5G meets Construction Machines: Towards a Smart working Site
Yusheng Xiang, Bing Xu, Tianqing Su, Christine Brach, Samuel S. Mao, Marcus Geimer
55G meets Construction Machines:Towards a Smart working Site st Yusheng Xiang
Institute of Vehicle System TechnologyKarlsruhe Institute of Technology
Karlsruhe, [email protected] nd Bing Xu
Institute of Vehicle System TechnologyKarlsruhe Institute of Technology
Karlsruhe, [email protected] rd Tianqing Su
Institue of Communication TechnologyTechnical University of Braunschweig
Brunswick, [email protected] th Christine Brach
Division of Mobile HydraulicsRobert Bosch GmbH
Elchingen, [email protected] th Samuel S. Mao
Department of Mechanical EngineeringUniversity of California, Berkeley
Berkeley, [email protected] th Marcus Geimer
Institute of Vehicle System TechnologyKarlsruhe Institute of Technology
Karlsruhe, [email protected]
Abstract —The fleet management of mobile working machineswith the help of connectivity can increase safety and productivity.Although in our previous study, we proposed a solution to useIEEE 802.11p to achieve the fleet management of constructionmachines, the shortcoming of WIFI may limit the usage ofthis technology in some cases. Alternatively, the fifth-generationmobile networks (5G) have shown great potential to solve theproblems. Thus, as the world’s first academic paper investigating5G and construction machines’ cooperation, we demonstratedthe scenarios where 5G can have a significant effect on theconstruction machines industry. Also, based on the simulationwe made in ns − , we compared the performance of 4G and5G for the most relevant construction machines scenarios. Lastbut not least, we showed the feasibility of remote-control andself-working construction machines with the help of 5G. Index Terms —5G, LTE, mmWave,, Remote control, self-working construction machinery
I. I
NTRODUCTION
The fleet management of mobile machines is the principalresearch direction of the internet of things in the constructionmachines industry. Besides using the ad-hoc network as thefirst version for mobile machines [1], 5G attracts huge atten-tion to be expected to achieve even higher-quality communica-tion. As mentioned in our previous study, WIFI technology canaccomplish realtime communication among mobile machinesso that they will work denser and faster. As a consequence,we can increase productivity and therefore reduce the durationof the construction projects. This is meaningful for the casesof repairing projects on the highway, mining projects, andtransportation in harbors. Since mobile machines are usuallyworking surrounded by dust and Lidars are quite sensitive tothis case, cameras are a more robust and promising approachtowards self-working machines or remote control of mobileconstruction machines. As we know, as the videos’ resolutionincreases, both image recognition algorithms, and humans canacquire information easier and more accurate. However, thecapacity, especially the uplink capacity of WIFI technology, limits the introduction of wireless high-definition (HD) videotransmission for construction machines. As we did not findcomprehensive research indicating how can 5G change themobile construction machines industry, we first analyze thepotential use cases for the implementation of 5G for theconstruction machines industry in this paper. Followed byillustrating the benefits by utilizing 5G with our simulationresults by means of ns − [2]. Last but not least, we show theblueprint of future smart working sites based on the simulationresults. Fig. 1 and Fig. 2 demonstrate the potential use casesof 5G in the field of mobile construction machines. User RoutereNodeBEPC Operating Computer NR Or?
UE1 UE2 UE3
Fig. 1. Remote control with live streaming: here cameras will be installed onthe mobile machines while the driver sits in a comfortable room to operatethe machines remotely. Thanks to 5G, HD video streaming can be sent withlow delay and high reliability.
II.
CURRENT TECHNOLOGIES a r X i v : . [ c s . N I] S e p plink: VideoDownlink: Optimized path and working command Cloud computation:
1. Segmentation
2. Localization3. Working command4. Path planning
Fig. 2. Self-working mobile machines: here, cameras will be fixed on the ground instead of being installed on the machines to avoid the obstruction of vision.The stream will be uploaded to the center commander and be processed on the cloud. Based on the stream from more than two cameras, the depth informationand motion of machines can be acquired. Afterward, the command signal will be sent directly to the machines. The research about instance segmentation ofconstruction machines can be found in [3].
5G technology, we only take the parameters and data that morethan at least half of the community agree with into account.To overcome the shortcoming of 4G [4], the basic require-ments for the 5G are drawn by [5]–[9]: higher transmissionrate, shorter latency, higher reliability, and more User Equip-ment (UE) connection. Correspondingly, the big 3 concepts:enhanced Mobile Broadband (eMBB), Ultra Reliable LowLatency Communications (URLLC), massive Machine TypeCommunications (mMTC) [10], were proposed. Accordingto the 3rd Generation Partnership Project (3GPP) 38.101agreement [11], 5G NR mainly uses two frequency bands: FR1frequency band and FR2 frequency band. The frequency rangeof the FR1 band is 450MHz-6GHz, which is also called thesub 6GHz sub frequency band; the frequency range of the FR2band is 24.25GHz-52.6GHz, usually called millimeter wave(mmWave) band. Currently, the most influential providers inthe field of 5G are Huawei for sub 6Ghz band and Qualcommfor the mmWave band, separately. Other competitors men-tioned quite often are Samsung, Ericsson, Datang, Nokia, Tele-com, Intel, and ZTE. As we know, the higher the frequency, thecloser the characteristic is to the light. That is, the propagationof the signal will be more similar to the light, which onlygoes straightforward so that the obstacles can easily block it.Also, the energy loss increases dramatically as the propagationdistance increases, and proportionally to the square of the fre-quency. Consequently, the coverage problem, which restrictsthe promotion of the high-frequency spectrum 5G, occurs dueto the nature of the mmWave. For this reason, most countries,such as China, Japan, and Korea, give priority to the sub 6Ghz band since the coverage is much larger, and thus more peoplecan benefit from 5G technology. Compared to 4G, which onlyhas 20MHz channel bandwidth, 5G is allocated about 100MHzin the sub 6Ghz area. Moreover, thanks to the novel Multiple-Input Multiple-Output (MIMO) technology, more antennas areused simultaneously to achieve a much higher transmissionrate than the previous 4G technology. Compared to the 4Ghandsets, which only have 2*2 or 4*4 antennas, 5G basestations and UEs have antenna array to increase the spectrumutilization [12], [13]. However, since such 5G UEs also use thesub 6Ghz band, there is principally not greatly different than4G, and thus some serious problems are inevitable. First of all,because the sub 6Ghz area is also used by 2G, 3G, 4G, andthus already very crowed, a further increase of the bandwidthis almost impossible. Although some communication operatorsgive 5G more channel bandwidth, which was belongs to 2Gand 3G to increase the bandwidth of 5G further, the bandwidthis surely not enough for the future potential requirements.In addition, the configuration of the antenna depends on thesignal frequency. At sub 6Ghz, the wavelength is more than1cm, so that the number of the antenna in the UE, in thiscase, is also limited. Therefore, soon after sub 6Ghz waspromoted, how to use the higher FR2 frequency regions, i.e.,higher than 28Ghz, has become a hot topic. Compared tothe sub 6Ghz region, it is quite easy to have 1Ghz channelbandwidth in the FR2 region so that the transmission ratiois expected to be much higher. In the mmWave frequencyband, taking the 28GHz frequency band as an example, theavailable spectrum bandwidth has reached 1GHz, while thevailable signal bandwidth of each channel in the 60GHzfrequency band is 2GHz [11]. In the case of constant spectrumutilization, if the mmWave frequency band is selected, thedata transmission rate can be doubled by directly doublingthe bandwidth. Since 3GPP has decided to continue to useorthogonal frequency division multiplexing (OFDM) technol-ogy for 5G NR [11], mmWave technology has become thebiggest novel idea of 5G. Although mmWave is already usedby satellite, they were considered as infeasible for the dailylife scenarios. Until recently, the novel technology unlocksthe high-frequency spectrum. Concretely, thanks to antennasarray, which constitutes a large number of antennas and thebeamforming technology [14], the energy can be concentratedin small regions. Moreover, because the antennas for mmWavecan be designed much smaller than the microwave antennas,the antennas in the mmWave antenna array are much denserand achieve a larger number for the same geometrical appara-tus. Along with a certain number of small cell base stations,mmWave comes to the forefront of commercial applications.The introduction of other important 5G technologies, suchas new numerology, LDPC/Polar codes, etc., can let OFDMtechnology better extend to the mmWave band. To adapt tothe large bandwidth characteristics of mmWave, 5G definesmultiple sub-carrier intervals, of which the larger sub-carrierintervals are specifically designed for mmWave, whereas thelower is for the compatibility of previous system deriving fromthe 4G era. One of the main goals of 5G is to support URLLCservices with stringent requirements for reliability and delay.LTE achieves a user plane two-way wireless delay of less than10ms, and the design goal of 5G is to reduce this delay by atleast 5 times, that is, less than 2ms. According to the 3GPPTS 38.211 protocol [11], the 5G NR physical layer providesmultiple sub-carrier spacing configurations [15]. By increasingthe sub-carrier spacing, the duration of OFDM symbols isreduced, thereby reducing the duration of a single time slotand reducing delay. The 3GPP protocol claims that the sub-carrier spacing is inversely proportional to the OFDM symbolduration, which is an inherent attribute of OFDM. For thecurrent network communication technology, the key capabilityindicators of the 5G system have been greatly improved. Theinformation transmission delay of the 5G network can reachmilliseconds, which meets the stringent requirements of thenetwork and guarantees the safety of controlled UE. Thepeak rate of 5G can reach 10-20 Gbit/s, and the numberof connections can reach 1 million/ km [16]. Apparently,although the technology can overcome the difficulties ofimplementing the mmWave, the base stations for mmWave areenergy-consuming equipment. Thus, Heterogeneous Network(HetNet) is also essential in the 5G era, i.e., most scientistsin the wireless community believe that both sub- and above6Ghz networks will coexist in a long time. The same as LTE,5G also has device to device network to solve the problemwhen UEs are outside of the coverage of base stations [17]. III. W HERE CAN CONSTRUCTION SITES BE BENEFITEDFROM
ROBLEM STATEMENT AND GOAL
In the previous study from Bermudez [25], they tested theperformance of the LTE network by the transmission of videodata. Their article evaluated two protocols’ behaviors, realtimemessaging protocol (RTMP) and realtime streaming protocol(RTSP), in a 4G environment. Based on their results, we cansay that the performance of LTE to transfer the HD video fromthe working side to the operator side in realtime is good butnot fully satisfying.Also, the through of LTE is in a steady-state growth situa-tion. That means, the simulation parameters of Bermudez [25]missed the extreme working critical condition. We still don’tknow whether the LTE network can always have an excellentperformance in a more stringent remote-control situation ornot. Therefore, the comparison between LTE and 5G for videotransmission in construction scenarios is necessary. For remotecontrol, the delay is always a significant indicator because itequals to the accuracy and reliability of the job and the safetyof the controlled machine [19]. Inspired from this and to fillthis research gap, we decide to choose the performance fromone of the new 5G technologies, mmWave, to compare withthe LTE network’s performance for construction machineryin remote-control and self-working scenarios. Meanwhile, wewill give the simulation a more stringent critical environment.Under the goal of finding out whether 5G network is moresuitable for remote control or self-working construction ma- chinery than LTE or not and if so, how good it is, our researchis not in existence already.V. M
ODELLING
In the scenarios shown in Fig. 1 or Fig. 2, that our UEs,i.e., construction machines, are under the sight of HD camerasor with the HD cameras. Here we assume our constructionmachines and cameras are both connected with the base stationand the operator. The operator will give the constructionmachines commands. Meanwhile, they will collect the videostreaming data from cameras. In case that the cameras are stickto the machine, the operator will give the order and receivethe video data simultaneously. Compare with the instructionfrom the operator, video streaming data will occupy a muchlarger bandwidth. Therefore, in our research, we will use videostreaming as the media, which can verify the performanceof both networks. Obviously, video streaming with differentresolution occupies different network bandwidth. Dependingon the different resolution requirements of video streaming,different pressure will be applied to the network.For our research, we will use ns − [2], [26] as oursimulation tool. To perform LTE simulation, we directly callthe LTE module inside ns − because there is already acomplete set of simulation modules and processes in ns − for 4G [27]. On the other hand, for the 5G network, since itis still quite novel, ns − has not yet developed an officialsimulation platform with all 5G modules. Fortunately, because ns − is an open-source platform, many professional networksimulation players can contribute to this platform based ontheir requirements, such as rewriting the algorithm, addingpatch packages, or doing other upgrades. Among them, weselected the model from Mezzavilla [28] to simulate the 5GmmWave performance. The following paragraphs will presentsome basic architecture details and model settings for bothnetwork models. Basic parameters are shown in table I andtable II. TABLE ILTE N
ETWORK P ARAMETERS , FROM
Parameters ValueBandwidth 25MHzDownlink Earfcn 100Uplink Earfcn 18100Scheduler
PfFfMacScheduler
TABLE II5G N
ETWORK P ARAMETERS , F
ROM
Parameters ValueBand n257Downlink NR-ARFCN 2054167 2104165Uplink NR-ARFCN 2054167 2104165Scheduler
MmWaveMacScheduler . Model Parameters1) Propagation Model:
For LTE, we use
FriisPropagation-LossModel [30]. Given an unobstructed visual path betweenthe transmitter and receiver, the free-space propagation modelcan predict the strength of the received signal. According toFriis [31], the received signal strength can be described as, P r ( d ) = P t G t G r λ (4 π ) d L (1)where P r ( d ) is defined as received signal power, P t is transmitpower, G t is transmit antenna gain, G r is receive antenna gain, λ is wavelength(m), d is the distance, and L is the system loss.As for 5G, we use MmWavePropagationLossModel [32],[33]. This mmWave model presents two kinds of path lossmodels. The first one is the one that we used, which is in astatistical characteristic of the Line of Sight (LOS) state. Theother one is
BuildingsObstaclePropagationLossModel [34],adding the obstacle between the gNB and the UE. Furtherpath-loss models of mmWave can be found in [32].
2) Transmission Control Protocol/Internet Protocol(TCP/IP):
The network transmission adopts the TCP/IPprotocol. The core protocols of the TCP/IP protocol are thetransport layer protocol (TCP and UDP) and the networklayer protocol (IP), which are usually implemented in thekernel of the operating system. Because the purpose of TCP isto achieve reliable data transmission, it has a set of handshakemechanism, send - confirmation, timeout - resend [35]. Inthe case of video streaming, the network spending of TCPtransmission is too large, thus impairing image quality andlatency. Therefore, the UDP transmission method is preferredfor realtime live streaming [36], [37].
3) Hybrid Automatic Repeat Request (HARQ):
For LTEand 5G, they both have two levels of retransmission mech-anisms: HARQ at the MAC layer and ARQ at the Radio LinkControl (RLC) layer [38], [39]. For 4G, the retransmissionof lost or erroneous data is mainly handled by the HARQmechanism of the MAC layer and supplemented by the ARQof the RLC. The HARQ mechanism of the MAC layer canprovide fast retransmission, and the ARQ mechanism of theRLC layer can provide reliable data transmission. In contrast,for 5G, the uplink HARQ mechanism is the same as thedownlink, and both are asynchronous HARQ. There will betwo kinds of changes [40]. First, the scheduling timing is moreflexible, especially in TDD mode, resulting in more resourceallocation flexibility. Second, the pressure of data bufferingwill increase. Unlike LTE’s uplink synchronous HARQ, asyn-chronous HARQ may have a longer retransmission interval.During this time, the UE must buffer the unACKed data, whichwill increase the buffering pressure.
4) Scenarios:
We setup three scenarios for both networkenvironments. In the first scenario, we choose 2Mbps asthe video streaming volume. 2Mpbs is nearly the level of720P video streaming bandwidth requirement [41]. Then weare changing the UE number from 2 to 20. In the secondscenario, we set the UE number as a constant condition. Bychanging the data volume to realize new scenario, from 1Mbps to 8Mbps, which includes the bandwidth requirement of720P(3Mbps), 1080P(5Mbps), and 3D 1080P(6Mbps) videos[42], we explore the network performance with a varyingresolution of the video. At last, we let UEs move to acquirethe knowledge of how mobility condition affects the networks.For scenarios 1 and 2, our mobile machines will be underthe sight of those high-definition cameras. Those cameras willcollect the working video data and transfer it to the operator.The UEs in scenario three will be cameras installed on mobilemachines. Here they will change their position together withthe construction machinery as collecting the video streaming.
TABLE IIIN
ETWORK S CENARIO Data Volume(Mbps) 2UE Number 2, 4, 6, 8, 10, 12, 14, 16, 18, 20
TABLE IVN
ETWORK S CENARIO UE Number 8Data Volume (Mbps) 1, 2, 3, 4, 5, 6, 7, 8
TABLE VN
ETWORK S CENARIO Data Volume(Mbps) 2UE Number 8UE Distance(m) 20, 40, 60, 80, 100, 120, 140, 160, 180, 200
VI. S
IMULATION R ESULTS
This section presents the results of our simulated networkscenarios in terms of throughput, packet loss rate, and delay.As for both network environments, we performed the simula-tion repeatedly and got the average value to improve accuracy.
Fig. 3. Network topology.
UE Number (a) Average throughput with increasingUE
UE Number (b) Packet loss rate with increasing UE
UE Number (c) Average delay with increasing UE (d) Average throughput with increasingData Volume (e) Packet loss rate with increasing DataVolume (f) Average delay with increasing DataVolumeFig. 4. Simulation results of scenario 1 and scenario 2
10 15 20 25 30 35 40 45 50 55 60
Speed(km/h) (a) Average throughput with increasingVelocity
10 15 20 25 30 35 40 45 50 55 60
Speed(km/h) (b) Packet loss rate with increasing Ve-locity
10 15 20 25 30 35 40 45 50 55 60
Speed(km/h) (c) Average delay with increasing Veloc-ityFig. 5. Simulation results of scenario 3
The network topology is shown in Fig. 3. From nodes 3to node 12 represent a set of remote devices, i.e., cameras,and the transmission data represents the video data sent bythe camera avatar, which is finally sent to the user terminal(node 1) through eNodeB or NR (node 2) and Evolved packetcore (EPC) (node 0).With the increase in the number of UE, the throughputsimulation results are shown in Fig. 4(a). In the beginning,the throughput of LTE and 5G networks has increased rapidly,and the throughput matches the total data volume, whichmeans both of them can complete the transmission of thevideo streaming task. As the number of UEs further increases,the 5G network can still transmit video service data better;however, the LTE network cannot provide enough transmissioncapacity for video service data, reaching a state of businesssaturation. It can be observed that the throughput remains basically unchanged as the UE number grows, about 17Mbps.In Fig. 4(b) simulation results of the packet loss rate as the UEnumber increase are shown. When the UE number is small,both LTE and 5G networks can keep the packet loss rateapproximately equal to 0, i.e., almost no packet loss occurs. Ifthe UE number increases, the 5G network can still maintainthe network with an almost low packet loss rate. Still, theLTE network will have more packet loss due to its networkresource constraints. It must discard the video service’s datapackets, causing the transmitted video to lose frames, freezeor completely lose the result of the video image, which willseriously affect the operator’s performance of the constructionmachinery. Besides that, high latency will make a video to beout of sync. In these cases, the operator cannot grasp the on-site working environment in realtime, resulting in the operatorto make wrong judgments about the working environment,hich is very dangerous for the work task and the constructionmachinery. The average delay of the 5G network is lowerthan that of the LTE network, as shown in Fig. 4(c). This isbecause the 5G network can provide larger network bandwidth,increase network transmission speed, and reduce data packetdelay. If the UE number is small, the average delay of the LTEnetwork is about twice that of the 5G network; however, whenthe number of users is large, the average delay of the LTEnetwork is much higher than that of the 5G network. At thistime, the LTE network cannot guarantee the video streamingservice.In the second simulation scenario, the number of UE num-ber is fixed to 8, and the video service data is increased from1Mbps to 8Mbps. The simulation result of throughput withincreasing video service rate is shown in Fig. 4(d). When thevideo service rates are 1Mbps and 2Mbps, the throughput ofthe LTE network and the 5G network can meet video streamingservices’ requirements. However, when the video service rateexceeds 3Mbps, the throughput of the LTE network does notcontinue to increase, and the throughput of the 5G network stillincreases with the video service rate, which can guarantee thetransmission of the video service. The simulation results of thepacket loss rate are demonstrated in Fig. 4(e), where we cansee that the 5G network has been able to maintain the packetloss rate at a low level. However, severe packet loss will occurfor LTE networks when a higher video service rate is required.In case that the video service rate is 5Mbps, the packet loss rateof LTE exceeds 50%. The average delay of video services ispresented in Fig. 4(f). 5G network continues to increase withthe increase of data volume, and they are all maintained ata low level, even when the video service rate is 5Mbps, theaverage delay still does not exceed 25ms. The video serviceaverage delay of the LTE network is significantly higher thanthat of the 5G network. In short, as the video service rate goeshigher, the improvement with 5G will be more significant.In the third scenario, we want to simulate the case thatconstruction machines carry the cameras with them whenthey change their positions. Here the video service data rateis 2Mbps, and the number of remote devices is still 8. Wesimulate the longest distance up to 200m since the longestpropagation distance of mmWave is considered as 200m [43].The simulation results of throughput with increasing speedare shown in Fig. 5(a). Due to lower frequency bands, LTEnetwork performances are affected only slightly with mobility.Also, when the UE velocity is lower than 40km/h, the through-put of the 5G network is still in a relatively stable declinestage. However, when the UE velocity exceeds 40km/h, thethroughput of the 5G network drops dramatically, and thusthe transmission of video services cannot be guaranteed at thistime. Fig. 5(b) presents, as the velocity increases, the packetloss rate is rising slowly for LTE networks. However, the 5Gnetwork will suffer a fast increasing packet loss rate when theUE moves faster than 30km/h. In Fig. 5(c), both the delay ofthe LTE network and 5G network increase steadily with thegrowth of velocity. However, we can notice that the delay ofthe 5G network still much advantageous compare with LTE. To sum up, 5G mmWave has significant advantages in termsof throughput, packet loss, and latency if the UEs are fixed.Although one of the requirements for 5G is the capacity todeal with high mobility, the mmWave 5G may still have aproblem if the beamforming technology, concretely, trackingalgorithm, is not perfect. In contrast, since 4G uses a lowerfrequency band, this problem is not so apparent for 4G, whichhints the suitability of using sub 6Ghz band 5G.VII. C
ONCLUSION
In this paper, we first indicate that 5G can be employedin the construction machines industry to improve the remotecontrol operation and work as an essential component toachieve self-working construction machines. By taking theremote-control and self-working of construction machineryas the scenes and using video streaming transmission as themedium, we compared the LTE network’s performance andthe 5G mmWave network. Based on our research, we foundthat 5G has the capability to accomplish a better quality of livestreaming so that both scenes can be significantly improved.Especially, 5G can let more cameras in the same network,indicating the possibility to acquire depth information from thevideo. Besides, since it is not difficult to let the machines al-ways under the cameras’ vision, we suggest letting the camerasunmoved avoid the shortcoming of mmWave. Otherwise, morerobust beamforming, i.e., dynamic beamforming, algorithm isneeded.Since we use video as the medium to test the performanceof the two networks, future work shall refine video factorsand explore how the structure of the different encoding videostyle will affect the networks. Besides, starting from thevideo phase, through the networks, and finally to the controloperator, a simulation analysis of the entire link can be carriedout to improve the content of this article. Moreover, as the6G technology is on the way [44]–[46], we will explore thepossibility to benefit the construction machine industry from6G technology in our next paper.R
EFERENCES[1] Y. Xiang, T. Su, C. Brach, X. Liu, and M. Geimer, “Realtime estimationof ieee 802.11p for mobile working machines communication respectingdelay and packet loss,” in
IEEE Intelligent Vehicle Symposium 2020 , LasVegas, USA, 2020.[2] T. R. Henderson, M. Lacage, G. F. Riley, C. Dowell, and J. Kopena,Eds.,
Network simulations with the ns-3 simulator . New York, NY:ACM, 2008.[3] Y. Xiang, Y. Huang, Z. Zhang, T. Su, C. Brach, S. Mao, and M. Geimer,“Kit moma v2: towards instance segmentation of construction ma-chines,” arXiv , 2020.[4] A. Ghosh, R. Ratasuk, B. Mondal, N. Mangalvedhe, and T. Thomas,“Lte-advanced: next-generation wireless broadband technology,”
IEEEWireless Communications , vol. 17, no. 3, pp. 10–22, 2010.[5] S. Chen and J. Zhao, “The requirements, challenges, and technologiesfor 5g of terrestrial mobile telecommunication,”
IEEE CommunicationsMagazine , vol. 52, no. 5, pp. 36–43, 2014.[6] A. Gupta and R. K. Jha, “A survey of 5g network: Architecture andemerging technologies,”
IEEE Access , vol. 3, pp. 1206–1232, 2015.[7] M. Lauridsen, L. C. Gimenez, I. Rodriguez, T. B. Sorensen, and P. Mo-gensen, “From lte to 5g for connected mobility,”
IEEE CommunicationsMagazine , vol. 55, no. 3, pp. 156–162, 2017.8] M. N. Tahir, K. Maenpaa, and T. Sukuvaara, “Evolving wireless vehic-ular communication system level comparison and analysis of 802, 11 p,4g 5g,” in . IEEE, 2019, pp. 48–52.[9] A. Vinel, J. Breu, T. H. Luan, and H. Hu, “Emerging technology for 5g-enabled vehicular networks,”
IEEE Wireless Communications , vol. 24,no. 6, p. 12, 2017.[10] M. Shafi, A. F. Molisch, P. J. Smith, T. Haustein, P. Zhu, P. de Silva,F. Tufvesson, A. Benjebbour, and G. Wunder, “5g: A tutorial overviewof standards, trials, challenges, deployment, and practice,”
IEEE Journalon Selected Areas in Communications
IEEE Com-munications Magazine , vol. 55, no. 6, pp. 94–101, 2017.[13] Y. Li, C. Sim, Y. Luo, and G. Yang, “12-port 5g massive mimo antennaarray in sub-6ghz mobile handset for lte bands 42/43/46 applications,”
IEEE Access , vol. 6, pp. 344–354, 2018.[14] S. Han, C. I, Z. Xu, and C. Rowell, “Large-scale antenna systems withhybrid analog and digital beamforming for millimeter wave 5g,”
IEEECommunications Magazine , vol. 53, no. 1, pp. 186–194, 2015.[15] J. Campos, “Understanding the 5g nr physical layer.”[16] E. Mohyeldin, “Minimum technical performance requirements for imt-2020 radio interface (s),” in
ITU-R Workshop on IMT-2020 TerrestrialRadio Interfaces , 2016, pp. 1–12.[17] R. I. Ansari, C. Chrysostomou, S. A. Hassan, M. Guizani, S. Mumtaz,J. Rodriguez, and Rodrigues, J. J. P. C., “5g d2d networks: Techniques,challenges, and future prospects,”
IEEE Systems Journal , vol. 12, no. 4,pp. 3970–3984, 2018.[18] W. Roh, J. Y. Seol, J. Park, B. Lee, J. Lee, Y. Kim, J. Cho, K. Cheun, andF. Aryanfar, “Millimeter-wave beamforming as an enabling technologyfor 5g cellular communications: theoretical feasibility and prototyperesults,”
IEEE Communications Magazine , vol. 52, no. 2, pp. 106–113,2014.[19] H. Yoshida, T. Yoshimoto, T. Innami, K. Ohashi, H. Furuya, andN. Mori, “Improving efficiency of remote construction by using adaptivevideo streaming,” in , 2019.[20] S. Dadhich, U. Bodin, and U. Andersson, “Key challenges in automationof earth-moving machines,”
Automation in Construction , vol. 68, pp.212–222, 2016.[21] I. Johansson, S. Dadhich, U. Bodin, and T. J¨onsson, “Adaptive videowith scream over lte for remote-operated working machines,”
WirelessCommunications and Mobile Computing , vol. 2018, pp. 1–10, 2018.[22] S. Dadhich, U. Bodin, F. Sandin, and U. Andersson, “From tele-remoteoperation to semi-automated wheel-loader,”
International Journal ofElectrical and Electronic Engineering and Telecommunications , vol. 7,no. 4, pp. 178–182, 2018.[23] Y. Xiang, Y. Wang, T. Su, R. Li, C. Brach, S. Mao, and M. Geimer,“Kit moma: A mobile machines dataset,” arXiv , 2020. [Online].Available: arXiv:2007.04198[24] Y. Xiang, T. Tang, T. Su, C. Brach, L. Liu, S. Mao, and M. Geimer,“Fast crdnn: Towards on site training of mobile construction machines,” arXiv , 2020. [Online]. Available: https://arxiv.org/pdf/2006.03169.pdf[25] H. F. Bermudez, R. Sanchez-Iborra, and J. L. Arciniegas, “Performancevalidation of ns3-lte emulation for live video streaming underqos parameters. in , networking and communications (wimob),” in , 2017, pp.300–307. [Online]. Available: http://ieeexplore.ieee.org/servlet/opac?punumber=8106925[26] L. F. Perrone, C. Cicconetti, G. Stea, and B. C. Ward, “On theautomation of computer network simulators,” in
Proceedings of the 2ndInternational Conference on Simulation Tools and Techniques
IEEECommunications Surveys & Tutorials
Evaluation ofX2-handover performance based on RSRP measurement with Friis pathloss using network simulator version 3 (NS-3) . 2014 2nd InternationalConference on Information and Communication Technology (ICoICT),IEEE, 2014.[31] H. T. Friis, “A note on a simple transmission formula,” in
Proceedingsof the IRE , vol. 34(5), 1946, pp. 254–256.[32] 3GPP, “Study on channel model for frequency spectrum above 6 ghz.”[33] G. R. MacCartney, J. Zhang, S. Nie, and T. S. Rappaport, “Path lossmodels for 5g millimeter wave propagation channels in urban micro-cells,” in , 2009, pp. 237–240.[36] D. Madhuri and P. C. Reddy, “Performance comparison of tcp, udpand sctp in a wired network,” in , 2016, pp. 1–6.[37] M. Sinky, A. Dhamodaran, B. Lee, and J. Zhao, “Analysis of h.264bitstream prioritization for dual tcp/udp streaming of hd video overwlans,” in , 2015, pp. 576–581.[38] M. Kim, S. Kim, and Y. Lim, “An implementation of downlink asyn-chronous harq for lte tdd system,” in . IEEE, 2012, pp. 271–274.[39] A. Anand and G. de Veciana, Eds.,
Resource Allocation and HARQOptimization for URLLC Traffic in 5G Wireless Networks , vol. 36. IEEEJournal on Selected Areas in Communications, IEEE, 2018.[40] J. Yeo, S. Park, J. Oh, Y. Kim, and J. Lee, “Partial retransmission schemefor harq enhancement in 5g wireless communications,” in , 2017.[41] C. Summerson, “How much data does netflix use?”
How-To Geek ,2018, pp. 1–5.[44] K. B. Letaief, W. Chen, Y. Shi, J. Zhang, and Y. Zhang, “The roadmapto 6g: Ai empowered wireless networks,”
IEEE Communications Mag-azine , vol. 57, no. 8, pp. 84–90, 2019.[45] B. Peng and T. Kuerner, “Three dimensional angle of arrival estimationin dynamic indoor terahertz channels using forward-backward algo-rithm,”
IEEE Transactions on Vehicular Technology , pp. 3791–3811,2016.[46] W. Saad, M. Bennis, and M. Chen, “A vision of 6g wireless systems:Applications, trends, technologies, and open research problems,”