WHITE PAPER
6G-SHINE: ADVANCING
CORE TECHNOLOGIES
FOR IN-X SUBNETWORKS
OCTOBER 2025
DOI: 10.5281/zenodo.17338153
URL: h ps://doi.o g/10.5281/zenodo.17338153
2
TABLE OF CONTENTS
EXECUTIVE SUMMARY ............................................................................... 3
ACKNOWLEDGEMENT .................................................................................................. 3
ABBREVIATIONS ......................................................................................... 4
1. INTRODUCTION ...................................................................................... 5
2. USE CASES FOR IN-X SUBNETWORKS ...................................................... 5
2.1 ARCHITECTURAL ENABLERS ................................................................................... 9
3. IN-X SHORT-RANGE RADIO CHANNEL CHARACTERIZATION .................. 12
4. PHYSICAL LAYER AND MEDIUM ACCESS CONTROL ENABLERS .............. 13
5. RADIO RESOURCE MANAGEMENT ........................................................ 16
6. MANAGEMENT OF TRAFFIC, SPECTRUM AND COMPUTATIONAL
RESOURCES .............................................................................................. 18
7. DEVELOPED PROOF-OF-CONCEPTS ....................................................... 19
8. OUTLOOK ............................................................................................. 20
3
EXECUTIVE SUMMARY
The 6G-SHINE p ojec se ou o design he ounda ional echnology componen s ha
will enable he nex gene a ion o in-X subne wo ks—sho - ange, low-powe adio cells
ope a ing a he e y edge o he 6G “ne wo k o ne wo ks.” These subne wo ks a e
en isioned o p o ide pe asi e wi eless co e age wi h unp eceden ed pe o mance
equi emen s. They should be able o ope a e independen ly, much like ad hoc ne wo ks,
wi h he ad an ages o seamless in eg a ion in o he b oade 6G sys em, which can
suppo spec um and compu a ional esou ce managemen , au hen ica ion and
au ho iza ion, a ic policy en o cemen , and da a o loading.
The p ojec ad anced esea ch ac oss mul iple echnical domains, including he physical
laye , medium access con ol, adio esou ce managemen , and ne wo k a chi ec u e.
The ambi ion was o c ea e a po olio o inno a ions ha can ans o m in-X subne wo ks
om concep in o eali y. This in ol ed no only de eloping mechanisms o local
communica ion bu also es ablishing how such subne wo ks can be e icien ly managed,
secu ed, and in e connec ed wi h he wide 6G amewo k.
Th ough his wo k, 6G-SHINE has laid he g oundwo k o u u e wi eless sys ems in
which g oups o de ices can o m in elligen , au onomous, and highly e icien
subne wo ks ope a ions. These ad ances open he doo o new applica ions in
consume , indus ial, and ehicula domains, whe e local connec i i y, obus ness, and
esponsi eness a e c i ical.
This whi epape is based on he 6G-SHINE p ojec deli e ables and p esen s a concise
o e iew o he main inno a ion de eloped in he p ojec .
Acknowledgemen
This whi epape is based on he deli e ables o he 6G-SHINE p ojec (Ho izon Eu ope
G an Ag eemen No. 101095738), publicly a ailable in he p ojec websi e. This p ojec
is unded by he Eu opean Union. Views and opinions exp essed he e a e howe e hose
o he au ho (s) only and do no necessa ily e lec hose o he Eu opean Union o he
Eu opean Heal h and Digi al Execu i e Agency. Nei he he Eu opean Union no g an ing
agency can be held esponsible o hem.
4
ABBREVIATIONS
Abb e ia ion
Meaning
3GPP
3 d Gene a ion Pa ne ship P ojec
AI
A i icial In elligence
AR
Augmen ed eali y
BLER
Block E o Ra e
CN
Co e Ne wo k
CSI
Channel S a e In o ma ion
DNN
Deep Neu al Ne wo ks
ECU
Elec onic Con ol Uni
GDRA
G adien Descen -based Resou ce Alloca ion
GNN
G aph Neu al Ne wo ks
GW
Ga eway Role
HC
Elemen wi h High Capabili ies
HPCU
High P ocessing Con ol Uni
ISAC
In eg a ed Sensing and Communica ion
LC
Elemen wi h Low Capabili ies
LOS
Line-o -Sigh
LSTM
Long Sho -Te m Memo y
MAC
Medium Access Con ol
ML
Machine Lea ning
NLOS
Non Line-o -Sigh
O-RAN
Open Radio Access Ne wo k
OFF
Compu e O loading
PoC
P oo -o -Concep
PPO
P oximal Policy Op imiza ion
RAN
Radio Access Ne wo k
RIS
Recon igu able In elligen Su ace
RRM
Radio Resou ce Managemen
SL
Sidelink
SN
In-X Wi eless Subne wo k
SNE
Subne wo k Elemen
SNM
Subne wo k Managemen
SNR
Signal- o-noise a io
TRL
Technology Readiness Le el
UE
Use Equipmen
URLLC
Ul a- eliable Low La ency Communica ion
VAA
Vehicle Access A ea
XR
Ex ended Reali y
5
1. INTRODUCTION
The 6G Sho ange ex eme communica ion IN En i ies (6G-SHINE) p ojec , unded
unde he SNS Wo k P og amme Phase 1 S eam B, a ge s he design o co e
echnology componen s o wi eless in-X subne wo ks—sho - ange, low-powe adio
cells deployed a he e y edge o he 6G “ne wo k o ne wo ks.” These subne wo ks aim
o deli e pe asi e wi eless co e age while mee ing unp eceden ed pe o mance
equi emen s. They should be able o ope a e independen ly simila ly o ad hoc
ne wo ks, while bene i ing om connec ion wi h he 6G pa en ne wo k, ha can ease
managemen o spec um and compu a ional esou ces, au hen ica ion / au ho iza ion,
policy en o cemen o a ic o/ om de ices in he subne wo k, a ic o loading.
6G-SHINE ad anced esea ch in he physical laye , medium access con ol, adio
esou ce managemen , and ne wo k a chi ec u e, wi h he ambi ion o assembling a
po olio o echnology componen s ha can u n in-X subne wo ks om concep in o
eali y.
The p ojec b ough oge he a conso ium o 12 pa ne s: Aalbo g Uni e si y (Denma k),
CNIT (I aly), Uni e sidad Miguel He nández de Elche (Spain), Apple (Ge many), Sony
(Sweden), Bosch (Ge many), In e digi al (UK), F aunho e IIS (Ge many), imec
(Belgium), Keysigh (Finland), Nokia (Denma k), and Cogninn (G eece).
In his whi epape , we p esen a concise o e iew o he inno a i e ideas and concep s
de eloped in he p ojec .
Sec ion 2 p esen s he main use cases en isioned o in-X subne wo ks, co e ing he
consume , indus ial, and in- ehicle domains, along wi h he main a chi ec u al enable s.
Sec ion 3 summa izes he adio channel cha ac e iza ion ac i i ies un in he p ojec ,
co e ing di e se use cases and equency bands.
The de eloped physical laye and medium access con ol enable s a e p esen ed in
sec ion 4, while sec ion 5 ocuses on adio esou ce managemen esea ch.
Sec ion 6 p esen s ou esea ch ou come o he managemen o a ic, spec um and
compu a ional esou ces. The ealized PoCs a e p esen ed in sec ion 7. Finally, Sec ion
8 p o ides concluding ema ks, and an ou look o u u e ac i i ies.
All p ojec deli e ables a e a ailable on he p ojec websi e (6gshine.eu), wi h e e ences
p o ided h oughou his whi epape o link he inno a ions p esen ed o hei
co esponding deli e ables. Fo a comp ehensi e desc ip ion o he inno a ions, he
eade can e e o he p ojec deli e ables.
2. USE CASES FOR IN-X SUBNETWORKS
In-X subne wo ks a e sho - ange, low powe adio cells o be ins alled in en i ies such
as obo s, p oduc ion modules, ehicles, and class ooms, o deli e highly localized and
high-pe o mance connec i i y. We ha e iden i ied 13 ele an use cases o in-X
subne wo ks, co e ing he consume , indus ial and in- ehicle domains. The use cases,
pic o ially depic ed in Figu e 1, 2 and 3, ha e ex ensi ely been desc ibed in deli e able
D2.2, along wi h hei pe o mance equi emen s, and a e summa ized in Table 1.
6
Table 1 6G-SHINE in-X subne wo ks use cases
No
Ti le
Sho Desc ip ions
C
-1
Imme si e Educa ion
Imme si e Educa ion aims o enhance he
in e ac i e expe ience o a g oup o s uden s and
eache (s) o knowledge exchange, le e aging
media con en and ela ed echnologies (e.g., XR
de ices).
C
-2
Indoo In e ac i e Games
XR in e ac i e gaming in an indoo en i onmen
whe e one o mo e playe s play in a place whe e
i has been equipped and p e
-
loaded wi h some
equipmen o acili a e he XR in e ac i e gaming.
C
-3
Vi ual Li e P oduc ion
One o mo e pe o me s ha can be loca ed in
di e en geog aphical a eas p oduce 3D ideo
con en ha can be li e
-
b oadcas ed o uploaded
o social media.
C
-4
AR Na iga ion
AR na iga ion powe ed by AI/ML concie ge based
on use inpu , including he in o ma ion a ound he
use . The ou pu is p o ided o he use ia an AR
de ice.
I
-1
Robo Con ol
The wi eless con ol o obo ope a ions, such as
he con ol o mul i
-
axis obo s o le e aging he
deg ees o eedom o e ed by po en ial
mo emen di ec ions he obo can accomplish.
I
-2
Uni Tes Cell
To pe o m quali y assu ance asks o p oduc
pa s in he manu ac u ing p ocess, as well as o
de ices used in he manu ac u ing p ocess.
I
-3
Visual Inspec ion Cell
A isual inspec ion cell pe o ms quali y assu ance
in he manu ac u ing p ocess h ough ideo eeds.
The ideo eeds a e p ocessed, and quali y con ol
is pe o med, by e en ually ou pu ing commands
o ac ua o s in case ac ions a e o be aken o
imp o
ing ope a ion quali y.
I
-4
Subne Co
-
exis ence in Fac o y
Hall
Tasks dis ibu ions among a swa m o smalle ,
specialized obo s. Each obo is con igu ed o
pe o m a speci ic unc ion. Wo king in conce ,
hese obo ic swa ms can assemble in ica e
p oduc s
I
-5
Subne wo k Segmen a ion and
Managemen
Combina ion o all he o he indus ial use cases
wi h a ocus on he secu i y and managemen
aspec s, pa icula ly on unc ional equi emen s.
7
V
-1
Wi eless Zone ECU: in
- ehicle
wi eless subne wo k zone
In
-
ehicle zone wi eless subne wo k ha is u ilized
by some senso s and ac ua o s loca ed in his
zone o connec wi elessly o he zone
Elec onic
Con ol Uni
(ECU)
ha manages and con ols
hem. The senso s and ac ua o s ha a e
wi elessly connec ed o he zone ECU a e
equipped wi h a 6G
-
capable wi eless
communica ion in e ace ha eplaces hei o me
wi ed communica ion in e ace.
V
-2
Collabo a i e Wi eless Zone
ECUs: Func ions ac oss
mul iple in
- ehicle zones
This use case co e s au omo i e sys ems and
applica ions ha equi e (o bene i om)
collabo a ion o o loading be ween unc ions,
senso s and ac ua o s loca ed a di e en zones
o he conside ed 6G
-SHINE e e ence in-
ehicle
E/E a chi ec u e
, ia a High P
ocessing Con ol
Uni (HP
CU).
V
-3
In e
-subne wo k Coo dina ion:
Collabo a ion/in e e ence/RRM
be ween subne wo ks in
in a/in e
- ehicle
communica ions
Two cases o subne wo k coo dina ion on
RRM/in e e ence handling whe e he sub-
ne wo ks can be wi hin he same ehicle (In a-
ehicula ) o in some o he ehicle(s) (In e -
ehicula ).
V
-4
Vi ual ECU: In
- ehicle senso
da a and unc ions p ocessing
a he 6G ne wo k edge
In eg a ing he in
-
ehicle ne wo k wi h he 6G
pa en ne wo k, o seamlessly ex end he in-
ehicle embedded compu e capabili ies o he
edge/cloud.
Figu e 1. Use cases o consume subne wo ks.
8
Figu e 2. Use cases o indus ial subne wo ks.
Figu e 3. Use cases o in- ehicle subne wo ks.
Consume Subne wo ks:
Use cases in consume subne wo ks ypically demand high da a a es and low la ency.
Compu a ion o loading is commonly employed, pa icula ly o ideo ende ing and
AI/ML da a p ocessing. A baseline le el o eliabili y is essen ial o minimize
e ansmissions, which could o he wise inc ease la ency. These scena ios in ol e a
di e se se o KPIs ac oss di e en links—such as one o mo e high- h oughpu da a
s eams (e.g., uplink/downlink ideo ames) accompanied by mul iple low- a e senso
da a s eams. Addi ionally, synch oniza ion among mul iple media s eams—including
ideo, audio, and senso da a—is o en equi ed. Mee ing he simul aneous demands o
high h oughpu (up o 6.37 Gbps), low la ency, and high eliabili y p esen s signi ican
echnical challenges.
9
Indus ial Subne wo ks:
Indus ial use cases gene ally p io i ize ul a-low la ency and high eliabili y. Ce ain
scena ios equi e communica ion cycles below 100 µs. Al hough he da a a es pe
de ice a e ypically low, he numbe o connec ed de ices pe access poin (AP) can be
high. Fu he mo e, mul iple APs may need o ope a e concu en ly, necessi a ing a well-
designed ne wo k a chi ec u e o suppo s ingen la ency equi emen s and enable
po en ial compu a ion o loading be ween nodes. I is wo h no ing ha indus ial use
cases in ol ing ideo eeds may also equi e ela i ely high da a a es—up o
app oxima ely 50 Mbps.
In-Vehicle Subne wo ks:
In- ehicle use cases place a s ong emphasis on ex eme eliabili y, wi h some scena ios
demanding up o 99.9% eliabili y. These deploymen s o en in ol e a mix o medium-
a e links—such as hose ca ying image o ideo da a om came as—and low- a e links
ela ed o in e ac ions wi h a ious senso s and ac ua o s.
The use cases de ined wi hin he 6G-SHINE amewo k a e expec ed o con ibu e
meaning ully o key alue indica o s (KVIs) ela ed o social, en i onmen al, and
economic sus ainabili y.
Social sus ainabili y is add essed h ough enhancemen s in use expe ience, ensu ing
ha u u e connec i i y se ices deli e g ea e inclusi i y, accessibili y, and quali y o li e
compa ed o exis ing echnologies.
En i onmen al sus ainabili y is pu sued h ough a ge ed educ ions in ca bon emissions,
including bo h ca bon dioxide and me hane, alongside a dec ease in o e all ene gy
consump ion ac oss ne wo k ope a ions and end-use applica ions.
Economic sus ainabili y is add essed h ough di e en mechanisms depending on he
subne wo k con ex . In indus ial subne wo ks, he ocus is on imp o ing p oduc ion
e iciency, he eby enhancing compe i i eness and esou ce u iliza ion. In in- ehicle
subne wo ks, economic bene i s s em om he ex ension o ehicle li ecycles and he
p omo ion o ci cula economy p inciples. In consume subne wo ks, economic
sus ainabili y is suppo ed by enabling new digi al se ices in sec o s such as educa ion
and gaming, po en ially unlocking new ma ke s and inno a ion oppo uni ies.
Ac oss all use cases, low-la ency and eliable communica ions a e iden i ied as
ounda ional enable s o hese sus ainabili y objec i es. In addi ion, high da a a es a e
ecognized as essen ial in speci ic scena ios, pa icula ly wi hin consume subne wo ks,
o suppo ad anced mul imedia applica ions and eal- ime in e ac i i y.
Rela ed deli e able: D2.1, “Ini ial de ini ion o scena ios, use cases and se ice
equi emen s o in-X subne wo ks” , D2.2, “Re ined de ini ion o scena ios, use cases
and se ice equi emen s o in- X subne wo ks”.
2.1 A chi ec u al enable s
In-X subne wo ks, along wi h hei in eg a ion in he 6G ne wo k o ne wo ks, ep esen
a majo a chi ec u al inno a ion in he wi eless communica ion landscape. In-X
subne wo ks a e specialized ne wo ks designed o p o ide localized, high-pe o mance
16
5. RADIO RESOURCE MANAGEMENT
Subne wo ks can spon aneously become e y dense, such as hose ins alled in obo s,
p oduc ion modules, and ehicles. In e e ence is he e o e a majo challenge o be
dealing wi h in o de o suppo he demanding equi emen s o dense subne wo ks.
Signi ican p og ess has been achie ed in ad ancing eliabili y, scalabili y, low la ency,
spec al e iciency, and esilience o in e e ence h ough a uni ied sui e o cen alized,
dis ibu ed, and goal-o ien ed adio esou ce managemen (RRM) solu ions de eloped
wi hin he 6G-SHINE p ojec . Figu e 6 p esen s a pic o ial depic ion o he RRM con ex
o dense in-X subne wo ks.
Cen alized solu ions ely on he p esence o RRM capabili ies a he 6G pa en ne wo k.
Subne wo ks epo pe iodically hei expe ienced channel condi ions o he pa en
ne wo k, ha can pe o m decisions on he adio esou ces hey should use locally.
App oaches based on deep neu al ne wo ks (DNN) allow o educe he compu a ional
bu den wi h espec o i e a i e heu is ics in he execu ion phase, p o ided ha he DNN
has been p e iously ained in he en i onmen o in e es . To mi iga e he e ec s o
ou da ed channel s a e in o ma ion (CSI) due o subne wo k mobili y, and combining
spa io- empo al a en ion-based LSTM p edic ion wi h a obus deep neu al ne wo k
(DNN)-based esou ce alloca ion scheme. Unde ealis ic CSI delays (e.g., 4-samples
lag), his me hod ou pe o med s a e-o - he-a app oaches, deli e ing 53% highe
minimum spec al e iciency han sys ems wi hou p edic ion and 94% highe han hose
using s anda d LSTM p edic o s. These esul s we e alida ed in ex emely dense
Figu e 6. Radio esou ce managemen in dense in-X subne wo ks.
17
scena ios—up o 25,000 subne wo ks pe km²—suppo ing he p ojec 's goal o scaling
10× beyond cu en 5G ul a-dense deploymen s.
A complemen a y dis ibu ed RRM solu ion ailo ed o subne wo ks ha a e no in he
co e age a ea o a pa en ne wo k, based on G aph Neu al Ne wo ks (GNNs), enables
au onomous powe con ol in en i onmen s whe e cen alized coo dina ion is no
easible. This app oach imp o ed spec al e iciency by ~7% in uni o m se ings and up
o 13.16% in he e ogeneous channel condi ions o e equal powe alloca ion baselines,
while le e aging p ac ical o e - he-ai message exchanges compa ible wi h exis ing
3GPP sys ems.
Fo mission-c i ical applica ions in indus ial au oma ion, a goal-o ien ed RRM s a egy
was in oduced, co-op imizing ne wo k quali y and applica ion-le el me ics such as obo
mission comple ion ime. Using a P oximal Policy Op imiza ion (PPO) ein o cemen
lea ning algo i hm o mobili y con ol, he sys em inc eased he likelihood o mee ing
s ingen URLLC a ge s—achie ing 20% highe p obabili y o sus aining he same BLER
unde a 0.5 ms la ency cons ain compa ed o baseline app oaches.
In ligh o he ising ele ance o unlicensed and hyb id spec um use, 6G-SHINE
de eloped echniques o enable dense subne wo k ope a ion in such en i onmen s.
Semi-s a ic channel access me hods we e shown o suppo up o 10× mo e XR sessions
han con en ional dynamic schemes. When combined wi h licensed-assis ed ope a ion,
an addi ional 67% inc ease in suppo ed subne wo ks was achie ed. Fu he capaci y
gains—up o 40%—we e ealized by mi iga ing in-band emissions h ough
imp o emen s in de ice on -end design and spec um coo dina ion s a egies.
To add ess he pe asi e issue o ex e nal in e e ence, a dual-p onged s a egy was
employed: obus esou ce alloca ion and esilien ecei e design. A G adien Descen -
based Resou ce Alloca ion (GDRA) algo i hm limi ed spec al e iciency deg ada ion o
jus 9.7%, ou pe o ming exis ing benchma ks ha saw losses exceeding 13%. On he
ecei e side, low-complexi y likelihood a io app oxima ion echniques we e de eloped
o ensu e eliable decoding e en unde impulsi e noise and jamming, wi h dynamic
adap a ion o eal- ime channel condi ions and minimal p ocessing bu den.
Toge he , hese inno a ions demons a e he easibili y o deploying dense, au onomous,
and in e e ence- esilien subne wo ks ac oss di e se domains—including indus ial,
consume , and ehicula se ings. The RRM solu ions de eloped no only exceed cu en
pe o mance benchma ks bu also p o ide a obus ounda ion o enabling dynamic and
scalable 6G ne wo k a chi ec u es.
Rela ed deli e ables: D4.1 “P elimina y esul s on he managemen o adio esou ces
in subne wo ks in he p esence o legi ima e and malicious in e e e s”, D4.3 “Final
esul s on he managemen o adio esou ces in subne wo ks in he p esence o
legi ima e and malicious in e e e s”.
18
6. MANAGEMENT OF TRAFFIC, SPECTRUM AND COMPUTATIONAL
RESOURCES
Recen de elopmen s ha e add essed he managemen o a ic, compu a ion, and
spec um esou ces bo h among subne wo ks wi hin he same domain and be ween
subne wo ks and he b oade 6G in as uc u e. These ad ancemen s ha e buil upon
ea lie a chi ec u al amewo ks o de ine a comp ehensi e ope a ional model o
subne wo ks, p io i izing use p i acy, au onomy, and educed dependency on he
cen al ne wo k.
A new ca ego y o use equipmen has been in oduced, wi h ailo ed enhancemen s in
con igu a ion, secu i y, and da a mul iplexing. A decen alized au hen ica ion amewo k
suppo s di ec de ice- o-de ice in e ac ions, minimizing eliance on co e ne wo k
unc ions. Subne wo k o ma ion, egis a ion, and mobili y managemen ha e been
u he e ined h ough dis ibu ed con ol and use -plane unc ions, wi h added suppo
o local IP ou ing. To op imize con ol plane ope a ions, especially o asks like loca ion
upda es, o loading s a egies we e p oposed. Subne wo ks we e also enabled o
coo dina e wi h neighbo ing subne wo ks, suppo ing p edic i e mobili y and coo dina ed
measu emen unc ions, ul ima ely imp o ing indi idual use pe o mance.
The challenges o handling mul i-modal da a—especially ele an in consume
applica ions—we e also add essed. Two me hods we e de eloped o align da a packe s
ac oss in e ela ed lows, ei he by en iching packe heade s o de ice-side
synch oniza ion o by in oducing a managemen -side con olle ha handles scheduling
and alignmen . These me hods enhance quali y o se ice e en o locally-con ained
a ic. Imp o emen s in uplink scheduling we e also p oposed, enabling mo e e icien
esou ce use by he pa en ne wo k.
In compu a ion o loading, he amewo k now suppo s bo h local and decen alized
app oaches ac oss subne wo ks. Me hods o selec ing compu e nodes and managing
he associa ion be ween ope a ional and compu a ional en i ies ha e been o malized.
To suppo his shi , he quali y o se ice amewo k was expanded o inco po a e
compu ing-speci ic equi emen s and cons ain s, wi h new pa ame e s and p ocedu es
in oduced o suppo ing compu a ion-awa e se ice le els.
In ehicula applica ions—whe e opologies a e ela i ely s a ic, bu eliabili y
equi emen s a e ex emely s ingen —no el de e minis ic ask o loading and esou ce
alloca ion schemes we e p oposed. These p io i ize ask deadlines a he han me e
la ency minimiza ion, allowing o be e wo kload dis ibu ion ac oss IoT-edge-cloud
a chi ec u es. This imp o es scalabili y and pe o mance in sa e y-c i ical sys ems.
Fu he enhancemen s o in- ehicle ne wo ks include de e minis ic ask scheduling
schemes ailo ed o zonal elec onic/elec ic a chi ec u es, enabling eliable se ice
le els e en unde inc easing compu a ional demands.
Fo imme si e XR applica ions, a compu e- and ne wo k-awa e a ic s ee ing
amewo k was in oduced. I dynamically selec s op imal se ice ins ances and
manages mobili y-awa e se ice ancho ing and mig a ion, ensu ing seamless execu ion
in he e ogeneous and esou ce-cons ained en i onmen s, he eby enhancing bo h use
expe ience and quali y o se ice.
19
Finally, esea ch on dynamic spec um sha ing compa ed egula o y policies ac oss
majo global egions, e alua ed exis ing sha ing mechanisms, and highligh ed eme ging
ends. A lexible spec um access p o ocol was de eloped o allow dynamic alloca ion
o licensed esou ces o subne wo ks based on eal- ime a ic demands, using he
concep o dynamic esou ce pools o enhance spec al e iciency and esponsi eness.
Rela ed deli e ables: D4.2 “P elimina y esul s on he managemen o a ic,
compu a ional and spec um esou ces among subne wo ks in he same en i y, and
be ween subne wo ks and 6G ne wo k”, D4.4 “ Final esul s on he managemen o
a ic, compu a ional and spec um esou ces among subne wo ks in he same en i y,
and be ween subne wo ks and 6G ne wo k”.
7. DEVELOPED PROOF-OF-CONCEPTS
A selec ed se o echnologies de eloped wi hin he p ojec we e demons a ed h ough
p oo -o -concep s (PoCs) in labo a o y acili ies. Since 6G-SHINE is a low echnology
eadiness le el (TRL) p ojec , i was ou o ou scope o de elop a ully in eg a ed
p o o ype o an in-X subne wo k. Ins ead, he ocus was on showcasing speci ic
echnology componen s, selec ed on he basis o hei p omised pe o mance and
easibili y o implemen a ion wi hin he p ojec ime.
The demons a ed PoCs a e he ollowing:
1. Low-la ency channel emula o – Valida es sho - ange channel modeling o
ehicula , indus ial, and consume applica ions, signi ican ly imp o ing he
la ency pe o mance o he PROPSIM channel emula o in cap u ing sho - ange
channel e ec s.
2. La ency-awa e MAC access – Demons a es imp o ed coexis ence in sha ed
bands, p o iding de e minis ic medium access o la ency-sensi i e applica ions,
wi h li le o no penaliza ion o bes -e o a ic.
3. Jamming- esilien PHY – Shows how an enhanced schedule and decode can
main ain obus pe o mance unde bo h wideband and na owband jamming
condi ions.
4. In a-subne wo k mac o-di e si y – Uses ne wo k coding and coope a i e
mechanisms o imp o e eliabili y and la ency in indus ial subne wo ks.
5. Cen alized RRM – In oduces dynamic esou ce alloca ion mechanisms o
in e e ence mi iga ion in ehicula and indus ial en i onmen s.
6. Cen alized/dis ibu ed in e e ence managemen – Demons a es AI-d i en,
decen alized echniques o op imize ansmission powe and channel usage in
subne wo ks ope a ing o e o e lapping esou ces.
7. Jamming de ec ion ia anomaly de ec ion – P esen s anomaly de ec ion
solu ions capable o ecognizing malicious jamme s and he eby igge ing
app op ia e mi iga ion echniques in subne wo ks.
These PoCs con i med ha he echnologies de eloped in he p ojec can deli e
pe o mance gains compa able o hose obse ed in heo e ical and simula ion s udies,
20
while accoun ing o he cons ain s o he labo a o y en i onmen and he unde lying
ha dwa e and so wa e.
Rela ed deli e ables: D5.2 “Fi s implemen a ion and e alua ion o PoCs”, D5.3
“Second implemen a ion and e alua ion o PoCs, inal esul s”.
8. OUTLOOK
The 6G-SHINE p ojec has signi ican ly ad anced he ision o in-X subne wo ks, sho -
ange low-powe cells a he e y edge o he 6G “ne wo k o ne wo ks.” Scien i ically,
he p ojec has con ibu ed new models, a chi ec u es, and me hods ac oss he physical,
MAC, and esou ce managemen laye s. These esul s a e he building blocks o making
in-X subne wo ks (and hei in eg a ion wi h a 6G pa en ne wo k) a solid eali y in he
igu e wi eless landscape, and o m a ounda ion o u u e esea ch on esilien , e icien ,
and in elligen sho - ange wi eless sys ems.
F om an indus ial and economic pe spec i e, 6G-SHINE has al eady le a angible ma k
on he s anda diza ion landscape. The p ojec in oduced he subne wo k concep in
3GPP, con ibu ed o ETSI ISAC, and deli e ed inpu s o IEEE and ETSI RIS and THz
g oups. In pa icula , ou p oposed in- ehicle subne wo k use case has been al eady
cap u ed on 3GPP TR 22.870 “S udy on 6G Use Cases and Se ice Requi emen s”,
sec ion 11.10. This ea ly engagemen ensu es ha he echnologies and use cases
explo ed in 6G-SHINE can in luence global s anda ds a he e y momen when 6G
discussions a e consolida ing. The conso ium also gene a ed subs an ial in ellec ual
p ope y and ini ia ed collabo a ions ha s eng hen Eu ope’s indus ial compe i i eness
in wi eless echnologies.
Socie ally, he p ojec p omo es a sus ainable and human-cen ic app oach o 6G, wi h
localized p ocessing o p i acy and sa e y, educed cabling o g eene deploymen s,
and suppo o new applica ions in educa ion, indus y, and mobili y. Beyond i s echnical
ou comes, 6G-SHINE has also con ibu ed o he b oade SNS and 6G-IA ision
ac i i ies, ensu ing ha he p ojec ’s inno a ions a e isible and aligned wi h Eu ope’s
s a egic ambi ions.
In summa y, 6G-SHINE has posi ioned in-X subne wo ks as a c edible building block o
6G, b idging low-TRL esea ch wi h s anda diza ion and pa ing he way o hei
indus ial and socie al up ake.
