity region in the presence of mobile POIs.
We conduct extensive simulation experiments using the OMNeT++ [21] for performance comparison and analysis in terms of the number of queries and the time spent in the validity region. The simulation results indicate that the convex hull based approaches, Adapt and Adapt:Ovhr, can reduce the number of queries and increase the time spent in the validity region during the query processing operations. The Reduced and Prob applied to the proposed query processing strategies can further improve the query performance with mobile POIs. Note that this paper is significantly extended1 based on our initial work [22].
The rest of paper is organized as follows. Prior query processing approaches are comprehensively reviewed and analyzed in Section 2.The basic query operation and time- and location-sensitive query operations are presented in Section 3. A set of validity region-based query processing strategies with static and mobile POIs is presented in Sections 4 and 5, respectively. Section 6 is devoted to performance evaluation and analysis. Finally, we further discuss the potential issues of query processing in MANETs followed by conclusion in Sections 7 and 8, respectively.
Section snippets
Related work
In this section, we comprehensively review prior non-region based querying approaches in terms of top-k query and k nearest neighbor (kNN) query, and then analyze region based querying approaches in diverse wireless and/or mobile networks.
Top-k query: Top-k query retrieves the k number of data items, ordered by the score based on a target attribute, from adjacent nodes in the network. A query issuing node floods a query to the rest of nodes in the network. When a node receives the query, it
Query operations
In this section, we first briefly present a basic query operation and then analyze time- and location-sensitive query operations in MANETs.
Validity region sensitive query
In the VRSense, we first propose four validity region-based query processing schemes with static POIs in MANETs.We also enhance and vary one of our proposed schemes for performance comparison.
Validity region sensitive query with mobile point of interests
In the VRSense, we also extend the proposed query processing strategies for mobile POIs. We first observe the mobility of POIs and its impact on the validity region and then present two techniques to reset the validity region accordingly.
Performance evaluation
We evaluate and compare the performance of proposed query processing strategies with both static and mobile POIs by changing key simulation parameters in MANETs.
Discussion
In this section, we further identify two major issues in order to see the full potential of the proposed query processing strategies. We also summarize the cons and pros of the proposed query processing strategies, present other related issues, and compare them with prior region-based query processing strategies proposed under different environments [15], [19], [20], [26].
Concluding remarks
In this paper, we investigated validity region sensitive query processing strategies to efficiently reduce the number of queries and update the freshness of queried data in MANETs. We first proposed basic rectangle and convex hull based validity regions and their corresponding query processing operations for static POIs.Then we extended them by combining rectangle and convex hull techniques and considering an opportunistic overhearing. In addition, we approximated the validity region and
Acknowledgment
This research was supported in part by International Cooperative Research Grant from Incheon National University (Incheon, Korea) in 2015.
References (28)
- et al.
Processing k nearest neighbor queries in location-aware sensor networks
Signal Process.
(2007) - et al.
An efficient algorithm for computing safe exit points of moving range queries in directed road networks
Inf. Syst.
(2014) - et al.
A safe exit approach for efficient network-based moving range queries
Data Knowl. Eng.
(2012) An efficient algorithm for determining the convex hull of a finite planar set
Inf. Process Lett.
(1972)- et al.
The infostation challenge: balancing cost and ubiquity in diverse wireless data
IEEE Pers. Commun.
(2000) - et al.
The application-Based clustering concept and requirements for intervehicle networks
IEEE Commun. Mag.
(2005) - et al.
Two-phase top-k query processing in mobile Ad Hoc networks
Proc. Int’l Conf. on Network-based Information Systems
(2011) - et al.
A robust routing method for top-k queries in mobile Ad Hoc networks
Proc. Mobile Data Management
(2013) - et al.
Efficient cluster-based top-k query routing with data replication in MANETs
Soft. Comput.
(2015) - et al.
Toward the optimal itinerary-based KNN query processing in mobile sensor networks
IEEE Trans. Knowl. Data Eng.
(2008)
Parallelizing itinerary-based KNN query processing in wireless sensor networks
IEEE Trans. on Knowl. Data Eng.
kNN Query processing methods in mobile ad hoc networks
IEEE Trans. Mobile Comput.
Distance-Constraint k-Nearest neighbor searching in mobile sensor networks
Sensors
Location-based Spatial Queries
Proc. ACM SIGMOD
Cited by (5)
Drone Helps Privacy: Sky Caching Assisted k-Anonymity in Spatial Querying
2022, IEEE Systems JournalMQry: Elastic Validity Region for Querying Mobile Point-of-Interests in Infrastructure-Less Networks
2020, IEEE Systems Journal