<?xml version='1.0'encoding='utf-8'?>encoding='UTF-8'?> <!DOCTYPE rfc [ <!ENTITY nbsp " "> <!ENTITY zwsp "​"> <!ENTITY nbhy "‑"> <!ENTITY wj "⁠"> ]> <rfc xmlns:xi="http://www.w3.org/2001/XInclude" ipr="trust200902" category="info" docName="draft-ietf-mops-ar-use-case-18" number="9699" consensus="true" obsoletes="" updates="" submissionType="IETF" xml:lang="en" tocInclude="true" symRefs="true" sortRefs="true" version="3"><!-- xml2rfc v2v3 conversion 3.11.1 --><front> <titleabbrev="MOPS AR Use Case">Media Operationsabbrev="XR Use Case">Use Case for an Extended Reality Application on Edge Computing Infrastructure</title> <seriesInfoname="Internet-Draft" value="draft-ietf-mops-ar-use-case-18"/>name="RFC" value="9699"/> <author fullname="Renan Krishna" initials="R." surname="Krishna"> <address> <postal> <country>United Kingdom</country> </postal> <email>renan.krishna@gmail.com</email><uri/></address> </author> <author initials="A." surname="Rahman" fullname="Akbar Rahman"> <organization>Ericsson</organization> <address> <postal> <street>349 Terry Fox Drive</street><city>Ottawa Ontario</city><city>Ottawa</city> <region>Ontario</region> <code>K2K 2V6</code> <country>Canada</country><region/></postal><phone/><email>Akbar.Rahman@ericsson.com</email><uri/></address> </author> <date/> <area>Operations and Management</area> <workgroup> MOPS</workgroup>month="December" year="2024"/> <area>OPS</area> <workgroup>mops</workgroup> <abstract><t> This<t>This document explores the issues involved in the use ofEdge Computingedge computing resources to operationalize a media usecasescase thatinvolveinvolves an Extended Reality (XR)applications.application. In particular, this document discussesthose applicationsan XR application that can run on devices having different form factors (such as different physical sizes and shapes) andneed Edgeneeds edge computing resources to mitigate the effect of problems such asathe need to support interactive communication requiring low latency, limited battery power, and heat dissipation from those devices.The intended audience for this document are network operators who are interested in providing edge computing resources to operationalize the requirements of such applications.This document also discusses the expected behavior of XRapplicationsapplications, which can be used to managethe traffic. In addition, the document discussestraffic, and the service requirementsoffor XR applications to be able to run on the network. Network operators who are interested in providing edge computing resources to operationalize the requirements of such applications are the intended audience for this document. </t> </abstract> </front> <middle> <section anchor="introduction" numbered="true" toc="default"> <name>Introduction</name> <t> Extended Reality (XR) is a term that includes Augmented Reality (AR), Virtual Reality(VR)(VR), and Mixed Reality (MR) <xref target="XR" format="default"/>. AR combines the real and virtual, isinteractiveinteractive, and is aligned to the physical world of the user <xref target="AUGMENTED_2" format="default"/>. On the other hand, VR places the user inside a virtual environment generated by a computer <xref target="AUGMENTED"format="default"/>.MRformat="default"/>. MR merges the real and virtualworldalong a continuum that connects a completely real environment at one end to a completely virtual environment at the other end. In this continuum, all combinations of the real and virtual are captured <xref target="AUGMENTED" format="default"/>. </t> <t> XR applicationswill bringhave several requirements for the network and the mobile devices running these applications. Some XR applicationssuch(such as AR applications) requireareal-time processing of video streams to recognize specific objects. This processing is then used to overlay information on the video being displayed to the user. In addition, other XR applicationssuch(such as AR and VRwillapplications) also require generation of new video frames to be played to the user. Both the real-time processing of video streams and the generation of overlay information are computationally intensive tasks that generate heat <xref target="DEV_HEAT_1"format="default"/>,format="default"/> <xref target="DEV_HEAT_2" format="default"/> and drain battery power <xref target="BATT_DRAIN" format="default"/> on the mobile device running the XR application. Consequently, in order to run applications with XR characteristics on mobile devices, computationally intensive tasks need to be offloaded to resources provided byEdge Computing.edge computing. </t> <t> EdgeComputingcomputing is an emerging paradigmwherewhere, for the purpose of this document, computing resources and storage are made available in close network proximity at the edge of the Internet to mobile devices and sensors <xref target="EDGE_1"format="default"/>,format="default"/> <xref target="EDGE_2" format="default"/>. A computing resource or storage is in close network proximity to a mobile device or sensor if there is a short and high-capacity network path to it such that the latency and bandwidth requirements of applications running on those mobile devices or sensors can be met. These edge computing devices use cloud technologies that enable them to support offloaded XR applications. In particular, cloud implementation techniques <xref target="EDGE_3" format="default"/> such as thefollowsfollowing can be deployed: </t><ul<dl spacing="normal"><li>Disaggregation (using SDN<dt>Disaggregation:</dt><dd>Using Software-Defined Networking (SDN) to break vertically integrated systems into independentcomponents- thesecomponents. These components can have open interfaceswhichthat are standard, welldocumenteddocumented, andnot proprietary), </li> <li>Virtualization (beingnon-proprietary.</dd> <dt>Virtualization:</dt><dd>Being able to run multiple independent copies of thosecomponentscomponents, such as SDN Controllerapps,applications and Virtual NetworkFunctionsFunctions, on a common hardwareplatform).</li> <li>Commoditization (beingplatform.</dd> <dt>Commoditization:</dt><dd>Being able to elastically scale those virtual components across commodity hardware as the workloaddictates).</li> </ul>dictates.</dd> </dl> <t> Such techniques enable XR applicationsrequiring low-latencythat require low latency and high bandwidth to be delivered by proximate edge devices. This is because the disaggregated components can run on proximate edge devices rather than on a remote cloud several hops away and deliverlow latency, high bandwidthlow-latency, high-bandwidth service to offloaded applications <xref target="EDGE_2" format="default"/>. </t> <t> This document discusses the issues involved when edge computing resources are offered by network operators to operationalize the requirements of XR applications running on devices with various form factors.A network operator forFor thepurposespurpose of thisdocumentdocument, a network operator is any organization or individual that manages or operates thecomputecomputing resources or storage in close network proximity to a mobile device orsensors.sensor. Examples of form factors includeHead Mounted Displays (HMD)the following: 1) head-mounted displays (HMDs), such asOptical-see throughoptical see-through HMDs andvideo-see-through HMDsvideo see-through HMDs, 2) hand-held displays, andHand-held displays. Smart phones3) smartphones with video cameras andlocation sensinglocation-sensing capabilities using systems such as a global navigation satellite system(GNSS) are another example of such devices.(GNSS). These devices have limited battery capacity and dissipate heat when running.BesidesAlso, as the user of these devices moves around as they run the XR application, the wireless latency and bandwidth available to the devicesfluctuatesfluctuates, and the communication link itself might fail. As a result, algorithms such as those based onadaptive-bit-rateAdaptive Bitrate (ABR) techniques that base their policy on heuristics or models of deployment perform sub-optimally in such dynamic environments <xref target="ABR_1" format="default"/>. In addition, network operators can expect that the parameters that characterize the expected behavior of XR applications are heavy-tailed. Heaviness of tails is defined as the difference from the normal distribution in the proportion of the values that fall a long way from the mean <xref target="HEAVY_TAIL_3" format="default"/>. Such workloads require appropriate resource management policies to be used on theEdge.edge. The service requirements of XR applications are also challenging when compared tothecurrent video applications. Inparticularparticular, severalQuality of ExperienceQuality-of-Experience (QoE) factors such as motion sickness are unique to XR applications and must be considered when operationalizing a network. This documentmotivatesexamines these issues witha use-case that isthe use case presented in the followingsections.section. </t> </section> <section anchor="use_case" numbered="true" toc="default"> <name>Use Case</name> <t>AThis use caseis now described thatinvolves anapplication withXRsystems' characteristics.application running on a mobile device. Consider a group of tourists who arebeing conducted intaking a tour around the historical site of the Tower of London. As they move around the site and within the historical buildings, they can watch and listen to historical scenes in 3D that are generated by the XR application and then overlaid by their XR headsets onto their real-world view. The headsetthencontinuously updates their view as they move around. </t> <t> The XR application first processes the scene that the walking tourist is watching inreal-timereal time and identifies objects that will be targeted for overlay of high-resolution videos. It then generates high-resolution 3D images of historical scenes related to the perspective of the tourist inreal-time.real time. These generated video images are then overlaid on the view of thereal-worldreal world as seen by the tourist. </t> <t> This processing of scenes and generation of high-resolution imagesis noware discussed in greaterdetail.detail below. </t> <section anchor="processsing_of_scenes" numbered="true" toc="default"> <name>Processing of Scenes</name> <t> The task of processing a scene can be broken down into a pipeline of three consecutivesubtasks namelysubtasks: tracking,followed by anacquisition of a model of the real world, andfinallyregistration <xref target="AUGMENTED" format="default"/>. </t><t> Tracking: The<dl newline="false" spacing="normal"> <dt>Tracking:</dt><dd>The XR application that runs on the mobile device needs to track the six-dimensional pose (translational in the three perpendicular axes and rotational about those three axes) of the user's head,eyeseyes, andtheobjects that are in view <xref target="AUGMENTED" format="default"/>. This requires tracking natural features (forexampleexample, points or edges of objects) that are then used in the next stage of thepipeline. </t> <t> Acquisitionpipeline.</dd> <dt>Acquisition of a model of the realworld: Theworld:</dt><dd>The tracked natural features are used to develop a model of the real world. One of the ways this is done is to develop a model based on an annotated point cloud (a set of points in space that are annotated with descriptors)based modelthat is then stored in a database. To ensure that this database can be scaled up, techniques such as combiningaclient-side simultaneous tracking and mappingand awith server-side localization are used to construct a model of the real world <xref target="SLAM_1"format="default"/>,format="default"/> <xref target="SLAM_2"format="default"/>,format="default"/> <xref target="SLAM_3"format="default"/>,format="default"/> <xref target="SLAM_4" format="default"/>. Another model that can be built is based on a polygon mesh and texture mapping technique. The polygon mesh encodes a 3D object'sshapeshape, which is expressed as a collection of small flat surfaces that are polygons. In texture mapping, color patterns are mappedon toonto an object's surface. A thirdmodellingmodeling technique uses a 2D lightfield that describes the intensity or color of the light rays arriving at a single point from arbitrary directions. Such a 2D lightfield is stored as a two-dimensional table. Assuming distant light sources, the single point is approximately valid for small scenes. For larger scenes, many 3D positions are additionallystoredstored, making the table 5D. A set of all such points (either a 2D or 5D lightfield) can then be used to construct a model of the real world <xref target="AUGMENTED"format="default"/>. </t> <t> Registration: Theformat="default"/>.</dd> <dt>Registration:</dt><dd>The coordinate systems, brightness, and color of virtual and real objects need to be aligned with eachother andother; this process is calledregistration"registration" <xref target="REG" format="default"/>. Once the natural features are tracked as discussed above, virtual objects are geometrically aligned with those features by geometric registration. This is followed by resolving occlusion that can occur between virtual andthereal objects <xref target="OCCL_1"format="default"/>,format="default"/> <xref target="OCCL_2" format="default"/>. The XR application also applies photometric registration <xref target="PHOTO_REG" format="default"/> by aligningthebrightness and color between the virtual and real objects. Additionally, algorithms that calculate global illumination of both the virtual and real objects <xref target="GLB_ILLUM_1"format="default"/>,format="default"/> <xref target="GLB_ILLUM_2" format="default"/> are executed. Various algorithms are also required to deal with artifacts generated by lens distortion <xref target="LENS_DIST" format="default"/>, blur <xref target="BLUR" format="default"/>, noise <xref target="NOISE"format="default"/> etc. are also required. </t>format="default"/>, etc.</dd> </dl> </section> <section anchor="generation" numbered="true" toc="default"> <name>Generation of Images</name> <t> The XR application must generate a high-quality video that has the properties describedin the previous stepabove and overlay the video on the XR device'sdisplay- adisplay. This step is calledsituated visualization."situated visualization". A situated visualization is a visualization in which the virtual objects that need to be seen by the XR user are overlaid correctly on the real world. This entails dealing with registration errors that may arise, ensuring that there is no visual interference <xref target="VIS_INTERFERE" format="default"/>, and finally maintaining temporal coherence by adapting to the movement of user's eyes and head. </t> </section> </section> <section anchor="Req" numbered="true" toc="default"> <name>Technical Challenges and Solutions</name> <t> As discussed insection 2,<xref target="use_case"/>, the components of XR applications perform tasks that are computationally intensive, such as real-time generation and processing of high-quality videocontent that are computationally intensive.content. This sectionwill discussdiscusses the challenges such applications can face as aconsequence.</t>consequence and offers some solutions. </t> <t>As a result of performing computationally intensive tasks on XR devices such as XR glasses, excessive heat is generated by thechip-setschipsets that are involved in the computation <xref target="DEV_HEAT_1"format="default"/>,format="default"/> <xref target="DEV_HEAT_2" format="default"/>. Additionally, the battery on such devices discharges quickly when running such applications <xref target="BATT_DRAIN" format="default"/>. </t> <t> A solution totheproblem of heat dissipation and battery drainageproblemis to offload the processing and video generation tasks to the remote cloud. However, running such tasks on the cloud is not feasible as the end-to-end delays must be within the order of a few milliseconds. Additionally, such applications require high bandwidth and low jitter to provide a high QoE to the user. In order to achieve such hard timing constraints, computationally intensive tasks can be offloaded toEdgeedge devices. </t> <t> Another requirement for our use case and similarapplicationsapplications, such as 360-degree streaming (streaming of video that represents a view in every direction in 3Dspace)space), is that the display on the XR device should synchronize the visual input with the way the user is moving their head. This synchronization is necessary to avoid motion sickness that results from atime-lagtime lag between when the user moves their head and when the appropriate video scene is rendered. This time lag is often called"motion-to-photon" delay."motion-to-photon delay". Studies have shown<xref target="PER_SENSE" format="default"/>, <xref target="XR" format="default"/>, <xref target="OCCL_3" format="default"/>that this delay can be at most20ms20 ms and preferably between7-15ms7-15 ms in order to avoidthemotion sicknessproblem.<xref target="PER_SENSE" format="default"/> <xref target="XR" format="default"/> <xref target="OCCL_3" format="default"/>. Out of these20ms,20 ms, display techniques including the refresh rate of write displays and pixel switching take12-13ms12-13 ms <xref target="OCCL_3"format="default"/>,format="default"/> <xref target="CLOUD" format="default"/>. This leaves7-8ms7-8 ms for the processing of motion sensor inputs, graphic rendering, andround-trip-timeround-trip time (RTT) between the XR device and theEdge.edge. The use of predictive techniques to mask latencies has been considered as a mitigating strategy to reduce motion sickness <xref target="PREDICT" format="default"/>. In addition,Edge Devicesedge devices that are proximate to the user might be used to offload these computationally intensive tasks. Towards this end, a 3GPP studyindicatessuggests anUltra ReliableUltra-Reliable Low Latency of0.1ms0.1 to1ms1 ms for communication between anEdgeedge server and User Equipment (UE) <xref target="URLLC" format="default"/>. </t> <t> Note that theEdgeedge device providing the computation and storage is itself limited in such resources compared to theCloud. So, forcloud. For example, a sudden surge in demand from a large group of tourists can overwhelmthatthe device. This will result in a degraded user experience as their XR device experiences delays in receiving the video frames. In order to deal with this problem, the client XR applications will need to useAdaptive Bit Rate (ABR)ABR algorithms that choosebit-ratesbitrate policies tailored in a fine-grained manner to the resource demands andplaybackplay back the videos with appropriate QoE metrics as the user moves around with the group of tourists. </t> <t> However, the heavy-tailed nature of several operational parameters (e.g., buffer occupancy, throughput, client-server latency, and variable transmission times) makes prediction-based adaptation by ABR algorithms sub-optimal <xref target="ABR_2" format="default"/>. This is because with such distributions, the law of large numbers (how longdoesittaketakes for the sample mean to stabilize) works too slowly <xref target="HEAVY_TAIL_2"format="default"/>,format="default"/> and the mean of sample does not equal the mean of distribution <xref target="HEAVY_TAIL_2"format="default"/>, andformat="default"/>; as aresultresult, standard deviation and variance are unsuitable as metrics for such operational parameters <xref target="HEAVY_TAIL_1" format="default"/>. Other subtle issues with these distributions include the "expectation paradox" <xref target="HEAVY_TAIL_1" format="default"/>where the(the longer the wait for an event, the longer a further need towaitwait) and theissue ofmismatch between the size and count of events <xref target="HEAVY_TAIL_1" format="default"/>.This makesThese issues make designing an algorithm for adaptation error-prone and challenging.Such operational parameters include but are not limited to buffer occupancy, throughput, client-server latency, and variable transmission times.In addition, edge devices and communication links mayfailfail, and logical communication relationships between various software components change frequently as the user moves around with their XR device <xref target="UBICOMP" format="default"/>. </t> </section> <section anchor="ArTraffic" numbered="true" toc="default"> <name>XR Network Traffic</name> <section anchor="traffic_workload" numbered="true" toc="default"> <name>Traffic Workload</name> <t> As discussedearlier,in Sections <xref target="introduction" format="counter"/> and <xref target="Req" format="counter" />, the parameters that capture the characteristics of XR application behavior are heavy-tailed. Examples of such parameters include the distribution of arrival times between XR applicationinvocation,invocations, the amount of data transferred, and the inter-arrival times of packets within a session. As a result, any traffic model based on such parametersare themselvesis also heavy-tailed. Using these models to predict performance under alternative resource allocations by the network operator is challenging. For example, both uplink and downlink traffic to a user device has parameters such as volume of XR data, burst time, and idle time that are heavy-tailed. </t> <t> <xref target="TABLE_1" format="default"/> below shows various streaming video applications and their associated throughput requirements <xref target="METRICS_1" format="default"/>. Since our use case envisages a 6 degrees of freedom (6DoF) video or point cloud,it can be seen fromthe table indicates that it will require 200 to1000Mbps1000 Mbps of bandwidth.As seen from the table,Also, the table shows that XRapplicationapplications, such as the one in our usecasecase, transmit a larger amount of data per unit time as compared totraditionalregular video applications. As a result, issues arisingout offrom heavy-tailedparametersparameters, such as long-range dependent traffic <xref target="METRICS_2"format="default"/>,format="default"/> and self-similar traffic <xref target="METRICS_3" format="default"/>, would be experienced attime scalestimescales of milliseconds and microseconds rather than hours or seconds. Additionally, burstiness at thetime scaletimescale of tens of milliseconds due to the multi-fractal spectrum of traffic will be experienced <xref target="METRICS_4" format="default"/>. Long-range dependent traffic can have longburstsbursts, and various traffic parameters from widely separatedtimetimes can show correlation <xref target="HEAVY_TAIL_1" format="default"/>. Self-similar traffic contains bursts at a wide range oftime scalestimescales <xref target="HEAVY_TAIL_1" format="default"/>. Multi-fractal spectrum bursts for trafficsummarizessummarize the statistical distribution of local scaling exponents found in a traffic trace <xref target="HEAVY_TAIL_1" format="default"/>. The operationalconsequencesconsequence of XR traffic having characteristics such as long-rangedependency,dependency and self-similarity is that the edge servers to which multiple XR devices are connected wirelessly could face long bursts of traffic <xref target="METRICS_2"format="default"/>,format="default"/> <xref target="METRICS_3" format="default"/>. In addition, multi-fractal spectrum burstiness at the scale ofmilli-secondsmilliseconds could induce jitter contributing to motion sickness <xref target="METRICS_4" format="default"/>. This is because bursty traffic combined with variable queueing delays leads to large delay jitter <xref target="METRICS_4" format="default"/>. The operators of edge servers will need to run a'managed"managed edge cloudservice'service" <xref target="METRICS_5" format="default"/> to deal with the above problems. Functionalities that such a managed edge cloud service could operationally provide include dynamic placement of XR servers, mobilitysupportsupport, and energy management <xref target="METRICS_6" format="default"/>. ProvidingEdge serversupport fortheedge servers in techniquesbeing developed at the DETNET Working Group at the IETFsuch as those described in <xref target="RFC8939" format="default"/>, <xref target="RFC9023" format="default"/>, and <xref target="RFC9450" format="default"/> could guarantee performance of XR applications. For example, these techniques could be used for the link between the XR device and the edge as well as within the managed edge cloud service. Another option forthenetwork operators could be to deploy equipment that supports differentiated services <xref target="RFC2475" format="default"/> or per-connectionquality-of-serviceQuality-of-Service (QoS) guarantees using RSVP <xref target="RFC2210" format="default"/>. </t> <t> Thus, the provisioning of edge servers (in terms of the number of servers, the topology, the placement of servers, the assignment of link capacity, CPUs, and Graphics Processing Units (GPUs)) should be performed with the above factors in mind. </t> <table anchor="TABLE_1"> <name>ThroughputrequirementRequirements forstreaming video applications</name>Streaming Video Applications</name> <thead> <tr><th> Application</th> <th> Throughput<th>Application</th> <th>Throughput Required</th> </tr> </thead> <tbody> <tr><td> <t>Real-world<td><t>Real-world objects annotated with text and images for workflow assistance(e.g.(e.g., repair)</t></td> <td> <t>1 Mbps</t></td> </tr> <tr><td> <t>Video Conferencing</t></td><td><t>Video conferencing</t></td> <td> <t>2 Mbps</t></td> </tr> <tr> <td> <t>3DModelmodel andData Visualization</t></td>data visualization</t></td> <td> <t>2 to 20 Mbps</t></td> </tr> <tr> <td> <t>Two-way 3DTelepresence</t></td>telepresence</t></td> <td> <t>5 to 25 Mbps</t></td> </tr> <tr> <td> <t>Current-Gen 360-degree video (4K)</t></td> <td> <t>10 to 50 Mbps</t></td> </tr> <tr> <td> <t>Next-Gen 360-degree video (8K, 90+Frames-per-second, High Dynamic Range, Stereoscopic)</t></td>frames per second, high dynamic range, stereoscopic)</t></td> <td> <t>50 to 200 Mbps</t></td> </tr> <tr> <td><t>6 Degree of Freedom Video<t>6DoF video orPoint Cloud</t></td>point cloud</t></td> <td> <t>200 to 1000 Mbps</t></td> </tr> </tbody> </table><t> Thus, the provisioning of edge servers in terms of the number of servers, the topology, where to place them, the assignment of link capacity, CPUs and GPUs should keep the above factors in mind. </t></section> <section anchor="traffic_performance" numbered="true" toc="default"> <name>Traffic Performance Metrics</name> <t> The performance requirements for XR traffic have characteristics that need to be considered when operationalizing a network. These characteristics arenow discussed.</t>discussed in this section.</t> <t>The bandwidth requirements of XR applications are substantially higher than those of video-based applications.</t> <t>The latency requirements of XR applications have been studied recently <xref target="XR_TRAFFIC" format="default"/>. The following characteristics wereidentified.:identified: </t> <ul spacing="normal"> <li>The uploading of data from an XR device to a remote server for processing dominates the end-to-end latency. </li> <li> A lack of visual features in the grid environment can cause increased latencies as the XR device uploads additional visual data for processing to the remote server.</li> <li>XR applications tend to have large bursts that are separated by significant time gaps.</li> </ul> <t> Additionally, XR applications interact with each other on atime scaletimescale ofa round-trip-timean RTT propagation, and this must be considered when operationalizing a network.</t> <t>The following<xref target="TABLE_2" format="default"/><xref target="METRICS_6" format="default"/>shows a taxonomy of applications with their associated required response times andbandwidths.bandwidths (this data is from Table V in <xref target="METRICS_6" format="default"/>). Response times can be defined as the time interval between the end of a request submission and the end of the corresponding response from a system. If the XR device offloads a task to an edge server, the response time of the server is theround-trip timeRTT from when a data packet is sent from the XR device until a response is received. Note that the required response time provides an upper boundonfor the sum of the time taken by computational taskssuch(such as processing ofscenes,scenes and generation ofimagesimages) and theround-trip time.RTT. This response time depends only on theQuality of Service (QOS)QoS required by an application. The response time is therefore independent of the underlying technology of the network and the time taken by the computational tasks. </t> <t> Our use case requires a response time of20ms20 ms at most and preferably between7-15ms7-15 ms, as discussed earlier. This requirement for response time is similar to the first two entriesofin <xref target="TABLE_2"format="default"/> below.format="default"/>. Additionally, the required bandwidth for our use caseas discussed in section 5.1, <xref target="TABLE_1" format="default"/>,is200Mbps-1000Mbps.200 to 1000 Mbps (see <xref target="traffic_workload"/>). Since our use case envisages multiple users running the XRapplicationsapplication on theirdevices,devices andconnectedconnecting toanthe edge server that is closest to them, these latency and bandwidth connections will grow linearly with the number of users. The operators should match the network provisioning to the maximum number of tourists that can be supported by a link to an edge server. </t> <table anchor="TABLE_2"> <name>Traffic Performance Metrics of Selected XR Applications</name> <thead> <tr> <th> Application</th> <th> Required Response Time</th> <th> Expected Data Capacity</th> <th> Possible Implementations/ Examples</th> </tr> </thead> <tbody> <tr><td> <t>Mobile XR based<td><t>Mobile XR-based remote assistance with uncompressed 4K (1920x1080 pixels) 120 fps HDR 10-bit real-time video stream</t></td><td> <t>Less<td><t>Less than 10 milliseconds</t></td><td> <t>Greater<td><t>Greater than 7.5 Gbps</t></td><td> <t>Assisting<td><t>Assisting maintenance technicians, Industry 4.0 remote maintenance, remote assistance in robotics industry</t></td> </tr> <tr><td> <t>Indoor<td><t>Indoor and localized outdoor navigation </t></td><td> <t>Less<td><t>Less than 20 milliseconds</t></td><td> <t>50<td><t>50 to 200 Mbps</t></td><td> <t>Theme Parks, Shopping Malls, Archaeological Sites, Museum guidance</t></td><td><t>Guidance in theme parks, shopping malls, archaeological sites, and museums</t></td> </tr> <tr><td> <t>Cloud-based Mobile<td><t>Cloud-based mobile XR applications</t></td><td> <t>Less<td><t>Less than 50 milliseconds</t></td><td> <t>50<td><t>50 to 100 Mbps</t></td><td> <t>Google<td><t>Google Live View, XR-enhanced Google Translate </t></td> </tr> </tbody> </table> </section> </section> <section anchor="conclusion" numbered="true" toc="default"> <name>Conclusion</name> <t> In order to operationalize a use case such as the one presented in this document, a network operator could dimension their network to provide a short and high-capacity network path from the edgecomputecomputing resources or storage to the mobile devices running the XR application. This is required to ensure a response time of20ms20 ms at most and preferably between7-15ms.7-15 ms. Additionally, a bandwidth of 200 to1000Mbps1000 Mbps is required by such applications. To deal with the characteristics of XR traffic as discussed in this document, network operators could deploy a managed edge cloud service that operationally provides dynamic placement of XR servers, mobilitysupportsupport, and energy management. Although the use case is technically feasible, economic viability is an important factor that must be considered. </t> </section> <section anchor="iana" numbered="true" toc="default"> <name>IANA Considerations</name> <t> This document has no IANA actions. </t> </section> <section anchor="Sec" numbered="true" toc="default"> <name>Security Considerations</name> <t> The security issues for the presented use case are similar toother streaming applicationsthose described in <xref target="DIST" format="default"/>, <xref target="NIST1" format="default"/>, <xref target="CWE" format="default"/>, and <xref target="NIST2" format="default"/>. This documentitself introduces nodoes not introduce any new security issues. </t> </section><section anchor="ack" numbered="true" toc="default"> <name>Acknowledgements</name> <t> Many Thanks to Spencer Dawkins, Rohit Abhishek, Jake Holland, Kiran Makhijani, Ali Begen, Cullen Jennings, Stephan Wenger, Eric Vyncke, Wesley Eddy, Paul Kyzivat, Jim Guichard, Roman Danyliw, Warren Kumari, and Zaheduzzaman Sarker for providing very helpful feedback, suggestions and comments. </t> </section></middle> <back> <references> <name>Informative References</name> <reference anchor="DEV_HEAT_1"target="">target="https://dl.acm.org/doi/10.1145/2637166.2637230"> <front> <title> Draining ourGlass: An Energyglass: an energy andHeatheat characterization of Google Glass</title> <author initials="R" surname="LiKamWa" fullname="Robert LiKamWa"> <organization/> </author> <author initials="Z" surname="Wang" fullname="Zhen Wang"> <organization/> </author> <author initials="A" surname="Carroll" fullname="Aaron Carroll"> <organization/> </author> <author initials="F" surname="Lin" fullname="Felix Xiaozhu Lin"> <organization/> </author> <author initials="L" surname="Zhong" fullname="Lin Zhong"> <organization/> </author> <dateyear="2013"/>year="2014"/> </front><seriesInfo name="In Proceedings of" value="5th<refcontent>APSys '14: 5th Asia-Pacific Workshop onSystemsSystems, pp.1-7"/>1-7</refcontent> <seriesInfo name="DOI" value="10.1145/2637166.2637230"/> </reference> <reference anchor="EDGE_1"target="">target="https://ieeexplore.ieee.org/document/7807196"> <front> <title> The Emergence of Edge Computing</title> <author initials="M" surname="Satyanarayanan" fullname="Mahadev Satyanarayanan"> <organization/> </author> <date year="2017"/> </front><seriesInfo name="In " value="Computer 50(1)<refcontent>Computer, vol. 50, no. 1, pp.30-39"/>30-39</refcontent> <seriesInfo name="DOI" value="10.1109/MC.2017.9"/> </reference> <reference anchor="EDGE_2"target="">target="https://ieeexplore.ieee.org/document/8812200"> <front> <title> The Seminal Role of Edge-Native Applications</title> <author initials="M" surname="Satyanarayanan" fullname="Mahadev Satyanarayanan"> <organization/> </author> <author initials="G" surname="Klas" fullname="Guenter Klas"> <organization/> </author> <author initials="M" surname="Silva" fullname="Marco Silva"> <organization/> </author> <author initials="S" surname="Mangiante" fullname="Simone Mangiante"> <organization/> </author> <date year="2019"/> </front><seriesInfo name="In " value="IEEE<refcontent>2019 IEEE International Conference on Edge Computing(EDGE)(EDGE), pp.33-40"/>33-40</refcontent> <seriesInfo name="DOI" value="10.1109/EDGE.2019.00022"/> </reference> <reference anchor="ABR_1"target="">target="https://dl.acm.org/doi/10.1145/3098822.3098843"> <front> <title> Neural Adaptive Video Streaming with Pensieve</title> <author initials="H" surname="Mao" fullname="Hongzi Mao"> <organization/> </author> <author initials="R" surname="Netravali" fullname="Ravi Netravali"> <organization/> </author> <author initials="M" surname="Alizadeh" fullname="Mohammad Alizadeh"> <organization/> </author> <date year="2017"/> </front><seriesInfo name="In " value="Proceedings<refcontent>SIGCOMM '17: Proceedings of the Conference of the ACM Special Interest Group on Data Communication, pp.197-210"/>197-210</refcontent> <seriesInfo name="DOI" value="10.1145/3098822.3098843"/> </reference> <reference anchor="ABR_2"target="">target="https://www.usenix.org/conference/nsdi20/presentation/yan"> <front> <title> Learning in situ: a randomized experiment in video streaming </title> <author initials="F" surname="Yan" fullname="Francis Y. Yan"> <organization/> </author> <author initials="H" surname="Ayers" fullname="Hudson Ayers"> <organization/> </author> <author initials="C" surname="Zhu" fullname="Chenzhi Zhu"> <organization/> </author> <author initials="S" surname="Fouladi" fullname="Sadjad Fouladi"> <organization/> </author> <author initials="J" surname="Hong" fullname="James Hong"> <organization/> </author> <author initials="K" surname="Zhang" fullname="Keyi Zhang"> <organization/> </author> <author initials="P" surname="Levis" fullname="Philip Levis"> <organization/> </author> <author initials="K" surname="Winstein" fullname="Keith Winstein"> <organization/> </author> <date month="February" year="2020"/> </front><seriesInfo name="In " value=" 17th<refcontent>17th USENIX Symposium on Networked Systems Design and Implementation (NSDI20),'20), pp.495-511"/>495-511</refcontent> </reference> <reference anchor="HEAVY_TAIL_1"target="">target="https://www.wiley.com/en-us/Internet+Measurement%3A+Infrastructure%2C+Traffic+and+Applications-p-9780470014615"> <front><title> Internet measurement: infrastructure, traffic<title>Internet Measurement: Infrastructure, Traffic andapplications</title>Applications</title> <author initials="M" surname="Crovella" fullname="Mark Crovella"> <organization/> </author> <author initials="B" surname="Krishnamurthy" fullname="Balachander Krishnamurthy"> <organization/> </author> <date year="2006"/> </front><seriesInfo name="John " value="Wiley<refcontent>John Wiley andSons Inc."/>Sons</refcontent> </reference> <reference anchor="HEAVY_TAIL_2"target="">target="https://arxiv.org/pdf/2001.10488"> <front><title> The Statistical<title>Statistical Consequences of FatTails</title>Tails: Real World Preasymptotics, Epistemology, and Applications</title> <author initials="N" surname="Taleb" fullname="Nassim Nicholas Taleb"> <organization/> </author> <dateyear="2020"/>year="2022"/> </front><seriesInfo name="STEM " value="Academic Press"/><refcontent>Revised Edition, STEM Academic Press</refcontent> </reference> <reference anchor="UBICOMP"target="">target="https://www.taylorfrancis.com/chapters/edit/10.1201/9781420093612-6/ubiquitous-computing-systems-jakob-bardram-adrian-friday"> <front><title> Ubiquitous<title>Ubiquitous Computing Systems</title> <author initials="J" surname="Bardram" fullname="Jakob Eyvind Bardram"> <organization/> </author> <author initials="A" surname="Friday" fullname="Adrian Friday"> <organization/> </author> <date year="2009"/> </front><seriesInfo name="In " value=" Ubiquitous<refcontent>Ubiquitous ComputingFundamentalsFundamentals, 1st Edition, Chapman and Hall/CRC Press, pp.37-94. CRC Press"/>37-94</refcontent> </reference> <reference anchor="SLAM_1"target="">target="https://ieeexplore.ieee.org/document/6909455"> <front><title> A minimal solution<title>A Minimal Solution to thegeneralized pose-and-scale problem </title>Generalized Pose-and-Scale Problem</title> <author initials="J" surname="Ventura" fullname="Jonathan Ventura"> <organization/> </author> <author initials="C" surname="Arth" fullname="Clemens Arth"> <organization/> </author> <author initials="G" surname="Reitmayr" fullname="Gerhard Reitmayr"> <organization/> </author> <author initials="D" surname="Schmalstieg" fullname="Dieter Schmalstieg"> <organization/> </author> <date year="2014"/> </front><seriesInfo name="In " value="Proceedings of the<refcontent>2014 IEEE Conference on Computer Vision and Pattern Recognition, pp.422-429"/>422-429</refcontent> <seriesInfo name="DOI" value="10.1109/CVPR.2014.61"/> </reference> <reference anchor="SLAM_2"target="">target="https://link.springer.com/chapter/10.1007/978-3-319-10593-2_2"> <front><title><title>gDLS: Ascalable solutionScalable Solution to thegeneralized poseGeneralized Pose andscale problem </title>Scale Problem</title> <author initials="C" surname="Sweeny" fullname="Chris Sweeny"> <organization/> </author> <author initials="V" surname="Fragoso" fullname="Victor Fragoso"> <organization/> </author> <author initials="T"surname="Hollerer"surname="Höllerer" fullname="TobiasHollerer">Höllerer"> <organization/> </author> <author initials="M" surname="Turk" fullname="Matthew Turk"> <organization/> </author> <date year="2014"/> </front><seriesInfo name="In " value="European Conference on Computer Vision,<refcontent>Computer Vision - ECCV 2014, pp.16-31"/>16-31</refcontent> <seriesInfo name="DOI" value="10.1007/978-3-319-10593-2_2"/> </reference> <reference anchor="SLAM_3"target="">target="https://ieeexplore.ieee.org/document/6636302"> <front><title> Model estimation<title>Model Estimation andselectionSelection towardsunconstrained real-time trackingUnconstrained Real-Time Tracking andmapping </title>Mapping</title> <author initials="S" surname="Gauglitz" fullname="Steffen Gauglitz"> <organization/> </author> <author initials="C"surname="Sweeny"surname="Sweeney" fullname="ChrisSweeny">Sweeney"> <organization/> </author> <author initials="J" surname="Ventura" fullname="Jonathan Ventura"> <organization/> </author> <author initials="M" surname="Turk" fullname="Matthew Turk"> <organization/> </author> <author initials="T"surname="Hollerer"surname="Höllerer" fullname="TobiasHollerer">Höllerer"> <organization/> </author> <dateyear="2013"/>year="2014"/> </front><seriesInfo name="In " value="IEEE transactions<refcontent>IEEE Transactions onvisualizationVisualization andcomputer graphics, 20(6),Computer Graphics, vol. 20, no. 6, pp.825-838"/>825-838</refcontent> <seriesInfo name="DOI" value="10.1109/TVCG.2013.243"/> </reference> <reference anchor="SLAM_4"target="">target="https://ieeexplore.ieee.org/document/6671783"> <front><title> Handling<title>Handling pure camera rotation in keyframe-basedSLAM </title>SLAM</title> <author initials="C" surname="Pirchheim" fullname="Christian Pirchheim"> <organization/> </author> <author initials="D" surname="Schmalstieg" fullname="Dieter Schmalstieg"> <organization/> </author> <author initials="G" surname="Reitmayr" fullname="Gerhard Reitmayr"> <organization/> </author> <date year="2013"/> </front><seriesInfo name="In " value="2013<refcontent>2013 IEEEinternational symposiumInternational Symposium onmixedMixed andaugmented realityAugmented Reality (ISMAR), pp.229-238"/>229-238</refcontent> <seriesInfo name="DOI" value="10.1109/ISMAR.2013.6671783"/> </reference> <reference anchor="OCCL_1"target="">target="https://onlinelibrary.wiley.com/doi/10.1111/1467-8659.1530011"> <front><title> Interactive<title>Interactive Occlusion andautomatic object placementfor augmented reality </title>Automatic Object Placement for Augmented Reality</title> <author initials="D.E" surname="Breen" fullname="David E. Breen"> <organization/> </author> <author initials="R.T" surname="Whitaker" fullname="Ross T. Whitaker"> <organization/> </author> <author initials="E" surname="Rose" fullname="Eric Rose"> <organization/> </author> <author initials="M" surname="Tuceryan" fullname="Mihran Tuceryan"> <organization/> </author> <date month="August" year="1996"/> </front><seriesInfo name="In " value="Computer<refcontent>Computer Graphics Forum, vol. 15, no.3 ,3, pp.229-238,Edinburgh, UK: Blackwell Science Ltd"/>11-22</refcontent> <seriesInfo name="DOI" value="10.1111/1467-8659.1530011"/> </reference> <reference anchor="OCCL_2"target="">target="https://ieeexplore.ieee.org/document/6948419"> <front><title> Pixel-wise<title>Pixel-wise closed-loop registration in video-based augmentedreality </title>reality</title> <author initials="F" surname="Zheng" fullname="Feng Zheng"> <organization/> </author> <author initials="D" surname="Schmalstieg" fullname="Dieter Schmalstieg"> <organization/> </author> <author initials="G" surname="Welch" fullname="Greg Welch"> <organization/> </author> <date year="2014"/> </front><seriesInfo name="In " value="IEEE<refcontent>2014 IEEE International Symposium on Mixed and Augmented Reality (ISMAR), pp.135-143"/>135-143</refcontent> <seriesInfo name="DOI" value="10.1109/ISMAR.2014.6948419"/> </reference> <reference anchor="PHOTO_REG"target="">target="https://ieeexplore.ieee.org/document/6165138"> <front><title> Online tracking<title>Online Tracking ofoutdoor lighting variationsOutdoor Lighting Variations foraugmented realityAugmented Reality withmoving cameras </title>Moving Cameras</title> <author initials="Y" surname="Liu" fullname="Yanli Liu"> <organization/> </author> <author initials="X" surname="Granier" fullname="Xavier Granier"> <organization/> </author> <date year="2012"/> </front><seriesInfo name="In " value="IEEE<refcontent>IEEE Transactions onvisualizationVisualization andcomputer graphics, 18(4), pp.573-580"/>Computer Graphics, vol. 18, no. 4, pp. 573-580</refcontent> <seriesInfo name="DOI" value="10.1109/TVCG.2012.53"/> </reference> <reference anchor="GLB_ILLUM_1"target="">target="https://ieeexplore.ieee.org/document/6671773"> <front><title> Differential irradiance caching<title>Differential Irradiance Caching for fast high-quality light transport between virtual and realworlds.</title>worlds</title> <author initials="P" surname="Kan" fullname="Peter Kan"> <organization/> </author> <author initials="H" surname="Kaufmann" fullname="Hannes Kaufmann"> <organization/> </author> <date year="2013"/> </front><seriesInfo name="In " value="IEEE<refcontent>2013 IEEE International Symposium on Mixed and Augmented Reality(ISMAR),pp. 133-141"/>(ISMAR), pp. 133-141</refcontent> <seriesInfo name="DOI" value="10.1109/ISMAR.2013.6671773"/> </reference> <reference anchor="GLB_ILLUM_2"target="">target="https://ieeexplore.ieee.org/document/6948407"> <front><title> Delta voxel cone tracing.</title><title>Delta Voxel Cone Tracing</title> <author initials="T" surname="Franke" fullname="Tobias Franke"> <organization/> </author> <date year="2014"/> </front><seriesInfo name="In " value="IEEE<refcontent>2014 IEEE International Symposium on Mixed and Augmented Reality (ISMAR), pp.39-44"/>39-44</refcontent> <seriesInfo name="DOI" value="10.1109/ISMAR.2014.6948407"/> </reference> <reference anchor="LENS_DIST"target="">target="https://link.springer.com/chapter/10.1007/978-3-7091-6785-4_2"> <front><title> Practical calibration procedures<title>Practical Calibration Procedures foraugmented reality.</title>Augmented Reality</title> <author initials="A" surname="Fuhrmann" fullname="Anton Fuhrmann"> <organization/> </author> <author initials="D" surname="Schmalstieg" fullname="Dieter Schmalstieg"> <organization/> </author> <author initials="W" surname="Purgathofer" fullname="Werner Purgathofer"> <organization/> </author> <date year="2000"/> </front><seriesInfo name="In " value="Virtual<refcontent>Virtual Environments 2000, pp.3-12. Springer, Vienna"/>3-12</refcontent> <seriesInfo name="DOI" value="10.1007/978-3-7091-6785-4_2"/> </reference> <reference anchor="BLUR"target="">target="https://diglib.eg.org/items/6954bf7e-5852-44cf-8155-4ba269dc4cee"> <front><title> Physically-Based<title>Physically-Based Depth of Field in AugmentedReality.</title>Reality</title> <author initials="P" surname="Kan" fullname="Peter Kan"> <organization/> </author> <author initials="H" surname="Kaufmann" fullname="Hannes Kaufmann"> <organization/> </author> <date year="2012"/> </front><seriesInfo name="In " value="Eurographics (Short Papers),<refcontent>Eurographics 2012 - Short Papers, pp.89-92."/>89-92</refcontent> <seriesInfo name="DOI" value="10.2312/conf/EG2012/short/089-092"/> </reference> <reference anchor="NOISE"target="">target="https://ieeexplore.ieee.org/document/4079277"> <front><title> Enhanced<title>Enhanced visual realism by incorporating camera imageeffects.</title>effects</title> <author initials="J" surname="Fischer" fullname="Jan Fischer"> <organization/> </author> <author initials="D" surname="Bartz" fullname="Dirk Bartz"> <organization/> </author> <author initials="W"surname="Straßer"surname="Strasser" fullname="WolfgangStraßer">Strasser"> <organization/> </author> <date year="2006"/> </front><seriesInfo name="In " value="IEEE/ACM<refcontent>2006 IEEE/ACM International Symposium on Mixed and Augmented Reality, pp.205-208."/>205-208</refcontent> <seriesInfo name="DOI" value="10.1109/ISMAR.2006.297815"/> </reference> <reference anchor="VIS_INTERFERE"target="">target="https://ieeexplore.ieee.org/document/4538846"> <front><title> Interactive focus<title>Interactive Focus andcontext visualizationContext Visualization foraugmented reality.</title>Augmented Reality</title> <author initials="D" surname="Kalkofen" fullname="Denis Kalkofen"> <organization/> </author> <author initials="E" surname="Mendez" fullname="Erick Mendez"> <organization/> </author> <author initials="D" surname="Schmalstieg" fullname="Dieter Schmalstieg"> <organization/> </author> <date year="2007"/> </front><seriesInfo name="In " value="6th<refcontent>2007 6th IEEE and ACM International Symposium on Mixed and Augmented Reality, pp.191-201."/>191-201</refcontent> <seriesInfo name="DOI" value="10.1109/ISMAR.2007.4538846"/> </reference> <reference anchor="DEV_HEAT_2"target="">target="https://www.mdpi.com/1424-8220/20/5/1446"> <front><title> Thermal model<title>Thermal Model andcountermeasuresCountermeasures forfuture smart glasses.</title>Future Smart Glasses</title> <author initials="K" surname="Matsuhashi" fullname="Kodai Matsuhashi"> <organization/> </author> <author initials="T" surname="Kanamoto" fullname="Toshiki Kanamoto"> <organization/> </author> <author initials="A" surname="Kurokawa"fullname=" Atsushifullname="Atsushi Kurokawa"> <organization/> </author> <date year="2020"/> </front> <refcontent>Sensors, vol. 20, no. 5, p. 1446</refcontent> <seriesInfoname="In " value="Sensors, 20(5), p.1446."/>name="DOI" value="10.3390/s20051446"/> </reference> <reference anchor="BATT_DRAIN"target="">target="https://ieeexplore.ieee.org/document/7993011"> <front><title> A survey<title>A Survey ofwearable devicesWearable Devices andchallenges.</title>Challenges</title> <author initials="S" surname="Seneviratne" fullname="Suranga Seneviratne"> <organization/> </author> <author initials="Y" surname="Hu" fullname="Yining Hu"> <organization/> </author> <author initials="T" surname="Nguyen"fullname=" Thamfullname="Tham Nguyen"> <organization/> </author> <author initials="G" surname="Lan"fullname=" Guohaofullname="Guohao Lan"> <organization/> </author> <author initials="S" surname="Khalifa"fullname=" Sarafullname="Sara Khalifa"> <organization/> </author> <author initials="K" surname="Thilakarathna"fullname=" Kanchanafullname="Kanchana Thilakarathna"> <organization/> </author> <author initials="M" surname="Hassan"fullname=" Mahbubfullname="Mahbub Hassan"> <organization/> </author> <author initials="A" surname="Seneviratne"fullname=" Arunafullname="Aruna Seneviratne"> <organization/> </author> <date year="2017"/> </front><seriesInfo name="In " value="IEEE<refcontent>IEEE Communication Surveys and Tutorials,19(4), p.2573-2620."/>vol. 19 no. 4, pp. 2573-2620</refcontent> <seriesInfo name="DOI" value="10.1109/COMST.2017.2731979"/> </reference> <reference anchor="PER_SENSE"target="">target="https://dl.acm.org/doi/10.1145/1012551.1012559"> <front> <title> Perceptual sensitivity to head tracking latency in virtual environments with varying degrees of scene complexity.</title> <author initials="K" surname="Mania"fullname="Katrinafullname="Katerina Mania"> <organization/> </author> <author initials="B.D." surname="Adelstein" fullname="Bernard D.Adelstein ">Adelstein"> <organization/> </author> <author initials="S.R." surname="Ellis"fullname=" Stephenfullname="Stephen R. Ellis"> <organization/> </author> <author initials="M.I." surname="Hill"fullname=" Michaelfullname="Michael I. Hill"> <organization/> </author> <date year="2004"/> </front><seriesInfo name="In " value="Proceedings<refcontent>APGV '04: Proceedings of the 1st Symposium on Applied perception in graphics andvisualizationvisualization, pp.39-47."/>39-47</refcontent> <seriesInfo name="DOI" value="10.1145/1012551.1012559"/> </reference> <reference anchor="XR"target="">target="https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3534"> <front><title> 3GPP TR 26.928: Extended<title>Extended Reality (XR) in5G.</title> <author initials="" surname="3GPP" fullname="3GPP"> <organization/>5G</title> <author> <organization>3GPP</organization> </author> <date year="2020"/> </front> <seriesInfoname="" value="https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3534"/>name="3GPP TR" value="26.928"/> </reference> <reference anchor="CLOUD"target="">target="https://dl.acm.org/doi/10.1145/3442381.3449854"> <front><title> Surrounded<title>Surrounded by the Clouds: A Comprehensive Cloud ReachabilityStudy.</title>Study</title> <author initials="L." surname="Corneo"fullname=" Lorenzofullname="Lorenzo Corneo"> <organization/> </author> <author initials="M." surname="Eder"fullname=" Maximilianfullname="Maximilian Eder"> <organization/> </author> <author initials="N." surname="Mohan"fullname=" Nitinderfullname="Nitinder Mohan"> <organization/> </author> <author initials="A." surname="Zavodovski"fullname=" Aleksandrfullname="Aleksandr Zavodovski"> <organization/> </author> <author initials="S." surname="Bayhan"fullname=" Suzanfullname="Suzan Bayhan"> <organization/> </author> <author initials="W." surname="Wong"fullname=" Walterfullname="Walter Wong"> <organization/> </author> <author initials="P." surname="Gunningberg"fullname=" Perfullname="Per Gunningberg"> <organization/> </author> <author initials="J." surname="Kangasharju"fullname=" Jussifullname="Jussi Kangasharju"> <organization/> </author> <author initials="J." surname="Ott"fullname=" Jörgfullname="Jörg Ott"> <organization/> </author> <date year="2021"/> </front><seriesInfo name="In" value="Proceedings<refcontent>WWW '21: Proceedings of the Web Conference 2021, pp.295-304"/>295-304</refcontent> <seriesInfo name="DOI" value="10.1145/3442381.3449854"/> </reference> <reference anchor="OCCL_3"target="">target="https://www.roadtovr.com/oculus-shares-5-key-ingredients-for-presence-in-virtual-reality/"> <front><title> Oculus<title>Oculus Shares 5 Key Ingredients for Presence in VirtualReality.</title>Reality</title> <author initials="B." surname="Lang" fullname="Ben Lang"> <organization/> </author> <date day="24" month="September" year="2014"/> </front><seriesInfo name="" value="https://www.roadtovr.com/oculus-shares-5-key-ingredients-for-presence-in-virtual-reality/"/><refcontent>Road to VR</refcontent> </reference> <reference anchor="PREDICT"target="">target="https://pubmed.ncbi.nlm.nih.gov/22624290/"> <front><title> The<title>The effect of apparent latency on simulator sickness while using a see-through helmet-mounted display:Reducingreducing apparent latency with predictivecompensation..</title>compensation</title> <authorinitials="T. J."initials="T.J." surname="Buker" fullname="Timothy J. Buker"> <organization/> </author> <author initials="D.A." surname="Vincenzi" fullname="Dennis A. Vincenzi "> <organization/> </author> <author initials="J.E." surname="Deaton" fullname=" John E. Deaton"> <organization/> </author> <date month="April" year="2012"/> </front><seriesInfo name="In " value="Human factors 54.2,<refcontent>Human Factors, vol. 54, no. 2, pp.235-249."/>235-249</refcontent> <seriesInfo name="DOI" value="10.1177/0018720811428734"/> </reference> <reference anchor="URLLC"target="">target="https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3453"> <front><title> 3GPP TR 23.725: Study<title>Study on enhancement of Ultra-Reliable Low-Latency Communication (URLLC) support in the 5G Core network(5GC).</title> <author initials="" surname="3GPP" fullname="3GPP"> <organization/>(5GC)</title> <author> <organization>3GPP</organization> </author> <date year="2019"/> </front> <seriesInfoname="" value="https://portal.3gpp.org/desktopmodules/Specifications/ SpecificationDetails.aspx?specificationId=3453"/>name="3GPP TR" value="23.725"/> </reference> <reference anchor="AUGMENTED"target="">target="https://www.oreilly.com/library/view/augmented-reality-principles/9780133153217/"> <front><title> Augmented Reality</title><title>Augmented Reality: Principles and Practice</title> <authorinitials="D. S."initials="D" surname="Schmalstieg" fullname="Dieter Schmalstieg"> <organization/> </author> <authorinitials="T.H." surname="Hollerer" fullname="Dennis A.Hollerer ">initials="T" surname="Höllerer" fullname="Tobias Höllerer"> <organization/> </author> <date year="2016"/> </front><seriesInfo name="" value="Addison Wesley"/><refcontent>Addison-Wesley Professional</refcontent> </reference> <reference anchor="REG"target="">target="https://direct.mit.edu/pvar/article-abstract/6/4/413/18334/Registration-Error-Analysis-for-Augmented-Reality?redirectedFrom=fulltext"> <front><title> Registration error analysis<title>Registration Error Analysis foraugmented reality.</title>Augmented Reality</title> <authorinitials="R. L."initials="R.L." surname="Holloway" fullname="Richard L. Holloway"> <organization/> </author> <date month="August" year="1997"/> </front><seriesInfo name="In " value="Presence:Teleoperators<refcontent>Presence: Teleoperators and VirtualEnvironments 6.4,Environments, vol. 6, no. 4, pp.413-432."/>413-432</refcontent> <seriesInfo name="DOI" value="10.1162/pres.1997.6.4.413"/> </reference> <reference anchor="XR_TRAFFIC"target="">target="https://ieeexplore.ieee.org/document/9158434"> <front> <title> Characterization of Multi-User Augmented Reality over Cellular Networks </title> <author initials="K." surname="Apicharttrisorn" fullname="Kittipat Apicharttrisorn"> <organization/> </author> <author initials="B." surname="Balasubramanian" fullname="Bharath Balasubramanian "> <organization/> </author> <author initials="J." surname="Chen"fullname=" Jiasifullname="Jiasi Chen"> <organization/> </author> <author initials="R." surname="Sivaraj"fullname=" Rajarajanfullname="Rajarajan Sivaraj"> <organization/> </author> <author initials="Y." surname="Tsai"fullname=" Yi-Zhenfullname="Yi-Zhen Tsai"> <organization/> </author> <author initials="R." surname="Jana" fullname="Rittwik Jana"> <organization/> </author> <author initials="S." surname="Krishnamurthy" fullname="Srikanth Krishnamurthy"> <organization/> </author> <author initials="T." surname="Tran" fullname="Tuyen Tran"> <organization/> </author> <author initials="Y." surname="Zhou" fullname="Yu Zhou"> <organization/> </author> <date year="2020"/> </front><seriesInfo name="In " value="17th<refcontent>2020 17th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON), pp.1-9. IEEE"/>1-9</refcontent> <seriesInfo name="DOI" value="10.1109/SECON48991.2020.9158434"/> </reference> <reference anchor="EDGE_3"target="">target="https://link.springer.com/book/10.1007/978-3-031-79733-0"> <front><title> 5G mobile networks:<title>5G Mobile Networks: Asystems approach.</title>Systems Approach</title> <author initials="L." surname="Peterson" fullname="Larry Peterson"> <organization/> </author> <author initials="O." surname="Sunay" fullname="Oguz Sunay"> <organization/> </author> <date year="2020"/> </front><seriesInfo name="In " value="Synthesis<refcontent>Synthesis Lectures on NetworkSystems."/>Systems</refcontent> <seriesInfo name="DOI" value="10.1007/978-3-031-79733-0"/> </reference> <reference anchor="AUGMENTED_2"target="">target="https://direct.mit.edu/pvar/article-abstract/6/4/355/18336/A-Survey-of-Augmented-Reality?redirectedFrom=fulltext"> <front><title> A<title>A Survey of AugmentedReality.</title>Reality</title> <authorinitials="R. T."initials="R.T." surname="Azuma" fullname="Ronald T. Azuma"> <organization/> </author> <date month="August" year="1997"/> </front><seriesInfo name="" value="Presence:Teleoperators<refcontent>Presence: Teleoperators and VirtualEnvironments 6.4,Environments, vol. 6, no. 4, pp.355-385."/>355-385</refcontent> <seriesInfo name="DOI" value="10.1162/pres.1997.6.4.355"/> </reference> <reference anchor="METRICS_1"target="">target="https://gsacom.com/paper/augmented-virtual-reality-first-wave-5g-killer-apps-qualcomm-abi-research/"> <front><title> Augmented<title>Augmented and Virtual Reality: The first Wave of KillerApps.</title> <author initials="" surname="ABI Research" fullname="ABI Research"> <organization/>Apps: Qualcomm - ABI Research</title> <author> <organization>ABI Research</organization> </author> <date month="April" year="2017"/> </front><seriesInfo name="" value="https://gsacom.com/paper/augmented-virtual-reality-first-wave-5g-killer-apps-qualcomm-abi-research/"/></reference> <reference anchor="METRICS_2"target="">target="https://ieeexplore.ieee.org/document/392383"> <front><title> Wide Area Traffic: The Failure<title>Wide area traffic: the failure of PoissonModelling.</title>modeling</title> <author initials="V." surname="Paxon" fullname="Vern Paxon"> <organization/> </author> <author initials="S." surname="Floyd" fullname="Sally Floyd"> <organization/> </author> <date month="June" year="1995"/> </front><seriesInfo name="In" value="IEEE/ACM<refcontent>IEEE/ACM Transactions on Networking, vol. 3, no. 3, pp.226-244."/>226-244</refcontent> <seriesInfo name="DOI" value="10.1109/90.392383"/> </reference> <reference anchor="METRICS_3"target="">target="https://ieeexplore.ieee.org/abstract/document/554723"> <front><title> Self-Similarity Through High Variability: Statistical Analysis<title>Self-similarity through high variability: statistical analysis and Ethernet LANTraffictraffic atSource Level.</title>source level</title> <author initials="W." surname="Willinger" fullname="Walter Willinger"> <organization/> </author> <author initials="M.S." surname="Taqqu" fullname="Murad S. Taqqu"> <organization/> </author> <author initials="R." surname="Sherman" fullname="Robert Sherman"> <organization/> </author> <author initials="D.V." surname="Wilson" fullname="Daniel V. Wilson"> <organization/> </author> <date month="February" year="1997"/> </front><seriesInfo name="In" value="IEEE/ACM<refcontent>IEEE/ACM Transactions on Networking, vol. 5, no. 1, pp.71-86."/>71-86</refcontent> <seriesInfo name="DOI" value="10.1109/90.554723"/> </reference> <reference anchor="METRICS_4"target="">target="https://www.sciencedirect.com/science/article/pii/S1063520300903427"> <front><title> Multiscale<title>Multiscale Analysis and DataNetworks.</title>Networks</title> <author initials="A.C." surname="Gilbert"fullname="A. C.fullname="A.C. Gilbert"> <organization/> </author> <date month="May" year="2001"/> </front><seriesInfo name="In" value="Applied<refcontent>Applied and Computational Harmonic Analysis, vol. 10, no. 3, pp.185-202."/>185-202</refcontent> <seriesInfo name="DOI" value="10.1006/acha.2000.0342"/> </reference> <reference anchor="METRICS_5"target="">target="https://research.google/pubs/site-reliability-engineering-how-google-runs-production-systems/"> <front><title> Site<title>Site Reliability Engineering: How Google Runs ProductionSystems.</title>Systems</title> <author initials="B." surname="Beyer" fullname="BetsyBeyer">Beyer" role="editor"> <organization/> </author> <author initials="C." surname="Jones" fullname="ChrisJones">Jones" role="editor"> <organization/> </author> <author initials="J." surname="Petoff" fullname="JenniferPetoff">Petoff" role="editor"> <organization/> </author> <author initials="N.R." surname="Murphy" fullname="Niall RichardMurphy">Murphy" role="editor"> <organization/> </author> <date year="2016"/> </front><seriesInfo name="" value="O'Reilly<refcontent>O'Reilly Media,Inc."/>Inc.</refcontent> </reference> <reference anchor="METRICS_6"target="">target="https://ieeexplore.ieee.org/document/9363323"> <front><title> A survey<title>A Survey onmobile augmented reality withMobile Augmented Reality With 5Gmobile edge computing: architectures, applications,Mobile Edge Computing: Architectures, Applications, andtechnical aspects.</title>Technical Aspects</title> <author initials="Y." surname="Siriwardhana" fullname="Yushan Siriwardhana"> <organization/> </author> <author initials="P." surname="Porambage" fullname="Pawani Porambage"> <organization/> </author> <author initials="M." surname="Liyanage" fullname="Madhusanka Liyanage"> <organization/> </author> <author initials="M." surname="Ylianttila" fullname="Mika Ylianttila"> <organization/> </author> <date year="2021"/> </front><seriesInfo name="In" value="IEEE<refcontent>IEEE Communications Surveys and Tutorials,Volvol. 23,No. 2"/>no. 2, pp. 1160-1192</refcontent> <seriesInfo name="DOI" value="10.1109/COMST.2021.3061981"/> </reference> <reference anchor="HEAVY_TAIL_3"target="">target="https://www.wiley.com/en-us/A+Primer+in+Data+Reduction%3A+An+Introductory+Statistics+Textbook-p-9780471101352"> <front><title> A<title>A Primer in DataReduction.</title>Reduction: An Introductory Statistics Textbook</title> <author initials="A." surname="Ehrenberg" fullname="A.S.C Ehrenberg "> <organization/> </author> <dateyear="1982"/> </front> <seriesInfo name="John" value="Wiley, London"/> </reference> <reference anchor="RFC9023" target="https://www.rfc-editor.org/info/rfc9023"> <front> <title>Deterministic Networking (DetNet) Data Plane: IP over IEEE 802.1 Time-Sensitive Networking (TSN)</title> <author fullname="B. Varga" initials="B." role="editor" surname="Varga"/> <author fullname="J. Farkas" initials="J." surname="Farkas"/> <author fullname="A. Malis" initials="A." surname="Malis"/> <author fullname="S. Bryant" initials="S." surname="Bryant"/> <date month="June" year="2021"/> <abstract> <t>This document specifies the Deterministic Networking IP data plane when operating over a Time-Sensitive Networking (TSN) sub-network. This document does not define new procedures or processes. Whenever this document makes statements or recommendations, these are taken from normative text in the referenced RFCs.</t> </abstract> </front> <seriesInfo name="RFC" value="9023"/> <seriesInfo name="DOI" value="10.17487/RFC9023"/> </reference> <reference anchor="RFC8939" target="https://www.rfc-editor.org/info/rfc8939"> <front> <title>Deterministic Networking (DetNet) Data Plane: IP</title> <author fullname="B. Varga" initials="B." role="editor" surname="Varga"/> <author fullname="J. Farkas" initials="J." surname="Farkas"/> <author fullname="L. Berger" initials="L." surname="Berger"/> <author fullname="D. Fedyk" initials="D." surname="Fedyk"/> <author fullname="S. Bryant" initials="S." surname="Bryant"/> <date month="November" year="2020"/> <abstract> <t>This document specifies the Deterministic Networking (DetNet) data plane operation for IP hosts and routers that provide DetNet service to IP-encapsulated data. No DetNet-specific encapsulation is defined to support IP flows; instead, the existing IP-layer and higher-layer protocol header information is used to support flow identification and DetNet service delivery. This document builds on the DetNet architecture (RFC 8655) and data plane framework (RFC 8938).</t> </abstract>year="2007"/> </front><seriesInfo name="RFC" value="8939"/> <seriesInfo name="DOI" value="10.17487/RFC8939"/> </reference> <reference anchor="RFC9450" target="https://www.rfc-editor.org/info/rfc9450"> <front> <title>Reliable and Available Wireless (RAW) Use Cases</title> <author fullname="CJ. Bernardos" initials="CJ." role="editor" surname="Bernardos"/> <author fullname="G. Papadopoulos" initials="G." surname="Papadopoulos"/> <author fullname="P. Thubert" initials="P." surname="Thubert"/> <author fullname="F. Theoleyre" initials="F." surname="Theoleyre"/> <date month="August" year="2023"/> <abstract> <t>The wireless medium presents significant specific challenges to achieve properties similar to those of wired deterministic networks. At the same time, a number of use cases cannot be solved with wires and justify the extra effort of going wireless. This document presents wireless use cases (such as aeronautical communications, amusement parks, industrial applications, pro audio and video, gaming, Unmanned Aerial Vehicle (UAV) and vehicle-to-vehicle (V2V) control, edge robotics, and emergency vehicles), demanding reliable<refcontent>John Wiley andavailable behavior.</t> </abstract> </front> <seriesInfo name="RFC" value="9450"/> <seriesInfo name="DOI" value="10.17487/RFC9450"/>Sons</refcontent> </reference> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.9023.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8939.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.9450.xml"/> <reference anchor="DIST"target="">target="https://dl.acm.org/doi/10.5555/2029110"> <front> <title> Distributed Systems: Concepts and Design</title> <author initials="G" surname="Coulouris" fullname="George Coulouris"> <organization/> </author> <author initials="J" surname="Dollimore" fullname="Jean Dollimore"> <organization/> </author> <author initials="T" surname="Kindberg" fullname="Tim Kindberg"> <organization/> </author> <author initials="G" surname="Blair" fullname="Gordon Blair"> <organization/> </author> <date year="2011"/> </front><seriesInfo name="" value="Addison Wesley"/><refcontent>Addison-Wesley</refcontent> </reference> <reference anchor="NIST1"target="">target="https://csrc.nist.gov/pubs/sp/800/146/final"> <front><title> NIST SP 800-146: Cloud<title>Cloud Computing Synopsis and Recommendations</title><author initials="" surname="" fullname="NIST"> <organization/><author> <organization>NIST</organization> </author> <date month="May" year="2012"/> </front> <seriesInfoname="" value="National Institute of Standards and Technology, US Department of Commerce"/>name="NIST SP" value="800-146"/> <seriesInfo name="DOI" value="10.6028/NIST.SP.800-146"/> </reference> <reference anchor="CWE"target="">target="https://www.sans.org/top25-software-errors/"> <front><title> CWE/SANS<title>CWE/SANS TOP 25 Most Dangerous SoftwareErrorss</title> <author initials="" surname="" fullname="SANS Institute"> <organization/>Errors</title> <author> <organization>SANS Institute</organization> </author><date year="2012"/></front><seriesInfo name="" value="Common Weakness Enumeration, SANS Institute"/></reference> <reference anchor="NIST2"target="">target="https://csrc.nist.gov/pubs/sp/800/123/final"> <front><title> NIST SP 800-123: Guide<title>Guide to General Server Security</title><author initials="" surname="" fullname="NIST"> <organization/><author> <organization>NIST</organization> </author> <date month="July" year="2008"/> </front> <seriesInfoname="" value="National Institute of Standards and Technology, US Department of Commerce"/> </reference> <reference anchor="RFC2210" target="https://www.rfc-editor.org/info/rfc2210"> <front> <title>The Use of RSVP with IETF Integrated Services</title> <author fullname="J. Wroclawski" initials="J." surname="Wroclawski"/> <date month="September" year="1997"/> <abstract> <t>This note describes the use of the RSVP resource reservation protocol with the Controlled-Load and Guaranteed QoS control services. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="2210"/>name="NIST SP" value="800-123"/> <seriesInfo name="DOI"value="10.17487/RFC2210"/> </reference> <reference anchor="RFC2475" target="https://www.rfc-editor.org/info/rfc2475"> <front> <title>An Architecture for Differentiated Services</title> <author fullname="S. Blake" initials="S." surname="Blake"/> <author fullname="D. Black" initials="D." surname="Black"/> <author fullname="M. Carlson" initials="M." surname="Carlson"/> <author fullname="E. Davies" initials="E." surname="Davies"/> <author fullname="Z. Wang" initials="Z." surname="Wang"/> <author fullname="W. Weiss" initials="W." surname="Weiss"/> <date month="December" year="1998"/> <abstract> <t>This document defines an architecture for implementing scalable service differentiation in the Internet. This memo provides information for the Internet community.</t> </abstract> </front> <seriesInfo name="RFC" value="2475"/> <seriesInfo name="DOI" value="10.17487/RFC2475"/>value="10.6028/NIST.SP.800-123"/> </reference> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2210.xml"/> <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2475.xml"/> </references> <section anchor="ack" numbered="false" toc="default"> <name>Acknowledgements</name> <t>Many thanks to <contact fullname="Spencer Dawkins"/>, <contact fullname="Rohit Abhishek"/>, <contact fullname="Jake Holland"/>, <contact fullname="Kiran Makhijani"/>, <contact fullname="Ali Begen"/>, <contact fullname="Cullen Jennings"/>, <contact fullname="Stephan Wenger"/>, <contact fullname="Eric Vyncke"/>, <contact fullname="Wesley Eddy"/>, <contact fullname="Paul Kyzivat"/>, <contact fullname="Jim Guichard"/>, <contact fullname="Roman Danyliw"/>, <contact fullname="Warren Kumari"/>, and <contact fullname="Zaheduzzaman Sarker"/> for providing helpful feedback, suggestions, and comments.</t> </section> </back> </rfc>