Elsevier

Information Sciences

Volume 230, 1 May 2013, Pages 78-93
Information Sciences

Location-aware private service discovery in pervasive computing environment

https://doi.org/10.1016/j.ins.2012.08.010Get rights and content

Abstract

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.

Introduction

Pervasive computing environments are characterized by increasing number of distributed devices, non-infrastructure support, and run-time dynamic topology. There is rapidly increasing number of mobile devices/terminals in a variety of different applications. As mobile devices are dynamically connected by wireless network, they do not have a fixed infrastructure but more work in an ad hoc manner [23]. The intelligent devices may need to frequently leave or join a network and incur dynamic network topology [28]. All these emerging properties in a pervasive computing environment pose a challenge on the design of efficient service discovery protocols.

In the last decade, several service discovery protocols have been proposed. These protocols mainly rely on multicast [8] or broadcast [11], [14] mechanisms, which may incur high traffic overhead during service discovery procedure [20]. Some traffic control approaches have been reported to reduce the number of transmitted messages. There are two main traffic control approaches: centralized approaches and distributed approaches. The centralized approach is not suitable for pervasive environment since there is no fixed structure to find a central node and the query service is beyond the capacity of a mobile device. In contrast, distributed approaches may appear to be more suitable for pervasive environments [23]. However, the conventional distributed approaches mainly use either multicast or flooding operations and hence they are not scalable. Scalable service discovery scheme is very important in pervasive computing environments. People usually have a need to find nearby Points of Interest, such as printer, fax, and projector, which they can easily touch. The service provider, which has minimum hop count to the requestor, should have higher priority to access because it is more reliable to use a minimum hop count path [18], [33] in pervasive environments. Thereafter, it becomes necessary to propose and analyze a scalable service discovery protocol which is aware of relative position between relay nodes and does not rely on multicast or flooding mechanisms. The DHTs (Distributed Hash Tables [25], [30]) technique is a scalable solution for distributed data storage. However, it is inappropriate for pervasive environments because the overlay networks does not reflect the physical topology. On the other hand, existing service discovery protocols have ignored node privacy concerns. In these methods, the nodes have to provide their identities during the service discover process. However, nodes may not be willing to reveal its privacy information until it correctly finds the service providing peer [21].

Privacy is a type of state in which the sensitive information of an entity is kept confidential from others. However, in pervasive environment, the users access the information and computational resources anytime and anywhere. Many of the devices and resources are unmanned system, and the services may be added or removed at any time. Hence, without a proper management, it is hard to involve only the necessary user and service provider in service discovery in pervasive systems. In order to reduce the number of irrelevant participants as much as possible, an overlay network is established. To protect user and service provider’s identity information, we use an agent to hide the entire devices in the cluster.

In this paper, we will propose a new Polar Grid based Service Discovery (PGSD) scheme. In the scheme, polar grid description is employed to build location-aware overlay networks, which is able to map the physical network topology. Polar coordinate is a 2-dimensional coordinate system in which each point on a plane is determined by a distance from a fixed point and an angle from a fixed direction. We use the hop counts away from a fixed node as distance. The nodes, which have the same distance around this fixed node, split the 360° equally to get angle property. Using directional controlled routing method, the network overload can be reduced [7]. To overcome keyword match having low quality discovery result [35], we use semantic service description. The PGSD using semantic-based service description method builds a topology-dependent structure without GPS information. We use semantic category tree to define service ID in order to classify similar services in one branch of the tree. The similar services will then have similar DHT key. The requestor also uses this category tree to generate query DHT key, which will find more useful services since the similar services are under the same branch of the category tree. Clustering strategies are incorporated in our service discovery protocol to localize traffic and hide user’s identity.

In addition, we present extensive numerical examples to demonstrate the advantages of the proposed scheme compared to the existing schemes and to show the determination of the crucial parameters. Numerical results indicate that our proposed scheme is able to achieve considerably higher discovery success ratio and lower cost, comparing to existing mechanisms.

The paper is organized as follows: Section 2 discusses the related work. Then, Section 3 describes the semantic-based service description method. The topology construction and search algorithm is described in detail in Section 4. Section 5 presents the results of performance analyses and simulations. Conclusions are drawn in Section 6.

Section snippets

Existing approaches for service discovery

The traditional approaches for service discovery in pervasive environments can be classified into two categories: centralized protocols and distributed protocols. The centralized protocols include Jini [3], UDDI [4], and Salutation [15]. Their operation needs a central lookup server. On the other hand, the distributed protocols rely on multicast or flooding, including UPnP [26] and Konark [13]. The existing distributed protocols can be further partitioned into two groups: structured and

Semantic-based service description

As there are hundreds of types of services, some works [6] use the class/subClassOf axiom of Web Ontology Language (OWL) to structure a general category classification. We add the location information on the ontology tree in order to support location-based searching.

The ontology tree gives two important properties of a service to show category information: service classification information and service position information. Position information enriches the query mode and reduces the time on

Proposed polar grid based service discovery protocol

In this section, we present a secure and scalable service discovery scheme that is able to efficiently reduce the traffic load in pervasive environments. We propose a new mechanism that constructs location-aware overlay networks using polar coordinate to avoid P2P overlay network mismatching problem. In order to enhance security, mobile nodes broadcast their requests to cluster heads without their own identity and those requests is encrypted by hash function when they are transmitted in the

Numerical results

In this section, our main objectives are to demonstrate the characteristics of the performance metric and their different behaviors in different topologies. The interaction between the performance and key parameters will be also explored. If not specified, we will use the parameters in Table 1 to simulate our protocol with NS-2 [39]. In the simulation, we randomly choose a node as a backup register node. We evaluate the performance with respect to the metrics message overhead, lookup time, and

Conclusions and future work

In this paper, we have proposed a new polar grid based service discovery protocol. The protocol constructs a private location-aware overlay networks for large-scale mobile networks. The overlay network is able to reflect the physical network and hence generates low message overhead. The similar service resources can be mapped to the DHT space which can be found more easily using semantic description method. Our results showed that PGSD scales very well with increasing traffic load, network

Acknowledgements

The work is partly supported by NSFC (No. 61003220), RFDP (No. 20090142120025), Fundamental Research Funds for the Central Universities (HUST:2010QN051) and Natural Science Foundation of Hubei Province of China (No. 2010CDB02302).

References (39)

  • D. Chakraborty et al.

    Toward distributed service discovery in pervasive computing environments

    IEEE Transactions on Mobile Computing

    (2006)
  • M. Chen et al.

    Directional controlled fusion in wireless sensor networks

    Mobile Networks and Applications

    (2009)
  • S. Cheshire et al.

    {DNS}-based Service Discovery

    (2011)
  • B. Chor et al.

    Private information retrieval

  • S.E. Czerwinski et al.

    An architecture for a secure service discovery service

  • R. Deepa et al.

    A secure and lightweight service discovery model for mobile ad hoc networks

  • W.R. Heinzelman et al.

    Energy-efficient communication protocol for wireless microsensor networks

  • S. Helal et al.

    Konark – a service discovery and delivery protocol for ad-hoc networks

  • H. Hof et al.

    Design of a secure distributed service directory for wireless sensor networks

    Wireless Sensor Networks

    (2004)
  • Cited by (17)

    • Fog node discovery and selection: A Systematic literature review

      2022, Future Generation Computer Systems
      Citation Excerpt :

      Thus, it should be an important criterion to consider in FND. Furthermore, while location-aware mechanisms have been studied widely in pervasive computing environments [78–80] and content-aware mechanisms have been discussed in Web services [81], they have not been applied to the different approaches for FND. One possible option to address this gap is to approach this as XaaS-FND, in which goal-specific services are developed that assist the edge devices in evaluating the available fog nodes according to the location, context and content of their requirements (meeting requirements FND1, FND2 and FND4).

    • Small-world: Secure friend matching over physical world and social networks

      2017, Information Sciences
      Citation Excerpt :

      From this point of view, Proximity-based Mobile Social Networks (PMSNs), which employ some short range communication techniques such as Bluetooth or WiFi to communicate with others freely and thus provide mobile users a novel way to enjoy these services and applications while reducing the overhead to the infrastructure. Since mobile users in most PMSNs-related services [27,29] and applications [18,19,23,28] need to share their personal data as the first step and employ profile matching-based approaches to discover and make new social interactions in vicinity, the privacy concerns increase sharply since there are always some sensitive information in the profile, such as user’s home address, symptom, religious faith, sexual orientation, etc. The leakage on such information may cause serious problems [13] like social reputation or economic damage, and make them victims of blackmail or even physical violence.

    • Bayesian resource discovery in infrastructure-less networks

      2017, Information Sciences
      Citation Excerpt :

      The performance evaluation of the above scheme, together with the analysis of the associated experimental results, are presented in Section 6, followed by the conclusion reported in Section 7. Resource discovery is a well-known and widely explored topic in distributed systems literature [7,14,21,30,32,34]. Many specific approaches, properly crafted for mobile ad hoc environments, have been proposed, often coupled with a companion routing facility.

    • Robust and affordable retail customer profiling by vision and radio beacon sensor fusion

      2016, Pattern Recognition Letters
      Citation Excerpt :

      Finally, we present conclusion in Section 6, arguing novelties introduced by our contribution and future developments. Localization and people tracking in an indoor environment (e.g., stores, museums, hospitals, public and private buildings in general) is an interesting challenge for AmI applications (e.g. for people with disabilities [6,37,56] and for indoor monitoring of elder people [43]). If the location of monitored subjects is well known, a smart system can provide them with correct and proper information about the surrounding environment.

    • Authentication scheme for mobile OFDM based on security information technology of physical layer over time-variant and multipath fading channels

      2015, Information Sciences
      Citation Excerpt :

      Such sets of numbers decorrelate from one transmit-receive path to another if the paths are separated by the order of an RF wavelength or more. A new service discovery scheme [26], considering both discovery efficiency and privacy protection, is proposed to build a location-aware overlay network. The PHY property of the wireless medium is a powerful source of domain-specific information that can be used to complement and enhance traditional security mechanism.

    View all citing articles on Scopus
    View full text