Pervasive and Mobile Computing

It gives us a great pleasure to present you this special issue that contains the extended versions of selected papers 1 presented at the Sixth Annual IEEE International Conference on Pervasive Computing and Communications (PerCom 2008) that 2 was held in Hong Kong, China, March 17–21, 2008. PerCom is a high-quality conference covering a broad spectrum of issues 3 in pervasive computing and communications: from underlying technologies to applications and services. 4 PerCom 2008 received 160 submissions, and following a thorough review process the program committee selected 19 5 regular papers and 6 concise papers, and 10 papers for poster presentation at the conference. Among the 19 regular papers, 6 we selected 7 papers and invited the authors of these papers to submit an extended version of their conference paper for 7 possible publication in the special issue. 8 After an additional in-depth review process, we selected the seven papers contained in this special issue, all of which 9 provide excellent contributions to various aspects of the pervasive computing and communications research: security, 10 privacy, indoor localization, software adaptation, RFID systems, context search for home entertainment and data querying 11 in pervasive computing systems. 12 In the first paper, ‘‘Structured Decomposition of Adaptive Application’’, Justin Mazzola Paluska, Hubert Pham, Umar Saif, 13 Grace Chau, Chris Terman and SteveWard describe an approach to automate certain high-level implementation decisions in 14 a pervasive application. The decisions may be postponed until run time, enabling adaptive applications. In the approach, an 15 application programmer can specify the behavior of an adaptive application as a set of open-ended decision points. Decisions 16 points are formalized as Goals, which may be satisfied by scripts called Techniques. The system provides a framework 17 in which Techniques may compete and interoperate at runtime in order to maintain an adaptive application. The paper 18 presents description of some applications that have been used to test the general applicability of Goals and Techniques. 19 In the second paper, ‘‘A Tamper-Proof and Lightweight Authentication Scheme’’, Ghaith Hammouri, Erdinç Őztűrk, and 20 Berk Sunar present a challenge response scheme that is based on 2-level noisy Physically Unclonable Functions (PUF). The 21 authors show that the scheme is secure against passive attacks provided that it is difficult to learn a threshold of halfspaces 22 under the uniform distribution. 23 ‘‘Shadow Attacks on Users’ Anonymity in Pervasive Computing Environments’’, by Daniele Riboni, Linda Pareschi and 24 Claudio Bettini is the third paper of the special issue. It considers attacks on privacy in mobile and pervasive computing 25 systems and proposes a defense to a type of attack on the notion of k-anonymity, called shadow attacks. K -anonymity is the 26 situation when an issuer of a request cannot be distinguished in a group of at least k potential issuers. The paper defines a 27 shadow attack and proposes a defense, for which its correctness is analyzed. 28 In the fourth paper, ‘‘An Effective Location Fingerprint Model for Wireless Indoor Localization’’, Nattapong Swangmuang 29 and Prashant Krishnamurthy presents a new analyticalmodel for predicting precision and accuracy for an indoor positioning 30 system. Location fingerprinting is a technique that connects location-dependent characteristics, such as received signal 31 strength (RSS) from known access points to a location and uses these characteristics to infer the location. The model applies 32 proximity graphs for approximating the probability of error distance given a fingerprint database using WLANs received 33 signals and its associated statistics. The analysis of the fingerprint structure can be used to modify inefficient location 34 fingerprints within a database. 35 In the fifth paper, ‘‘Data Quality and Query Cost in Pervasive Sensing Systems’’, David Yates, Erich Nahum, James Kurose 36 and Prashant Shenoy examine the costs and benefits of caching data in a wireless sensor system server. The paper shows 37 that for applications that are driven by delay, policies that emulate a cache hit by computing and returning approximate 38 values of a sensor data provide quality improvement and cost savings. In contrast, when data accuracy drives data quality, 39 then a tradeoff between data quality and query costs occurs. The paper measures the benefit and costs of seven different 40 caching and lookup policies as a function of application quality requirements. The paper shows the manner of the tradeoff 41 between costs and quality, and shows a class of policies for which the query rate can be boundedwhen servicing an arbitrary 42 workload of user queries. 43