Smart device to smart device communication

Shahid Mumtaz · Jonathan Rodriguez Editors Smart Device to Smart Device Communication Smart Device to Smart Device Communication Shahid Mumtaz Jonathan Rodriguez • Editors Smart Device to Smart Device Communication 123 Editors Shahid Mumtaz Jonathan Rodriguez Aveiro Portugal ISBN 978-3-319-04962-5 ISBN 978-3-319-04963-2 DOI 10.1007/978-3-319-04963-2 Springer Cham Heidelberg New York Dordrecht London (eBook) Library of Congress Control Number: 2014936421 Ó Springer International Publishing Switzerland 2014 This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained from Springer Permissions for use may be obtained through RightsLink at the Copyright Clearance Center Violations are liable to prosecution under the respective Copyright Law The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made The publisher makes no warranty, express or implied, with respect to the material contained herein Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com) Preface The Internet of Things envisages over billion connected devices that will spur the growth in mobile data traffic to rise exponentially with current predictions suggesting a 1000x increase over the next decade This foreseen market growth has urged mobile operators to examine new ways to plan, deploy, and manage their networks for improving coverage, boosting their network’s capacity, and reducing their capital and operating expenditures (CAPEX and OPEX) To provide a solution toward meeting new and evermore stringent end-user requirements, mobile stakeholders are already preparing the technology roadmap for next generation networks expected to be deployed by 2020 and beyond, which is collectively referred to as ‘‘5G.’’ 5G has a broad vision and envisages design targets that include 10-100 x peak rate data rate, 1000 x network capacity, 10 x energy efficiency, and 10-30 x lower latency These technologies will encompass all aspects of radio access network and applications: from wireless network infrastructure and topologies to physical layer transmission techniques, including spectrum availability, channel modeling, device innovations, and green radio Taking a step toward this vision, Device-to-Device communication in licensed band is one-key enabler toward a more disruptive and cost-effective communication paradigm A key motivation for D2D connectivity is the potential for operators to offload traffic from the core network and the framework for a new communication paradigm to support social networking through localization The current ad-hoc mode of communication does not support this functionality due to configuration complexity LTE-A, Qualcomm and IEEE 802.15.4g (SUN) are currently addressing the standardizing of D2D communication over licensed band A major breakthrough was achieved in due course when 3GPP (LTE-A release, 12 June 2012) agreed on starting a study item for D2D technology This book, inspired by the Eureka Celtic GREEN-T research initiative, brings together academic and industrial stakeholders to identify and discuss technical challenges in D2D communications, and their position on the 5G roadmap toward meeting the 1000x challenge This book is organized in a well-defined structure as shown by Fig 1, that not only elaborates on the progress toward D2D technology solutions, but also details potential use cases, business models, and real time applications In particular, Chap presents an overview tutorial on D2D communication aspects that v vi Preface Fig Organization of book includes the extension of the 3GPP SAE architecture to support D2D scenarios; definition of the D2D protocol stack; design aspects on D2D communication; link adaptation; power control; and channel measurement methods in D2D Moreover, it will elaborate on the use cases, business, and application opportunities that exist to outline the market potential for this technology In Chap 2, we provide a detailed analysis of the evolved LTE-A access, core and protocol architecture to support D2D communication In addition, a comprehensive literature review on coexistence issues between D2D and cellular communication is given Chapter explains the node/peer discovery and mode selection for D2D communication in the LTE-A band In the node discovery section, we explain the existing research on direct discovery that provides the baseline for the novel FlashLinQ technique This chapter also reviews and classifies the state-of-the-art research on mode selection and then introduces a queuing model under busty traffic conditions, and highlights the challenges and open issues to serve as guidelines for future research So far, we have discussed the D2D protocol stack, and its node discovery and mode selection approach After selecting the preferable mode, it is important to control the interference between different D2D pairs, and toward other cellular user In this respect, Chap explains interference management in D2D network, characterizing this interference and highlighting open challenges on this area Thereafter, Chap explains the establishment and maintenance of D2D communications It will elaborate the random access and the retransmission approach for D2D communications, and will present some novel proposals for these schemes Preface vii Until now we have addressed potential scenarios for the D2D communication paradigm, and its position on the 5G roadmap toward enabling cost effective communications for proximity-based services However, in Chap we will detail potential use-cases that give us an in-depth analysis for system requirements and architectural design Specifically, we elaborate on application areas of D2D communications which include cellular network offloading and coverage extension, proximity-based social networking, and providing national security and public safety in infrastructure-less situations So far, we have discussed D2D connectivity over licensed band However, Chap will present different configurations of D2D communication, i.e., control is still performed by the base station, but data is transferred locally using unlicensed band This kind of D2D architecture helps the operator to free some licensed band for other services Moreover, D2D communication can be viewed as one more layer within a HetNet environment which offloads the traffic from both the small cell and macrocell using licensed or un-licensed band, as explained in Chap Hence, in Chap we analyze the performance of incorporating D2D communication in HetNets; comparisons will be made against a full small-cell deployment in HetNets in terms of capacity and backhaul power consumption The last two chapters will explain the different applications of D2D communication: Chap will explain D2D communication in mobile cloud architectures This chapter will introduce the concept of mobile cloud as an efficient platform for cooperative content distribution by exploiting D2D communication Both energy and spectral efficiency aspects of communications will be taken into account, in addition to the throughput enhancement offered by mobile clouds Similarly, Chap 10 will explain the application of D2D communication for smart grids The editors believe that D2D can offer a palette of interesting colors that can paint new business opportunities for mobile stakeholders promoting it as a strong candidate technology for next generation wireless communication system Acknowledgments This book is the first of its kind tackling smart device-to-smart device communications, and its inspiration comes from the editors’ experience at the forefront of European research on D2D However, the editors would also like to thank not only the collaborators that have contributed with chapters toward this book, but also the 4TELL Research Group at the Instituto de Telecomunicações –Aveiro that have provided valuable comments and contributions toward the compilation of this book The editors would also like to acknowledge the Eureka Celtic GREEN-T that has progressed the state-of-the-art on this fast evolving subject ix Contents Introduction to D2D Communication Shahid Mumtaz and Jonathan Rodriguez LTE-A Access, Core, and Protocol Architecture for D2D Communication Dimitris Tsolkas, Eirini Liotou, Nikos Passas and Lazaros Merakos 23 Node/Peer Discovery, Mode Selection, and Signaling for D2D Communication in LTE-A Band Lei Lei and Yiru Kuang 41 Interference Management in D2D Communication Daesik Hong and Seokjung Kim 89 Establishment and Maintenance of D2D Communication Shaoyi Xu 113 Network Assisted Device-to-Device Communications: Use Cases, Design Approaches, and Performance Aspects Gabor Fodor, Stefano Sorrentino and Shabnam Sultana 135 Network-Assisted D2D Over WiFi Direct Alexander Pyattaev, Olga Galinina, Kerstin Johnsson, Adam Surak, Roman Florea, Sergey Andreev and Yevgeni Koucheryavy 165 Device-to-Device Communication in Heterogeneous Networks Yusuf A Sambo, Muhammad Z Shakir, Fabien Héliot, Muhammad A Imran, Shahid Mumtaz and Khalid A Qaraqe 219 xi xii Contents D2D-Based Mobile Clouds for Energy- and Spectral-Efficient Content Distribution Hamidreza Bagheri, Marcos Katz, Frank H P Fitzek, Daniel E Lucani and Morten V Pedersen Interdependency Between Mobile and Electricity Distribution Networks: Outlook and Prospects S Horsmanheimo, N Maskey and L Tuomimäki 237 281 294 S Horsmanheimo et al Fig Examples of calcuation results in Raasepori (a) Computed coverage (b) Computed redundancy redundancy rasters that indicate number of base stations heard at any given location Fig (a) shows computed coverage and Fig.9 (b) redundancy rasters The redundancy calculation indicated that networks, which are primarily dedicated to provide coverage like GSM 900 and UMTS 900 offer higher redundancy level in rural areas Redundancy is low in coastal areas as neighboring cells are missing from seaside and hence, grids on coastal areas are more vulnerable to failures when cellular networks go down 3.2 Fault Analysis The resiliency and redundancy of commercial cellular networks were assessed using fault scenarios in feeder and substation levels The fault analysis was executed by programmatically switching off primary substations, feeders or secondary substations Switching off a feeder or a primary substation caused associated base stations to go down causing gaps to the coverage If the loss was wide enough, communication links in the affected area were lost preventing remote control of power distribution network entities or communication with field teams For each network, four fault cases were performed which are: • • • • Case Case Case Case A: A feeder breaks down B: A primary substation breaks down C: 30 % of primary substations break down due to heavy storm D: 60 % of primary substations break down due to heavy storm The Fig 10 shows that the standard GSM module has the best redundancy and higher average cell counts in the DL direction than UMTS Both networks tolerate cases A and B faults cases comprehensively as the surrounding base stations are able to provide coverage support to the outage area For cases C and D, however, redundancy and cell counts decrease rapidly as multiple base stations mounted on single mast goes down and coverage is lost at same instant Providing battery backup to critical base stations would have helped to decrease the slope of redundancy percentage for cases C and D The detailed results are discussed in [14, 15, 16] Interdependency Between Mobile and Electricity 295 Fig 10 Redundancy percentage and average cell counts with different networks Feasibility Study of Commercial Cellular Networks for Smart Grid Communication in Urban Area The interdependency study was extended to urban areas in order to discover differences between rural and urban areas, and to analyze how utilities could exploit existing and also forthcoming mobile networks in grid communication For the study, an electricity grid layout was obtained from a utility and was used for selecting critical nodes for measurement sites The deployment of the mobile networks in rural and urban area is different The rural networks are merely coverage limited whereas urban networks are typically capacity limited In rural areas, the coverage is mainly established using sectored macrocells on high masts pointing toward main roads and inhabited areas In urban area, higher population density leads to denser mobile and electricity distribution networks The energy consumption and use of mobile broadband services are significantly higher than in rural area The capacity is provided using a larger number of small cells such as micro and indoor cells The cell antennas are typically mounted on rooftops or on ceilings Overlapping of cells is kept minimal in order to reduce interference The traffic loads in urban area are more volatile causing high traffic load peaks in the different parts of networks Our main interest was LTE, which should be capable of providing sufficient data rates and latencies for grid communication LTE networks have recently been deployed by operators in urban areas Operators have taken different strategies to create LTE networks in parallel with legacy 2G and 3G networks as well as with WLAN networks These strategies also affect utilities when they are making decisions pertaining to grid communication 296 S Horsmanheimo et al Fig 11 Measured signal strength and cell counts in GSM network (a) RXLEV_FULL (b) Cell Counts 4.1 Outdoor Measurements The trial area was about km2, including office, university, research center, and residential buildings A 3D model from buildings was created to support more detailed coverage calculations both indoors and outdoors The urban study was more focused on analyzing measurements rather than making coverage calculations The measurements were more comprehensive in urban area and there was less information available about mobile operators’ base station configurations The trial covered multiple operators and multiple radio technologies including GSM, UMTS, LTE and WLAN The analysis concentrated on aspects such as availability, redundancy, capacity, and latency of different radio access technologies As an example, Figs 11, 12, and 13 show the measured RXLEV_FULL (GSM), RSCP (UMTS), and RSRP (LTE) values of one of the operators’ networks In the figures, warm colors (red) indicate high received powers and cold (blue) low ones The values vary significantly along the measurement route due to shadowing of the buildings The LTE received power values (RSRP) are in average smaller than the respective ones in GSM (RXLEV_FULL) and UMTS (RSCP) networks This is merely due to lower transmission powers used in LTE network The receiver sensitivity in an LTE device is approximately 10–20 dBm better than in a UMTS or GSM device Therefore, base stations can use lower transmission powers On average, the measured cell counts are smaller than in rural area as the operators are trying to minimize the overlapping of cells in order to reduce interference and increase the capacity The highest cell counts are measured in GSM networks and lowest ones in LTE networks Deployment of LTE networks is still progressing, and the numbers of the users with LTE-enabled terminals are still rather small compared to GSM- or UMTS-enabled terminals The measurements included also unlicensed WLAN networks Figure 14 shows the measured received powers (RSSI) and cell counts in WLAN High values were obtained near residences and office buildings In those areas, WLAN network Interdependency Between Mobile and Electricity 297 Fig 12 Measured signal strength and cell counts in UMTS network (a) RSCP (b) Cell counts Fig 13 Measured signal strength and cell counts in LTE network (a) RSRP (b) Cell counts Fig 14 Measured signal strength and cell counts in WLAN network (a) RSSI (b) Cell counts 298 S Horsmanheimo et al Fig 15 Indoor measurement in Otaniemi (a) Indoor measuremens sites (b) Mobile robot used in measurement could be applicable for grid communication The main concern is its reliability, security, and sharing of radio resources with other users The outdoor measurements indicated that the coverage is not an issue in urban areas although buildings are often obstructing the line-of-sight between a base station and a terminal For the capacity assessment, LTE measurements were also repeated with Mbps UDP traffic to the UL direction The measurements indicated that all operators’ LTE networks can easily handle Mbps data rates even in low-coverage areas 4.2 Indoor Route Measurements A significant number of grid entities, such as transformers and disconnectors, are placed indoors, typically in the basement or on the ground floor of a building The next phase of our study was to analyze mobile networks’ performance in those places The measurement sites were chosen so that they present critical nodes in a grid having multiple transformers and disconnectors They were also selected based on their building type (an old university or new office building) and the distance from base stations Figure 15(a) shows three of the measurement sites at trial area The red line shows the measured indoor path and gray colors the constructed interior model of the buildings A mobile robot was used for indoor measurements The robot was using different sensors, e.g., 2D lidar and IMU (Inertial measurement unit) sensors, to obtain precise locations for measurement samples The 2D lidar data was also used Interdependency Between Mobile and Electricity 299 Fig 16 Measured RSRP values in LTE network for two operators (a) Operator A (b) Operator B to construct a 2D/3D model from the building’s interior Figure 15(b) shows picture shows the mobile robot used in the measurements Indoor measurements showed that there are significant differences in coverage areas between operators and radio access technologies High received powers were 300 S Horsmanheimo et al Fig 17 Measured RXLEV_FULL values and cell counts in GSM network (a) RXLEV_FULL (b) Cell counts measured in sites that were located on the first floor of relatively new buildings in near proximity of office rooms In contrast, low received powers were measured in basements of old buildings having thick walls Average LTE coverage was weaker Interdependency Between Mobile and Electricity 301 Fig 18 Measured RSCP values and cell counts in UMTS network (a) RSCP (b) Cell counts than GSM or UMTS coverage Nevertheless, the LTE network was still available over 80 % of time In some cases, the LTE device was making a handover to other operator’s LTE network The device was locked on LTE and all handovers to other radio access technologies were declined 302 S Horsmanheimo et al Fig 19 Measured RSRP values and cell counts in LTE network (a) RSRP (b) Cell counts The measurement results indicated that operators are using different strategies on how to deploy LTE cells Some operators were relying on 2G and 3G networks, and covered only hotspots with LTE cells whereas another operator was using more LTE cells to cover the whole area Figure 16 shows measured RSRP values Interdependency Between Mobile and Electricity 303 Fig 20 Measured RSSI values and cell counts in WLAN network (a) RSSI (b) Cell counts of two different operators The red circles indicate the locations where intrahandovers occurred during the measurement Figures 17, 18, and 19 show differences in received powers and cell counts in one operator’s GSM, UMTS, and LTE networks In the figures, we can see that the 304 S Horsmanheimo et al Fig 21 Indoor stationary measurement (a) Meaurement in utilities premises (b) Phases of 24 h meaurement highest received powers as well as cell counts were obtained from GSM and the lowest ones in LTE This operator is using GSM as a primary network technology to provide coverage for mobile calls The deployment of LTE networks is still undergoing as the number of LTE-enabled devices is still relatively small compared to GSM- and UMTS-enabled devices20) The received powers and cell counts were also measured in WLAN networks as shown in Fig 20 The values were high when the site was located on the first floor of the office building in close proximity of office rooms or residences The values got significantly worse when moved to the basements of old buildings WLAN networks can offer additional capacity in ‘‘best effort’’ manner when there are WLAN APs available 4.3 Indoor Stationary Measurements The traffic load in mobile networks varies depending the time of the day To get better insight about its effects on grid automation, a 24 h measurement was conducted at each site The measurement device was placed next to the grid entities as shown in Fig 21 During the measurement, data connections were periodically established to GSM, UMTS and LTE networks, and Mbps UDP traffic was sent Interdependency Between Mobile and Electricity 305 Fig 22 Received powers in UMTS and LTE networks presented with serving cell-specific colors (a) UMTS (b) LTE for half a minute followed by multiple ping messages During the measurement, received power (RXLEV_FULL, RSCP or RSRP), latency, throughput, and jitter values were measured from the active network 306 S Horsmanheimo et al Fig 23 An example of RTT value distribution in different networks and RTT values in time of day (a) RTT distribution (b) RTT comparison Figure 22 shows that during a 24 h stationary measurement, the device was served by multiple cells (cells depicted in graphs with different colors)—not always by the strongest one This phenomenon was discovered in both 2G and 3G networks and in case of all operators The time of the day or the network load was not affecting it The phenomenon appeared to be scheduler related In LTE, the similar effect could also be detected, but it was significantly more moderate than in 2G and 3G networks The biggest differences between radio access technologies can be seen when we examine latency distributions In Fig 23, LTE is presented with red, UMTS with green, and GSM with blue color The UMTS and GSM had a very wide latency distribution Especially high latencies were measured right after a data connection setup The latency distribution of LTE was narrower due to LTE network’s shorter data connection setup time The median latency in LTE was approximately 40 ms that is in line with the values reported by vendors However, there existed single latency values up to 150–220 ms that could hamper 100 ms grid communication These delays are occurring during data connection setups Therefore, in order to avoid them, it might be feasible to keep the connection always on Conclusions The rural area simulations with fault analysis showed that UMTS-900 and GSM900 networks can tolerate the loss of a feeder or a primary substation as long as the radius of the fault area is smaller than the average base station radius The neighboring base stations can provide additional support to the outage area Areas next to sea, large lakes, and international borders are the most vulnerable Structures of electricity distribution and mobile networks are similar in rural areas They both are denser in residential areas and along the roads The simulations indicated that ensuring power supply to critical base stations will improve the Interdependency Between Mobile and Electricity 307 resiliency of telecommunication networks, which in turn will have a significant effect on clearance and repair work and wireless remote control of electricity distribution entities The key factors of telecommunication networks’ resiliency are the speed of the clearance work, the duration of battery backups, and coverage redundancy and base station radius We must keep in mind that the coverage is not the only aspect to look at After a storm, voice and data traffic tend to increase as worried citizens start making more data and voice calls This typically causes congestions in the partly operational telecommunication networks Therefore, capacity and link quality aspects, e.g., delay and jitter are also important when the reliability of grid communication is considered The urban area simulations indicated that changes in the radio environment are more vivid in urban area than in rural area Buildings and walls are obstructing the propagation of radio waves and the used cells are smaller in order to provide sufficient capacity Operators are using different strategies to deploy LTE networks Some operators are relying on 2G and 3G networks and use 4G only for hotspots whereas some operators are aiming to offering high data rates in whole area with denser LTE network There is no one-size-fits-all communications technology for grid communication There exist a lot of alternatives, especially in urban areas, but their exploitation requires planning Neither coverage nor capacity is likely to be the limiting factor The most critical parameter seems to be latency Today’s LTE is capable of meeting the requirements set for slow automatic interactions (B100 ms), but it cannot fulfill the requirements set for protection messages (B10 ms) Future LTE releases are designed to meet also those stringent requirements Research efforts put on unifying TETRA and LTE for public safety communications, and later to run TETRA over LTE are concrete steps to make LTE applicable also for timecritical grid communication The forthcoming 5G technology has much to offer for grid communication by enabling virtual zero latency, ultra-reliability, massive capacity and always on connections, and support for time-critical cloud services in multi-technology networks References IEEE & Smart Grid, http://smartgrid.ieee.org/ieee-smart-grid S Coira et al., in Utility Networks Challenges in the Evolution Towards the Deployment of Smart Metering and Smart Grid Services (Ericsson) GTM White Paper, Trends in Utility Smart Grid Communications Management, GTM Research paper (2013) Ericsson White Paper, LTE for Utilities, Supporting Smart Grids, September 2013 IEC 61850-5 ed2, in Communication Networks and Systems in Substations—Part 5: Communication Requirements for Functions and Device Models 61850-5 iEC:2003(E) (2003–2007) 308 S Horsmanheimo et al J Weimer et al., A virtual laboratory for micro-grid information and communication infrastructures Paper presented at the 2012 3rd IEEE PES international conference and exhibition on Innovative Smart Grid Technologies (ISGT Europe), 14–17 October 2012, pp 1, Utilities Telecom Council, A study of Utility Communication Needs: Key Factors That Impact Utility Communication Networks, September 2010 NSN White Paper, Enhance Mobile Networks to Deliver 1000 Times More Capacity by 2020 (2013) Huawei White Paper, 5G: A Technology Vision (2013) 10 Nokia Solutions and Networks, Looking Ahead to 5G, Building a Virtual Zero Latency GigaBit Experience, White Paper, December 2013 11 Ericsson white Paper, 5G Radio Access, Research and Vision, June 2013 12 NSN White Paper Technology Vision 2020, Technology Vision for the Gigabit Experience, June 2013 13 NSN White Paper, Nokia Solutions and Networks Flatten Network Energy Consumption, Technology Vision 2020, Dec 2013 14 S Horsmanheimo et al., A tool for assessing interdependency of mobile communication and electricity-distribution network Paper presented at the IEEE international conference on smart grid communications, SmartGridComm, Oct 2013 15 S Horsmanheimo et al., Evaluation of interdependencies between mobile communication and electricity distribution networks in fault scenarios Paper presented at the ISGT ASIA (2013) 16 J Penttonen, Inherent Redundancy in Public Wireless, Networks, Applications in Smart Grid, White Paper (2011)