Truly,Grant-Free,Technologies,and,Protocols,for,6G

MA Yihua, YUAN Zhifeng, LI Weimin, LI Zhigang

DOI：
10.12142/ZTECOM.202104012

https：//kns.cnki.net/kcms/detail/34.1294. TN.20211022.1500.004.html， published online October 22， 2021

Manuscript received：
2021-04-16

Abstract：
The further integration of telecommunications and industry has been consider? able and is expected to bring significant benefits to society and economics in 6G. It also forms some evolution trends for next-generation communication systems， including further rises in machine-type communications （MTC）， uplink-dominated systems， and decentral? ized structures. However， the existing access protocols are not friendly to these trends. This paper analyzes the problems of existing access protocols and provides novel access technologies to solve them. These technologies include contention-based non-orthogonal multiple access （NOMA）， data features， enhanced pilot design and successive interference cancellation （SIC） of diversity. With these key enablers， truly grant-free access can be re ? alized， and some potential modifications of protocols are then analyzed. Finally， this paper uses massive and critical scenarios in digital transformations to show the great necessity of introducing novel access technologies into future communication protocols.

Keywords：
decentralization; digital transformations; future access protocols; MTC; uplink

Citation （IEEE Format）：
Y. H. Ma， Z. F. Yuan， W. M. Li， et al.， “Truly grant-free technologies and protocols for 6G， ”ZTE Communications， vol. 19， no. 4， pp. 105 – 110， Dec. 2021. doi：
10. 12142/ZTECOM.202104012.

1 Introduction

The targets of communications vary from connecting hu ? mans to connecting everything， and digital transforma? tionsareexpectedtopenetratemanyscenariosin 6G[ 1 –3]. Digital transformations are expected to bring significant benefits to society.Meanwhile， digital transforma? tions raise many new requirements， as well as set the course for future communication protocols.

To support seamless and handy digital services， there are some trends in future communications， including further evo? lution from human-oriented communications to machine-type communications（MTC）， from the downlink-dominated to up? link-dominated，andfromcentralizationtodecentralization. This paper focuses on the truly grant-free， or contention-based grant-free， technologies and protocols， which are crucial to ful? filling ultra-low latency and ubiquitous connectivity for future communications. In current 5G standards， the grant-free defi? nition is not truly yet as it requires extra pre-configuration of grants.

Novel access technologies are required to replace the exist? ing ones for realizing these trends， and this paper introducesfour key enablers：
the contention-based non-orthogonal multi? ple access （NOMA）， the prior knowledge of data， the enhanced pilot design， and the joint use of diversity and successive inter? ferencecancellation（SIC）.NOMA was first proposedinRef. [4]， and it usually requires accurate power control and resource allocation. As a comparison， the contention-based NOMA has no such requirements and achieves a higher flexibility. The pri? or knowledge of data can be used to improve transmission per? formance.Data-onlytransmissionwasproposedinRef.[5]， which is able to remove pilot overheads. Apart from the data-on? ly scheme， a pilot is still crucial to the full use of a large receiv ? ing antenna array where the spatial combining search space is huge. Therefore， thepilotshouldbe welldesignedtogaina good detection and estimation performance[6 –7]. Moreover， the joint use of diversity and SIC[8]can deal with the fluctuation of loading as it is random in contention-based transmissions. This strategy is able to average the loading in different slots via itera ? tive detection and cancellation.

With these key enablers， users are able to transmit packets withoutbuildingradioresourcecontrol（RRC）connections. That is to say， a connection-free transmission can be achieved，andthetransmissionnolongerreliesonRRCconnections. Thisgreatenhancementsimplifiesthetransmissionproce ? dures a lot， as well as reduces the overheads and latency for building RRC connections. It also leads to some possible mod ? ifications of protocols in the future， which are also analyzed in this paper.

The motivation of this paper is to discuss the importance of truly grant-free technology for future communications， summa? rize key enablers for it， and suggest some protocol evolution considerations. The rest of this paper is organized as follows. Section2introduces themain trendsof futureradioaccess networks and analyzes the problems of existing access proto ? cols. In Section 3， some novel access technologies are provid ? ed to solve the problems. Section 4 briefly introduces the im ? pact of these technologies on future protocols. InSection 5， two important scenarios are discussed to show the advantages of the proposed schemes. Section 6 concludes this paper and provides some future research directions.

2 Three Main Trends

In this section， three main trends of future networks are in ? troduced. The problems of existing access protocols are ana? lyzed under these trends.

2.1 Further Rises in MTC

The first trend is that the main participants of communica? tions vary from humans to machines. Although massive ma? chine-typecommunication（mMTC）isincludedin5G，more massive and critical MTC will develop and be in demand to ? wards 2030 and beyond[ 1]. Unlike human communications， the potential users can be massive and the reliability requirement can be very high. However， the classical random access proto ? col requires a handshaking procedure， which is not suitable for the high-efficiency transmissions of massive potential us ? ers or low-latency transmissions of high-mobility networks.

2.2 Uplink-Dominated System

The second trend is that the overall performance is becom?ing dominated by uplink instead of downlink transmissions. For many MTC applications， uplink is the main bottleneck[9] . Moreover， in massive multiple input multiple output （MIMO） systems， time division duplex （TDD） is much easier to realize; it uses uplink-downlink reciprocity to obtain downlink chan?nel information. Direct downlink channel estimation is very in? efficient as the overheads of downlink pilots increase with the antenna number of the base station （BS） [10] . Therefore， the pilot in the uplink becomes very important for obtaining the chan?nel information and effectively using the capability of massive antennas of the BS. In the current protocol， the uplink pilot or demodulation reference signal （DMRS） is orthogonal among users. It limits the number of pilots and is not suitable for con?tention-based transmission.

2.3 Decentralized Structure

The last trend is from centralization to decentralization. This trend has many aspects， including the distributed anten?na or cell-free design for ubiquitous connectivity， device-todevice transmission not relying on the central controller， and decentralized information management and control， e. g. the blockchain[11] . Although network centralization has brought us many good aspects like easy management and global control， the costs should not be neglected， including coverage， latency and privacy risk. These costs greatly limit the performance and credibility of networks.

3 Novel Access Technologies

Thissectionanalyzesfournovelaccesstechnologiesas shown in Fig. 1. The basic idea and advantages of these tech ? nologies are discussed.

3.1 Contention-Based NOMA

NOMA is a very effective technology to increase spectrum efficiency when there is a near-far effect. Due to the complexi? ty limitation， a non-orthogonal power domain has not been ful? lyutilized especially for the uplink in current protocols. NO ? MA still acts as an important role in the future， as the com? plexity limitation is expected to be solved by advanced algo ? rithms and powerful hardware in the future.

It is complex and inefficient to implement accurate power control and resource allocation when there are massive users or the latency requirement is high. Therefore， a scheme allow? ing users to transmit freely is in high demand. To acquire this convenience for end devices， the transmission itself is inevita? ble to be non-orthogonal. In this case， extra sensing and local power control can be required to utilize the power domain[ 12] ， but it is not suitable for low-cost and low-power devices. To support a more flexible transmission not relying on any sens ? ing and power control， a joint use of power domain， code do? main and spatial domain should be considered as in Fig. 1（a）.

3.2 Data Features

To get rid of pilots or reduce the pilot overheads， data fea? tures should be used. There are two mainstream ways to real? ize this. One is to utilize the prior knowledge and statistical in ? formation of data[5， 13] ， e. g. the constellation shape， correlation matrix， constant modulus， etc. This method is compatible with existing protocols， and the modification to existing standards is relatively small. The other is a data-driven method that uses deep learning （DL）. The end-to-end auto-encoder is one impor? tant application of DL at the physical （PHY） layer， and Ref. [ 14] shows that pilot-free transmission can also be realized by an auto-encoder.

During the exploitation of data features， novel waveform po? tentiallyarises[15 – 16]. Discrete Fourier transform spreading or? thogonalfrequencydivisionmultiplexing（DFT-s-OFDM） plays an important role in 5G for its low peak to average powerratio （PAPR）. However， it makes the data feature hard to use. OnesolutionisrealFourierrelatedtransformspreading OFDM （RFRT-s-OFDM） [15]. As shown in Fig.1（b）， this novel waveform can maintain the data feature. Channel equalization and time/frequency offset correction can be conducted using thedatafeaturesinRFRT-s-OFDMasshowninFig.1（b）， while the low PAPR advantage is kept.

3.3 Enhanced Pilot Design

To support mMTC and massive MIMO， novel pilot designs are proposed to support more users and reduce allocation over? heads as shown in Fig. 1（c）. One mainstream research area is non-orthogonalpilotsbasedoncompressedsensing（CS） [6]. The sparsity of user activities， multiple paths and angles of ar? rival can be used to increase the performance of user detec ? tion and channel estimation. However， a CS-based non-orthog? onalpilotschemeleadstolargecomputationalcomplexity when the pilot pool is large or the receiving antenna array is large. Also， the inter-cell interference and time/frequency off? set problems are hard to solve.

The other research direction is the special non-orthogonal pilot design with partial orthogonality. To be specific， multi-pi? lot[7]is proposed which consists of multiple pilots. Multiple or? thogonal pilots are usually employed， although multiple non- orthogonal pilots also work. The detection of every orthogonal pilot is very simple， and the detection complexity is reduced a lot compared with general non-orthogonal pilot design. More ? over， an orthogonal pilot has been used and verified during a long period，andmanyexistingengineering methodscan be used to ensure the performance of multi-pilot use.

3.4 SIC of Diversity

The joint use of diversity and SIC was first proposed in Ref. [ 17]， and then some novel schemes have improved the perfor? mance by optimizing a bipartite graph of Fig. 1（d）. The joint use allows users to transmit multiple replica packets at any time slot， and SIC of packets is used after demodulating any packet in each round. This strategy was designed for satellite communications where the round-trip time is very long. The achievable loading of this strategy is approaching1 with the cost of replicas. This method greatly reduces the transmission delay and increases transmission efficiency. It is also named modern random access， which is seen as one potential next- generation random access protocol[ 10].

Unlike satellite communications， the channel gain of differ? ent users variesa lotin widely-used terrestrialcommunica? tions. Therefore， the near-far effect becomes a practical factor that requires consideration. NOMA is able to utilize the near- far effect to separate different users. The combination of NO ? MA and multiuser diversity with SIC requires a joint design of the medium access control （MAC） and PHY layers. An exam ? ple was shown in Ref. [ 18] which jointly uses the code domain NOMA and diversity with SIC.

4 Impacts on Protocols

With the novel access technologies， the transmission proce ? dure can be simplified a lot， which deeply affects the future protocols.

4.1 RRC Idle/Inactive

With the key enablers in Section 3， connection-free transmission becomes possible which enables high-efficiency and instant MTC without any connection establishment. That is to say， thetransmissioncanberealizedinanidleorinactive RRC state. These enablers can also evolve around the randomaccess protocols with a much higher successful rate and lower latency especially in dense environment， i. e.， they are benefi? cial for RRC connection establishment. A related standardiza? tion work is the two-step random access channel （RACH）[ 19]. However， it only acts as a substitution of the four-step RACH， and the collision problem has not been studied as orthogonal multiple access and orthogonal pilots are still in use.In regard to collisions， the protocols relating to multiple access and pilot sequences are expected to evolve.

4.2 Uplink and Sidelink NOMA

Asmentionedbefore，thenon-orthogonaldomainhasnot been well utilizedin5G. Actually， thediscussion of uplink NOMA has not reached a consensus in 5G standards[20]. One reason is that the performance and complexity comparison is not enough to decide a winner among different uplink NOMA schemes. The demands of mMTC should be further analyzed， and some crucial key performance indicators should be espe ? cially emphasized. The contention-based grant-free feature[21] is a potential scheme in the future as it is exceedingly friendly to low-cost and low-power MTC devices. Moreover， contention- based NOMA is also a potential standardization direction for sidelink（SL）withoutcentralcontrol，e.g.，LTE-VSL.

4.3 Comprehensive Synchronization

The connection-free transmission brings some synchroniza? tion （SYNC） challenges. In the current protocols， synchroniza? tion is realized by theSYNCsignal and measured time ad ? vance （TA） from BS[23]. TA is hard to be obtained in connec ? tion-free transmissions， so a comprehensive SYNC is required. Some possible modifications can be based on the information of the UE status （e.g.， position and speed）， BS position， SYNC signals of multiple BSes， etc. Moreover， an overall framework is required to jointly use all the prior knowledge relating to SYNC which can be obtained by the end device.

5 Case Studies

Two representative cases of both massive and critical MTC are shown in this section. The advantages of novel access tech ? nologiesshowthattheymayplayimportantrolesinfuture standards.

5.1 Case Study of MMTC

This case study is for mMTC， and a Poisson arrival model is used with an average arrival rate of λ. NOMA provides a large gain when there is a near-far effect， e. g.， a near and strong user can reuse the channel used by a far and weak us ? er， whichgreatlyimprovesthespectrumefficiency. Therefore， the near-far effect is included in comparison. As shown in Fig. 2， the outage performances of two different contention- based grant-free NOMA schemes[ 12， 18]are compared. As men ? tioned before， power domain NOMA with a contention-based feature requires extra sensing of channel gain and power con ? trol according to estimated channel gain. The aim is to make the receive power belong to some predetermined power lev ? els. It results in an extremely high transmission power， and Ref. [ 12] provides sub-channel selection and distance-based methods to solve it. As a comparison， code domain NOMA[ 18] is just required to randomly select the spread code and can work without power control.

The simulation results show that code domain NOMA per? forms better than power domain. For a target outage rate of 0. 1， the achievable arrival rate of power domain NOMA is less than 1 while that of code domain NOMA is greater than 2. Al? so， when combined with the SIC of the diversity strategy， the achievable arrival rates at the outage of 10-2are 2.4 and 3.4 for the power domain and code domain NOMA. In this compar? ison， power domain NOMA plus SIC of diversity achieves a lowererror floorbecauseperfectsensingandpowercontrol are assumed. The outage performance of code domain NOMA isboundedbyacorrespondingorthogonalmultipleaccess （OMA） transmission of perfect scheduling. As the near-far ef? fectisutilized，codedomainNOMAplusSICofdiversity achieves an arrival rate greater than 3 with the outage perfor? mance very close to the OMA bound.

5.2 Case Study of V2V

This case study is for vehicle-to-vehicle （V2V） communica? tions without central control. V2V communications should beextremely reliable for safety， and V2V without central control is especially important due to the robustness in a non-cellular domain and ultra-low latency requirement. In LTE-V， sensing- based semi-persistent scheduling （SPS）[22]is employed， and ev? ery user sensing the spectrum resources randomly selects idle resources in a given period. A sensing-based method is also studied and a potential candidate in 5G V2V is chosen[24]. The problem of a sensing-based method is that it is not friendly to multiple antennas， and the reliability is relatively low in ultra- dense environments. To solve these problems， Ref. [25] pro? poses a novel distributed antenna deployment and full-duplex contention-based NOMA transceiver scheme which jointly use power， code and spatial domains to achieve V2V communica? tions with ultra-low latency and high reliability in ultra-dense scenarios. The block error rate （BLER） comparison is shown in Fig. 3. The channel model is based on LTE-V standards[22] ， and 1 500 and 700 vehicles are dropped in the urban and free ? way scenarios defined in Ref. [ 10]， respectively.

In Fig. 3， three protocol-definedscenarios are all consid ? ered，includingurbannon-line-of-sight（UN），urbanline-of- sight （UL） and freeway line-of-sight （FL）. In this comparison， the full-duplex NOMA scheme achieves a much higher reli ? ability than the method in the current protocol. Also， the over? heads of the full-duplex NOMA scheme are reduced only 1/5， which means a much more frequent transmission of 250 pack? ets per second （pps） can be supposed， which helps to reduce the end-to-end latency once the information is generated. In this case， NOMA turns the near-far effect into advantages， and achieves a high spectrum efficiency and ultra-high reliability of near vehicles.

6 Conclusions

Tofulfilldigitaltransformations，itisnecessarytomake some great modifications in current protocols. One crucial as ? pect is to enhance the access protocols as it is a bottleneck of seamless and instant digital services. This paper provides sev? eral novel access technologies， and potential impacts on proto? cols are also briefly analyzed. Moreover， some typical use cas ? es are shown to verify the necessity of these modifications of future protocols. The standardization process of novel access technologies to boost digital transformations requires a much wider application to make it more urgent， more careful consid? erations given to privacy and security， and a joint work of in? dustry and academics to refine technologies.

References

[ 1] MAHMOOD N H， B?CKER S， MUNARI A， et al. White paper on critical and massivemachinetypecommunicationtowards6G[EB/OL].（2020-05-04） [2020-11-11]. https：//arxiv.org/abs/2004.14146

[2] ITU-T. FG-NET2030– Focus group on technologies for network 2030. [EB/ OL].[2020- 11- 11].https：//www. itu. int/en/ITU-T/focusgroups/net2030/Pages/ default.aspx

[3] DAVID K， BERNDT H. 6G vision and requirements：
is there any need for be? yond 5G？ [J]. IEEE vehicular technology magazine. 2018， 13（3）：
72 – 80. DOI：10. 1109/MVT.2018.2848498

[4] SAITO Y， KISHIYAMA Y， BENJEBBOUR A， et al. Non-orthogonal multiple access （NOMA） for cellular future radio access [C]//IEEE VTC Spring. Dresden， Germany：
IEEE， 2013. DOI：
10. 1109/VTCSpring.2013.6692652

[5] YUAN Z F， LI W M， HU Y Z， et al. Blind multi-user detection based on receive beamforming for autonomousgrant-free high-overloadingmultipleaccess[C]// IEEE 2nd 5G World Forum （5GWF）. Dresden， Germany：
IEEE， 2019：
520 – 523. DOI：
10. 1109/5GWF.2019.8911643

[6] KE M， GAO Z， WU Y， et al. Compressive sensing-based adaptive active user detectionandchannelestimation：massiveaccessmeetsmassiveMIMO[J]. IEEE transactions on signal processing， 2020， 68：
764 – 779. DOI：10. 1109/TSP.2020.2967175

[7] YUAN Z F， LI W M， LI Z G， et al. Contention-based grant-free transmission withindependentmulti-pilotscheme[EB/OL].（2020-04-07） [2020- 11- 11]. https：//arxiv.org/abs/2004.03225

[8] CLAZZER F， MUNARI A， LIVA G， et al. From 5G to 6G：
has the time for mod? ern random access come？ [C]//6G Wireless Summit. Levi， Finland：
6G Flagship，2019

[9] DAWY Z， SAAD W， GHOSH A， et al. Toward massive machine type cellular communications[J]. IEEE wireless communications， 2017， 24（ 1）：120– 128. DOI：
10. 1109/MWC.2016. 1500284WC

[ 10] ZHANG W C， XIANG J Y， LI Y-N R Y， et al. Field trial and future enhance ? ments for TDD massive MIMO networks [C]//26th Annual International Sympo? sium on Personal， Indoor， and Mobile Radio Communications （PIMRC）. Hong Kong， China：
IEEE， 2015：
2339 –2343. DOI：
10. 1109/PIMRC.2015.7343689

[ 11] MACDONALD M， LIU- THORROLD L， JULIEN R. The blockchain：
a com? parisonof platformsand their uses beyondbitcoin[EB/OL].[2020- 11- 11]. https：//www.researchgate.net/publication/313249614_The_Block? chain_A_Comparison_of_Platforms_and_Their_Uses_Beyond_Bitcoin

[ 12] CHOI J. NOMA-based random access with multichannel ALOHA[J]. IEEE journal on selected areas in communications， 2017， 35（12）：
2736 – 2743. DOI：
10. 1109/JSAC.2017.2766778

[ 13] MA Y H， YUAN Z F， HU Y Z， et al. A data-assisted algorithm for truly grant- free transmissions of future mMTC [C]//IEEE Global Communications Confer? ence.Taipei，China：IEEE，2020.DOI：10. 1109/GLOBE?COM42002.2020.9348198

[ 14] YE H， LI G Y， JUANG B-H F. Bilinear convolutional auto-encoder based pi ?lot-free end-to-end communication systems[C]//IEEE International Conference on Communications （ICC）. Dublin， Ireland：
IEEE， 2020. DOI：10. 1109/ICC40277.2020.9149030

[15] MA Y H， YUAN Z F， HU Y Z， et al. A real fourier-related transform spreadingOFDM multi-user shared access system[C]//90th Vehicular TechnologyConference （VTC2019-Fall）. Honolulu， USA：
IEEE， 2019：
1 –5

[ 16] YUAN Z， HU Y， MA Y， et al. Autonomous grant-free high overloading multiple access based on conjugated data symbols [C]//WS-20， IEEE InternationalConference on Communications. Dublin， Ireland：
IEEE， 2020

[ 17] CASINI E， GAUDENZI R D， HERRERO O D R. Contention resolution diversity slotted ALOHA （CRDSA）：
an enhanced random access scheme for satelliteaccesspacketnetworks[J].IEEEtransactionsonwirelesscommunications，2007， 6（4）：
1408 – 1419. DOI：
10. 1109/TWC.2007.348337

[ 18] MA Y H， YUAN Z F， LI W M， et al. NOMA based modern random access notrelying on sensing or power control [J]. IEEE iInternet of Things journal， 2021，8（20）：
15382 – 15395. DOI：
10. 1109/JIOT.2021.3073367

[ 19] 3GPP. Consideration on 2-step RACH procedures：
3GPP R1- 1903879 [S]. 2019

[20] 3GPP. Study on non-orthogonal multiple access（NOMA） for NR：
3GPP TR38.812 Release 16 [S]. 2018

[21] YUAN Y F， YUAN Z F， TIAN Li. 5G non-orthogonal multiple access （NOMA）studyin3GPP[J].IEEEcommunicationsmagazine，2020，58（7）：90– 96.DOI：
10. 1109/MCOM.001. 1900450

[22] 3GPP. Vehicle to vehicle （V2V） services based on LTE sidelink; user equipment （UE） radio transmission and reception：
3GPP TS 36.785 Release 14 [S].2016

[23] 3GPP. NR; Physical layer procedures for control：
3GPP TS 38.213 Release 16[S]. 2020

[24] 3GPP. Study on NR Vehicle-to-Everything （V2X）：
3GPP TR 38.885 Release16 [S]. 2020

[25] YUAN Z F， MA Y H， HU Y Z， et al. High-efficiency full-duplex V2V communication[C]//2nd6GWirelessSummit.Levi，Finland：IEEE，2020.DOI：10. 1109/6GSUMMIT49458.2020.9083762

Biographies

MA Yihua （yihua.ma@zte.com.cn） received the B.Eng. degree from Southeast University， China in 2015 and the M.Sc. degree from Peking University， China in 2018. Since 2018 he has been with ZTE Corporation， China. He is now a se?nior research engineer in the Department of Wireless Algorithm， ZTE. His main research interests include mMTC， grant-free transmissions， NOMA and cell-free massive MIMO.

YUAN Zhifeng received the M.S. degree in signal and information processing from Nanjing University of Post and Telecommunications， China in 2005. From 2004 to 2006， he was mainly engaged in FPGA/SOC ASIC design. He has been a member of the wireless technology advance research department at ZTE Cor?poration since 2006 and has been responsible for the research of the new multi?ple access group since 2012. His research interests include wireless communi? cations， MIMO system， information theory， multiple access， error control cod?ing， adaptive algorithms and high-speed VLSI design.

LI Weimin received the M.S. degree in communication and information system from Nanjing University of Posts and Telecommunications， China in 2010， and is currently working as a technology pre-research senior engineer at ZTE Corpo? ration. His research interests include power control， interference control， multi?ple access and grant-free transmission.

LI Zhigang received the B.S. degree from Jiangsu University， China in 2016 and the M.S. degree from Harbin Institute of Technology， China in 2019， and is currently working as a technology pre-research engineer at ZTE Corporation. His current research interests include wireless communication， multiple access and grant-free transmission.

推荐访问:Free Grant Protocols