Cellular systems - 5G and beyond Part 2¶

约 557 个字预计阅读时间 2 分钟

Stefan Parkval, IEEE fellow, Sernior Expert Ericsson Research¶

1.Standardization in practice¶

Why¶

standard can be de-facto, de jure, voluntary, where cellular network is the last one
Technology development - selecting the best solution
interoperability - samsung phone in Ericsson network
mass market - scale effect of economy

What¶

logical arch
protocol on interfaces
RF/ Radio transmitter, law consideration
behavior, functionality

not do:(where competetion exists) - physical implementation - algorithm

Where¶

standard-developing org.: 3GPP, IEEE, ETSI, O-RAN
- nonprofit
- technical
- global/reginal/national
Regulatory bodies: ITU, FCC, ECC, …
- governmental
- spectrum usage
- products
Insdustry fora: wifi, GSM world
- promoting and lobbying for specific technologies

Process¶

stage 1 Requirement
stage 2 logical architechture, fucntional split, interfaces, rough solution
stage 3 details: header format, coding scheme, quantitizaed requirements
Test: test cases

ITU¶

founded in 1865, second oldest international organization stll running
Rhine navigation is the first)
standardization
global (almose) radio specturm allocation

ITU-R¶

Radio regulations:
- different frequency bands
- mandatory technical parameters to be observed
Reports
Recommodations:
- approves standards, such as, from 3GPP

IMT-2000 (3G)¶

6 technologies: CDMA, CDMA-TDD, …

IMT-advanced (4G)¶

LTE

WiMAN (wimax): never used

IMT-2020 (5G)¶

NR

DECT-2020: never used

IMT-2030 (6H)¶

24-26 : requirement

27-28 : proposal

29-30 : Eval&Spec

3GPP¶

founded in lalet 90s, for 2G standards

(although ‘G’ concept hadn’t been invented then)

Org¶

TSG: RAN, SA (Service&overall system), CT (core network)

RAN WG1(RAN1), RAN WG2, ....

SA1, SA2, … SA6

Standardizatin in practice¶

contribution driven
decision by cnosensus, including small talks in coffe-breaks
- tech, commercial, political …
socials
one week meetings, long meeting dats
no voting

ORAN¶

not buy basestation products, but buy components, with common interfaces across different companies.

started 2018
operator-driven, including CMCC
all big player except huawei

problem: - two complicated. - highly integration is required by real-time performance, robustness ....

But at least open frontaul is good (from the Ericsson’s perspective)

Because anyway further decoupling of fraontaul and baseband is necessary.

What is the role of academia in standarization

none, but impact the standardization personal reflection from speaker: academia is too close to 3GPP, no one is aiming at 2040!

6G¶

officially started in march 2025
now 6G requirements stage

requirement¶

joint capabilities
application determined

princeiples¶

standalone (should not have NSA 6G)
interface an 5G Core, which is flexible enough, and just deployed by operators

different from NSA between 4G and 5G.

6G phone only connects 6G RAN, shall not support any 5G phone, to avoid NSA issue like the transition from 4G to 5G.
intent-based and programmable

more granularity on commercialized service
selected open interfaces
operatee in all existing 3GPP bands, and in new cmWave bands
spectrum sharing between 5G and 6G
new and evolved use cases

overall design guideline¶

Keep: waveform, coding(LDPC; polar), modulation(OFDM), mimo and multi-antenna features
Change: aggregation (UL-DL decoupling! best DL may not be the best UL), energy performance, scheduling and control signalling
Add: ovservability (collect detail statistics from network) resillient communications isac service-level agreement
- intent driven netwwork to automated SLA management: to this task, the network will provide this kind of service and charge in this scheme…

Massive IoT¶

Ambidient IoT, replace NBIoT or CAT-M, without sacrificing 6G performance

should consider from day 1 of 6G

Non-terrestrial access¶

as a complement to terrestrial access

deployment: focus on LEO, both handheld and VSAT

avoid dependency on externall technologies such as GNSS

ISAC¶

reuse the communication specturm and infra