February 2011 Wireless Cover 1 2/7/11 12:20 PM Page 1 • IMS EMERGENCY SERVICES: A PRELIMINARY STUDY IEEE February 2011, Vol. 18, No. 1 • MIXING NETWORK CODING AND COOPERATION FOR RELIABLE WIRELESS COMMUNICATIONS • SPECTRUM SENSING FOR COGNITIVE RADIO SYSTEMS • COOPERATIVE COMMUNICATION IN MULTIHOP COGNITIVE RADIO NETWORKS WIRELESS BASED ON MULTICARRIER MODULATION COMMUNICATIONS • TOPOLOGICAL-BASED ARCHITECTURES FOR WIRELESS MESH NETWORK IMS EMERGENCY SERVICES ® A Publication of the IEEE Communications Society In cooperation with IEEE Computer and VehicularTechnology Societies LYT-TOC-Feb 2/7/11 11:41 AM Page 1 FEBRUARY 2011/VOL . 18, NO. 1 IEEE WIRELESS COMMUNICATIONS ACCEPTED FROM OPEN CALL 46 CHALLENGES, OPPORTUNITIES, AND 6 SOLUTIONS FOR CONVERGED SATELLITE AND IMS EMERGENCY SERVICES: TERRESTRIAL NETWORKS A PRELIMINARY STUDY TARIK TALEB, YASSINE HADJADJ-AOUL, AND TOUFIK AHMED YI-BING LIN, MENG-HSUN TSAI, AND YUAN-KUANG TU 54 15 INTERFERENCE COORDINATION FOR MIXING NETWORK OFDM-BASED MULTIHOP CODING AND COOPERATION FOR LTE-ADVANCED NETWORKS RELIABLE WIRELESS COMMUNICATIONS KAN ZHENG, BIN FAN, JIANHUA LIU, YICHENG LIN, FRANCESCO ROSSETTO AND MICHELE ZORZI AND WENBO WANG 22 64 SECURING UNDERWATER DISTRIBUTED AUTOMATED WIRELESS COMMUNICATION NETWORKS INCIDENT DETECTION WITH VGRID MARI CARMEN DOMINGO BEHROOZ KHORASHADI, FRED LIU, DIPAK GHOSAL, 30 MICHAEL ZHANG, AND CHEN-NEE CHUAH SPECTRUM SENSING FOR 74 COGNITIVE RADIO SYSTEMS: TOPOLOGICAL-BASED ARCHITECTURES FOR TECHNICAL ASPECTS AND WIRELESS MESH NETWORKS STANDARDIZATION ACTIVITIES OF THE AMIR ESMAILPOUR, NIDAL NASSER, AND TARIK TALEB IEEE P1900.6 WORKING GROUP 82 KLAUS MOESSNER, HIROSHI HARADA, CHEN SUN, YOHANNES D. ALEMSEGED, HA NGUYEN TRAN, SYNCHRONIZATION OF DOMINIQUE NOGUET, RYO SAWAI, AND NAOTAKA SATO MULTIHOP WIRELESS SENSOR NETWORKS AT THE APPLICATION LAYER 38 ÁLVARO MARCO, ROBERTO CASAS, JOSÉ LUIS SEVILLANO MULTICARRIER MODULATION AND RAMOS, VICTORIAN COARASA, ÁNGEL ASENSIO, COOPERATIVE COMMUNICATION IN AND MOHAMMAD S. OBAIDAT MULTIHOP COGNITIVE RADIO NETWORKS TAO LUO, FEI LIN, TAO JIANG, MOHSEN GUIZANI, AND WEN CHEN MESSAGE FROM THE EDITOR-IN-CHIEF — 2 SCANNING THE LITERATURE — 4 Cover image: Getty Images IEEE Wireless Communications • February 2011 1 LYT-EDIT NOTE-February 2/7/11 10:39 AM Page 2 MESSAGE FROM THE ED I TOR- IN -CH I E F CALL FOR MORE WIRELESS INNOVATIONS irst of all, I am happy to report that my munications and networking, and decision mak- term as the Editor-in-Chief has been ing and control. The second technology on the F extended for one more year. I also want top 11 list is social networking, which has close to report that we have been working on ties to wireless technologies as most users have moving into Manuscript Central, which will already been doing their social networking with be a great step toward a more efficient smartphones or mobile devices, leading to the paper handling process. As a new year ever more popular mobile social networking. As starts, we also look forward to more excit- we can see, wireless technologies have already ing news on wireless communications, net- revolutionized the way we live and have trans- working, and their applications. You may formed our society into a completely different have already read the first issue of IEEE kind for everything imaginable. We, the engi- Spectrum this year. Among the top 11 iden- neers and scientists, are at center stage to uti- tified technologies (although it is not clear lize the technologies we have created to make a why it is 11 rather than the more commonly significantly different society, good or bad, and used 10, perhaps to avoid a copyright law- it is up to us to reshape it and our living envi- suit from David Letterman?), smartphones ronments for better quality of life. We need you are ranked on top. Unfortunately, the col- YUGUANG MICHAEL FANG to invent new technologies, including wireless umn does not spell out the drivers of future technologies, and we need new ideas and new smartphone technologies: the deep integration technologies of innovations from you. Our magazine offers you the platform smartphones with smart environments and human interactions for you to spread the word. We are searching for articles on for ubiquitous data collection, processing, data mining, com- new innovations in the wireless area and solicit new special IEEE Director of Magazines IEEE Wireless Communications (ISSN 1536-1284) is pub- A Yndurgzueaj nJga jsMzcizcyhkEa,ed Ali tFGoaHrn-g Ui,n U.- oCnfih vSi.c eoif. f& F lToercidh.a P, UolaSnAd WCOIMRMUENILCAETISONSS lE12i71snht2ghe-i7 dnF0 elb5eoi-rmo8sr9,o ,9 InN9nt;che .wl eyH- mYbeyoa ridTklq:h, ueNj.a pIYrnote sr1rtci0set LYT-EDIT NOTE-February 2/7/11 10:39 AM Page 3 MESSAGE FROM THE ED I TOR- IN -CH I E F issues on hot topics that are of great interest to our reader- Bin Fan, Yicheng Lin, and Wenbo Wang, presents an ship. Some of the topics we are particularly interested in are overview of the interference coordination strategies for wireless technologies in cyber-physical systems, social net- OFDM-based multihop cellular networks and proposes sever- works, smart grids, healthcare and medical systems, urban al static or semi-static interference coordination schemes sensing and public safety, and cloud computing systems. With based on the framework of Long Term Evolution (LTE)- the upcoming electronic paper handling process, I hope we Advanced networks with multihop relaying to improve cover- can make the publication process much faster. For those who age and increase the data rate over cell edge areas. are engaging in hot wireless innovations, I encourage you to “Distributed Automated Incident Detection with Vgrid,” organize special issues to share your excitement with our gen- by Behrooz Khorashadi, Fred Liu, Dipak Ghosal, Chen-Nee eral audiences. Chuah, and Michael Zhang, studies an ad hoc distributed Due to the unforeseen delay in getting articles for the automated incident detection algorithm for highway traffic planned special issue, we have decided to accommodate arti- using vehicles that are equipped with wireless communica- cles from our open call to cut the long queue of accepted tions, processing, and storage capabilities. By requesting vehi- papers. We have selected 10 articles in this issue; their brief cles with such capability to broadcast beacon information summaries follow. containing their speed, location, and lane information, the “IMS Emergency Services: A Preliminary Study,” by Yi- detection algorithm can make better decisions and yield bet- Bing Lin, Meng-Hsun Tsai and Yuan-Kuang Tu, presents a ter performance. study on how to support emergency call and walkie-talkie ser- “Topological-Based Architectures for Wireless Mesh Net- vices over IP Multimedia Subsystem in wireless cellular sys- work,” by Amir Esmailpour, Nidal Nasser, and Tarik Taleb, tems. provides an overview on architectural design for wireless mesh “Mixing Network Coding and Cooperation for Reliable networks, summarizes the state-of-the-art research findings, Wireless Communications,” by Francesco Rossetto and and calls for further research on this topic. Michele Zorzi, gives an overview of how to take advantage of “Synchronization of Multihop Wireless Sensor Networks at both cooperation and network coding to improve the perfor- the Application Layer,” by Álvaro Marco, Roberto Casas, mance and error correction capabilities of radio networks and José Luis Sevillano, Victorián Coarasa, Ángel Asensio, and highlights the main challenges for future research. Mohammad S. Obaidat, proposes a method for accurate syn- “Securing Underwater Wireless Communication Net- chronization of large multihop networks, which operates at works,” by Mari Carmen Domingo, reviews some important the application layer while minimizing message exchange. security design issues specific to underwater wireless commu- I hope you enjoy reading these articles. I also wish you a nication networks and discusses possible research challenges. productive 2011! “Spectrum Sensing for Cognitive Radio Systems: Technical Aspects and Standardization Activities of IEEE 1900.6 Work- BIOGRAPHY ing Group,” by Klaus Moessner, Hiroshi Harada, Chen Sun, YUGUANG MICHAEL FANG [F’08] ( LYT-SCANNING-February 2/7/11 10:40 AM Page 4 SCANN ING TH E L I T ERATU R E EDITED BY YANCHAO ZHANG Using Classification to Protect the input and output data traces used to con- ing reliability of the pulse oximeters struct it may be private data of partici- (median 80.85 percent, range 0.46–97.69 Integrity of Spectrum Measure- pants (e.g., their individual food intake, percent). Sensing failures usually ments in White Space Networks lifestyle choices, and resulting weight). occurred in short bursts, although longer O. Fatemieh, A. Farhadi, R. Chandra, and This paper proposes and experimentally periods were also present due to sensor C. Gunter, in the 18th Annual Network studies a technique that attempts to keep disconnections. The authors show that & Distributed System Security Sympo- such input and output data traces pri- the sensing reliability could be signifi- sium (NDSS), San Diego, CA, February vate, while allowing accurate model con- cantly improved through oversampling 2010 struction. This is significantly different and by implementing a disconnection from perturbation-based techniques in alarm system that incurs minimal inter- The emerging paradigm for using the that no noise is added. The main contri- vention cost. A retrospective data analy- wireless spectrum more efficiently is bution of the paper is to show a certain sis indicated that the system provided based on enabling secondary users to data transformation at the client side sufficient temporal resolution to support exploit white space frequencies that are that helps keeping the client data private the detection of clinical deterioration in not occupied by primary users. A key while not introducing any additional three patients who suffered from signifi- enabling technology for forming net- error to model construction. The authors cant clinical events including transfer to works over white spaces is distributed particularly focus on linear regression intensive care units. spectrum measurements to identify and models which are widely used in partici- assess the quality of unused channels. patory sensing applications. They use the CodeOn: Cooperative Popular Content This spectrum availability data is often data set from a map-based participatory aggregated at a central base station or sensing service to evaluate their scheme. Distribution for Vehicular Networks database to govern the usage of spec- The service in question is a green naviga- Using Symbol Level Network Cod- trum. This process is vulnerable to tion service that constructs regression ing integrity violations if the devices are models from participant data to predict malicious and misreport spectrum sens- the fuel consumption of vehicles on road M. Li, Z. Yang, and W. Lou, IEEE Jour- ing results. This paper presents CUSP, a segments. They evaluate the proposed nal on Selected Areas in Communica- new technique based on machine learn- mechanism by providing empirical evi- tions (JSAC), special issue on Vehicular ing that uses a trusted initial set of signal dence that: i) an individual data trace is Communication Networks, vol. 29, no. 1, propagation data in a region as input to generally hard to reconstruct with any January 2011 build a classifier using support vector reasonable accuracy, and ii) the regres- Driven by both safety concerns and com- machines. The classifier is subsequently sion model constructed using the trans- mercial interests, one of the key services used to detect integrity violations. Using formed traces has a much smaller error offered by vehicular networks is popular classification eliminates the need for than one based on additive data-pertur- content distribution (PCD). The funda- arbitrary assumptions about signal propa- bation schemes. mental challenges to achieve high speed gation models and parameters or thresh- content downloading come from the olds in favor of direct training data. highly dynamic topology of vehicular ad Reliable Clinical Monitoring Using Wire- Extensive evaluations using TV transmit- hoc network (VANET) and the lossy ter data from the FCC, terrain data from less Sensor Networks: Experiences nature of the vehicular wireless commu- NASA, and house density data from the In A Step-Down Hospital Unit nications. This paper introduces U.S. Census Bureau for areas in Illinois CodeOn, a novel push-based PCD O. Chipara1, C. Lu, T. Bailey, and G. and Pennsylvania show that CUSP is scheme where contents are actively Roman, in the 8th ACM Conference on effective against attackers of varying broadcasted to vehicles from road side Embedded Networked Sensor Systems sophistication, while accommodating access points and further distributed (SenSys), Zurich, Switzeland, Novem- regional terrain and shadowing diversity. among vehicles using a cooperative ber 2010 VANET. In CodeOn, we employ a This paper presents the design, deploy- recent technique, symbol level network Privacy-Preserving Regression ment, and empirical study of a wireless coding (SLNC), to combat the lossy Modeling of Participatory Sensing Data clinical monitoring system that collects wireless transmissions. Through exploit- H. Ahmadi, N. Pham, R. Ganti, T. pulse and oxygen saturation readings ing symbol level diversity, SLNC is Abdelzaher, S. Nath, and J. Han, in from patients. The primary contribution robust to transmission errors and the 8th ACM Conference on Embedded of this paper is an in-depth clinical trial encourages more aggressive concurrent Networked Sensor Systems (SenSys), that assesses the feasibility of wireless transmissions. In order to fully enjoy the Zurich, Switzeland, November 2010 sensor networks for patient monitoring benefits of SLNC, we propose a suite of in general hospital units. The authors techniques to maximize the downloading Many participatory sensing applications present a detailed analysis of the system rate, including a prioritized and localized use data collected by participants to con- reliability from a long-term hospital relay selection mechanism where the struct a public model of a system or phe- deployment over seven months involving selection criteria are based on the use- nomenon. For example, a health 41 patients in a step-down cardiology fulness of vehicle-possessed contents, application might compute a model relat- unit. The network achieved high reliabil- and a lightweight medium access proto- ing exercise and diet to amount of weight ity (median 99.68 percent, range col that naturally exploits the abundant loss. While the ultimately computed 95.21–100 percent). The overall reliabili- concurrent transmission opportunities. model could be public, the individual ty of the system was dominated by sens- We also propose additional mechanisms 4 IEEE Wireless Communications • February 2011 LYT-SCANNING-February 2/7/11 10:40 AM Page 5 SCANN ING TH E L I T ERATU R E to reduce the protocol overhead without tive throughput in the original transmis- transient contact with APs, the authors sacrificing the performance. sion. As such, an extra and “free” coor- proactively push content to the APs the dination channel can be built. vehicles are likely to visit in the near Free Side Channel: Bits over Inter- Based on this idea, the authors pro- future. In this way, vehicles can enjoy pose a DC-MAC to leverage this “free” the full wireless capacity instead of ference channel for efficient medium access in being bottlenecked by Internet connec- K. Wu, H. Tan, Y. Liu, J. Zhang, Q. Zhang, a multiple-user wireless network. They tivity, which is either slow or even and L. Ni, in the 16th Annual International theoretically analyze the capacity of this unavailable. The authors develop a new Conference on Mobile Computing and channel under different environments algorithm for predicting the APs that Networking (MobiCom), Chicago, Illinois, with various modulation schemes. will soon be visited by the vehicles. They September 2010 then develop a replication scheme that Enabling High-Bandwidth Vehicular leverages the synergy among (i) Internet Interference is a critical issue in wire- connectivity (which is persistent but has less communications. In a typical multi- Content Distribution limited coverage and low bandwidth), ple-user environment, different users U. Shevade, Y. Chen, L. Qiu, Y. Zhang, (ii) local wireless connectivity (which may severely interfere with each other. V. Chandar, M. Han, H. Song, and Y. has high bandwidth but transient dura- Coordination among users therefore is Seung, in the 6th International Conference tion), (iii) vehicular relay connectivity an indispensable part of interference on emerging Networking EXperiments and (which has high bandwidth but high management in wireless networks. It is Technologies (CoNEXT), Philadelphia, PA, delay), and (iv) mesh connectivity known that coordination among multi- November 2010 among APs (which has high bandwidth ple nodes is a costly operation, taking a but low coverage). The authors demon- significant amount of valuable commu- This paper presents VCD, a novel sys- strate the effectiveness of the VCD sys- nication resources. In this paper, the tem for enabling high-bandwidth con- tem using trace-driven simulation and authors have an interesting observation tent distribution in vehicular networks. Emulab emulation based on real taxi that by generating intended patterns, In VCD, a vehicle opportunistically traces. They further deploy VCD in two some simultaneous transmissions (i.e., communicates with nearby access points vehicular networks, one using 802.11b “interference”) can be successfully (APs) to download the content of inter- and the other using 802.11n, to demon- decoded without degrading the effec- est. To fully take advantage of such strate its effectiveness. Open Cal l f rom the IEEE Communications Society re you enthusiastic? Have you per formed quality reviews Afor technical periodicals? Demonstrated solid technical accomplishments? Have a reputation for high ethical standards and for rel iabil i ty? You may be ready . . . The IEEE Communicat ions Society is looking for volunteers who are interested in becoming par t of a prest ig ious Communicat ions Society editor ia l board. Duties include: A commitment to handle at least two manuscr ipts per month; arrange for three reviews or more in a t imely fashion; and the abi l i ty to make firm and fai r decis ions. Qualifications: Subject matter exper t ise, edit ing exper ience, technical program committee exper ience; references, representat ive papers . Apply at : www.comsoc.org/editor The decis ion to appoint an editor rests with the Editor- in-Chief of the journal/magazine. Please note that i t wi l l not be possible to send indiv idual acknowledgments to al l appl icants . IEEE Wireless Communications • February 2011 5 www.comsoc.org/editor LIN LAYOUT 2/7/11 12:04 PM Page 6 ACCEPTED FR O M OPEN CALL IMS EMERGENCY SERVICES: A PRELIMINARY STUDY YI-BING LIN, NATIONAL CHIAO TUNG UNIVERSITY MENG-HSUN TSAI, NATIONAL CHENG KUNG UNIVERSITY YUAN-KUANG TU, CHUNGHWA TELECOM ABSTRACT trols the connection for media channels in a media gateway (MGW, 4, Fig. 1). The MGW Emergency call and walkie-talkie are two ser- connects toward the legacy fixed and mobile net- Fixed vices utilized in emergency situations. During works to provide user data transport. The home network Typhoon Morakot in 2009, we experienced the subscriber server (HSS, 2, Fig. 1) is the master deficiency of emergency call service that cannot database containing all user-related mobile sub- continuously track callers in real time and scription and location information. In 2004 (3) MGC walkie-talkie communications where a speaker Chunghwa Telecom deployed the first commer- may not be granted the permission to talk. These cial IMS in Taiwan to provision commercial issues can be resolved by the emergency call and telecommunications services such as voice, video, GW push-to-talk over cellular services in IP Multime- and Internet-based multimedia services. The ini- dia Subsystem. This article conducts a prelimi- tial capacity was 125,000 subscribers. Current nary study on how these two services can be IMS capacity can accommodate about 500,000 effectively exercised in IMS. subscribers in daily commercial operation. During Typhoon Morakot in August 2009, INTRODUCTION Taiwan experienced serious damage from flood- ing and mudslides (Fig. 2, left), and the rescue Deficiencies in IP Multimedia Subsystem (IMS) supports IP- missions solely relied on GSM and satellite com- based multimedia services. IMS was originally munications that offer basic services (Fig. 2, right) communications designed by the Third Generation Partnership such as emergency call without location tracking. Project (3GPP) to deliver Internet services over From Typhoon Morakot, we learned that it is during emergencies general packet radio service (GPRS) in 3G net- desirable to accommodate emergency services in can be resolved by works such as Universal Mobile Telecommunica- IMS with a 3G network including emergency call tions System (UMTS). IMS was later updated to and push-to-talk over cellular (PoC). the emergency call support other access networks including wireless A GSM user in Taiwan can make an emer- LAN, CDMA2000, and fixed line. For the pur- gency call by dialing 110, 112, or 119. However, and the Push-to-talk pose of this article, Fig. 1 illustrates a simplified the existing GSM emergency call service only IMS network architecture (the reader is referred identifies the location of the caller at the time of over Cellular services to [1] for detailed descriptions of IMS). call setup and does not track the user’s location The IMS (Fig. 1b) connects to both mobile during the call. In Typhoon Morakot many peo- in IP Multimedia and fixed telecommunications networks (Fig. 1a) ple waiting for rescue could not be accurately Subsystem network. for fixed mobile convergence (FMC). IMS is not located through their phone calls, which intended to standardize applications or services. increased the difficulty of the rescue missions. The authors conduct Instead, it provides a standard approach for PoC is a walkie-talkie-like service defined in voice/multimedia application access from user Open Mobile Alliance (OMA) specifications [3, a preliminary study equipment (UE; 1, Fig. 1) in wireline and wire- 4]. In 2004 Chunghwa Telecom first launched less networks. This goal is achieved by having this service in Asia using 2.5G technology. Our on how these two horizontal control that isolates the access net- experience with 2.5G PoC included long PoC works from the service and application networks call setup time (13 s) and handoff time (2.5–3 s services can be (Fig. 1c). for reconnecting a PoC client when it moved effectively exercised In the IMS, the transport of user data is sepa- from one base station to another). These prob- rated from that for control signals, where IETF lems have been resolved by IMS-based PoC in IMS. protocols such as Session Initiation Protocol established on 3G networks. For example, the (SIP) [2] are used to ease the integration with 3G PoC call setup time is less than 6 s. Although the Internet. For example, the call session con- it is not clear if PoC will be a successful residen- trol function (CSCF, 5, Fig. 1) is a SIP server, tial service, it has proven useful for business cor- which is responsible for call control. The media porations and government organizations such as gateway control function (MGCF, 3, Fig. 1) con- the National Security Bureau in Taiwan. 6 1536-1284/11/$25.00 © 2011 IEEE IEEE Wireless Communications • February 2011 LIN LAYOUT 2/7/11 12:04 PM Page 7 In the past three years, we have studied emer- gency call [5] and PoC [6]. During Typhoon Morakot in 2009, emergency calls and walkie- talkies were important means of communications (a) (1) UE in rescue missions. Clearly, it is desirable to sup- port emergency services in IMS for disastrous Fixed Mobile network events. Therefore, this article conducts a prelimi- network nary investigation on IMS emergency call and PoC. (b) (2) HSS LOCATION TRACKING FOR EMERGENCY CALL (3) MGCF An important feature of emergency call is that the system can track the location of a calling UE (4) MGW (1, Fig. 1) during the conversation. To support IMS emergency call, three network nodes are deployed. When the UE originates an emergen- (5) CSCF cy call, the call is established by a special CSCF called an emergency-CSCF (E-CSCF, 5, Fig. 1), which dispatches the call to the nearest public (c) safety answering point (PSAP) based on the location information of the UE. The PSAP (7, (8) Fire department Fig. 1) is an IMS application server that process- (9) PoC server es emergency calls according to the types of Police department emergency events. For example, in a fire event (7) PSAP (6) GMLC the PSAP connects the UE (the caller) to the fire department (8, Fig. 1). The PSAP interacts Figure 1. A simplified IMS network architecture (dashed lines: control signal- with the E-CSCF by using SIP, and the voice ing; solid lines: user data/control signaling): a) fixed and mobile telecom net- conversation path is set up through the MGW works; b) IP Multimedia Subsystem; c) service and application networks. (4, Fig. 1) to the UE by using Real-Time Trans- port Protocol (RTP) [7]. The gateway mobile location center (GMLC, 6, Fig. 1) supports a location service (LCS) [8]. Through Signaling resolve these issues, the active location reporting System Number 7 (SS7) Mobile Application Part scheme was proposed in [5]. This scheme reports (MAP) [1], the GMLC interacts with the HSS the UE’s location upon change of its SA. This and mobile network to obtain the accurate loca- section describes location polling and active tion of UE. The GMLC provides the location location reporting, and comments on their per- information to the PSAP and E-CSCF. formance. The LCS merits further discussion. This ser- vice utilizes one or more positioning methods EMERGENCY CALL SETUP between the mobile network (Fig. 1a) and UE to Figure 3 illustrates IMS emergency call setup determine the location of the UE [9]. The cell- message flow with the following steps [10]: ID-based positioning method determines the Step A.1 The UE establishes IP connectivity to UE’s position based on the coverage of service the IMS through the mobile network [1]. areas (SAs). An SA includes one or more cells Step A.2 The UE sends a SIP INVITE message (base stations). The observed time difference of to the E-CSCF. This message includes the arrival (OTDOA) and uplink time difference of supported positioning methods of the UE arrival (U-TDOA) positioning methods utilize (cell-ID-based in our example). trilateration to determine the UE’s position Step A.3 The E-CSCF uses the received infor- based on the time differences between downlink mation to select a GMLC and sends the Emer- and uplink signal arrivals, respectively. The gency Location Request message to the Assisted Global Positioning System (A-GPS) GMLC. method speeds up GPS positioning by down- Steps A.4 and A.5 The GMLC exchanges the SS7 loading GPS information through the mobile MAP_SEND_ROUTING_INFO_FOR_LCS and network. In Chunghwa Telecom A-GPS is uti- acknowledgment message pair with the HSS lized for location-based services. to identify the mobile network node responsi- Without loss of generality, we consider the ble for connection to the UE. In UMTS this cell-ID-based method. After emergency call node is a serving GPRS support node (SGSN). setup, the PSAP may need to monitor the UE’s Step A.6 The GMLC sends the SS7 MAP_PRO- location in real time. In the 3GPP 23.167 specifi- VIDE_SUBSCRIBER_LOCATION message. cation [10], the UE’s location is tracked through Step A.7 The mobile network and UE exercise polling, where the PSAP periodically queries the the cell-ID-based positioning procedure to UE’s location. For description purposes, we obtain the location estimate information of refer to the 3GPP 23.167 approach as the loca- the UE (i.e., the SA identity of the UE). tion polling scheme. In this scheme, if the UE Step A.8 The mobile network returns the SA iden- does not change its location between two tity to the GMLC through SS7 MAP_PRO- queries, the second query is wasted (this is called VIDE_SUBSCRIBER_LOCATION_ack message. redundant polling). On the other hand, if the UE Step A.9 The GMLC selects a suitable PSAP has visited several locations between two loca- according to the SA of the UE and replies the tion queries, the PSAP may lose track of the UE Emergency Location Response message (with in this time period (this is called mistracking). To the selected PSAP address) to the E-CSCF. IEEE Wireless Communications • February 2011 7 LIN LAYOUT 2/7/11 12:04 PM Page 8 Redundant polling creates extra network traffic without providing useful location infor- mation. Furthermore, mis-tracking may result in wrong positioning in case of emergency Figure 2. Telecommunications services in Typhoon Morakot: (left) deploying temporary cables in flooded situations. These areas; (right) GSM/satellite communications through a vehicular base station. issues are resolved by Active Location Steps A.10–A.12 The E-CSCF forwards the SIP need to query the HSS to identify the mobile INVITE to the PSAP to set up the call. The network node responsible for connection to the Reporting. PSAP and the UE exchange the 200 OK and UE (i.e., steps B.2 and B.3 are eliminated). The the SIP ACK messages through the E-CSCF. active location reporting scheme is illustrated in After the PSAP has received the SIP ACK Fig. 5 with the following steps: message, the emergency call is established. Step C.1 When the UE moves to a new SA, the Step A.13 The GMLC sends the location infor- mobile network detects this movement at the mation obtained at step A.8 to the PSAP after cell tracking mode and then triggers the posi- the call has been established, where the PSAP tioning procedure. address is resolved by the GMLC at step A.9. Step C.2 After the positioning procedure is exe- cuted, the UE’s SA identity is obtained. LOCATION POLLING Step C.3 The mobile network sends the SS7 UE may move during an emergency call, and the MAP_SUBSCRIBER_LOCATION_REPORT PSAP needs to monitor the UE’s location in real message with the SA identity to the GMLC. time. In location polling, the PSAP periodically Step C.4 From the UE-PSAP mapping table, the queries the UE’s location. In each query, the fol- GMLC retrieves the PSAP address of the UE lowing steps are executed (Fig. 4) [8]: stored at step A.9 and then sends the updated Step B.1 The PSAP sends the Location Informa- location information to the PSAP. tion Request message to the GMLC. When the emergency call is terminated, the Steps B.2–B.6 These steps retrieve the UE’s SA following steps are executed: identity, which are similar to steps A.4–A.8 in Step C.5 When the IMS call is released, the UE Fig. 3. moves from the cell-connected mode to the Step B.7 The GMLC returns the SA identity of idle mode, and the mobile network no longer the UE to the PSAP. tracks the movement of the UE. Steps B.8–B.10 When the emergency call is Step C.6 The E-CSCF sends the Emergency terminated, the E-CSCF exchanges the Emer- Location Release message to the GMLC to gency Location Release and Response mes- terminate location tracking. sage pair with the GMLC to terminate Step C.7 The GMLC returns the Emergency location tracking. Location Response message to the E-CSCF and then deletes the (UE, PSAP) mapping ACTIVE LOCATION REPORTING from the UE-PSAP table. To resolve redundant polling and mistracking Note that steps C.1 and C.2 in active location issues in location polling, the active location reporting automatically detects the movement of reporting scheme was proposed in [5], which UE, which is different from steps B.1–B.5 in reports the UE’s location upon change of its SA. location polling. This scheme introduces a new locationEstimate- Type initiateActiveReport (to trigger active loca- PRELIMINARY PERFORMANCE EVALUATION tion reporting) in the MAP_PROVIDE_ It is clear that redundant polling creates extra SUBSCRIBER_LOCATION message (at step A.6). network traffic without providing useful location Since the IP connectivity exists during the IMS information. Furthermore, mistracking may emergency call, the UE is in the cell-connected result in wrong positioning in case of emergency state and is tracked by the mobile network at the situations. These issues are resolved by active cell level [1]. Therefore, the mobile network can location reporting. However, it is desirable to detect when the UE moves from one base station evaluate the performance of location polling to to another, and report the new SA identity to the justify the modifications to the existing location GMLC. In this approach the GMLC maintains a tracking procedure in active location reporting. UE-PSAP mapping table to store the (UE, Suppose that the SA residence time has a PSAP) pair at step A.9. The GMLC does not Gamma distribution with mean 1/μ and variance 8 IEEE Wireless Communications • February 2011 LIN LAYOUT 2/7/11 12:04 PM Page 9 In the PoC service several predefined Mobile network group members participate in one PSAP GMLC E-CSCF HSS UE PoC session. Since A.1. IMS connection PoC utilizes establishment A.2. SIP INVITE half-duplex A.3. Emergency Location Request communications, only one PoC Report initial location member speaks A.4. MAP_SEND_ROUTING_INFO_FOR_LCS at a time, A.5. MAP_SEND_ROUTING_INFO_FOR_LCS_ack and others listen. A.6. MAP_PROVIDE-SUBSCRIBER_LOCATION A.7. Positioning A.8. MAP_PROVIDE-SUBSCRIBER_LOCATION_ack A.9. Emergency location response Establish emergency call A.10. SIP INVITE A.11. 200 OK A.11. 200 OK A.12. SIP ACK A.12. SIP ACK A.13. Location information Figure 3. IMS emergency call setup. Vm (other distributions have shown similar es. Similarly, when λ is large (e.g., λ ≥ 5 μ), results [5]). The inter-query interval is a fixed effect 1 is significant, and β is a decreasing func- period 1/λ in location polling. Several output tion of Vm. The impact of Vm on β is more sub- measures are studied. Let α be the mistracking tle when λ = μ. In this case β increases and then probability. An SA crossing is mistracked if decreases as Vm increases, which implies that there is no query between this and the next SA when Vm is small, effect 2 is more significant. On crossings (i.e., the system does not know that the the other hand, when Vm is large, effect 1 domi- user has visited this SA). Let β be the probability nates. An important observation is that when that redundant queries exist between two SA 0.1/μ2 < Vm < 10/μ2, both α and β values are crossings. The larger the α or β values, the worse non-negligibly large, and poor performance of the performance of location polling. location polling cannot be ignored. Figure 6a shows intuitive results that as the Besides mistracking and redundant polling polling frequency λ increases, α decreases and β probabilities, we would also like to investigate increases. We describe two effects of Vm: the following output measures: • Effect 1: When the SA residence time intervals • Ti: The expected period in which the PSAP become more irregular (i.e., Vm increases), cannot correctly track the UE’s location (i.e., more SA residence time intervals without any Ti is the period between an SA crossing and query are observed. when the next query arrives). In this period • Effect 2: When Vm increases, if a query arrives the system does not know the user’s correct in an SA residence time interval, more than location. one query will tend to arrive in this interval. • NR: The expected number of redundant Effect 1 implies that as Vm increases, more queries between two SA crossings (i.e., NR is SA crossings are mistracked, and we have a non- the number of queries issued within two con- trivial observation that α increases as Vm increas- secutive SA crossings). IEEE Wireless Communications • February 2011 9