Saturday 15 April 2017

Self-backhauling: Integrated access and backhaul links for 5G


One of the items that was proposed during the 3GPP RAN Plenary #75 held in Dubrovnik, Croatia, was Study on Integrated Access and Backhaul for NR (NR = New Radio). RP-17148 provides more details as follows:

One of the potential technologies targeted to enable future cellular network deployment scenarios and applications is the support for wireless backhaul and relay links enabling flexible and very dense deployment of NR cells without the need for densifying the transport network proportionately. 

Due to the expected larger bandwidth available for NR compared to LTE (e.g. mmWave spectrum) along with the native deployment of massive MIMO or multi-beam systems in NR creates an opportunity to develop and deploy integrated access and backhaul links. This may allow easier deployment of a dense network of self-backhauled NR cells in a more integrated manner by building upon many of the control and data channels/procedures defined for providing access to UEs. An example illustration of a network with such integrated access and backhaul links is shown in Figure 1, where relay nodes (rTRPs) can multiplex access and backhaul links in time, frequency, or space (e.g. beam-based operation).

The operation of the different links may be on the same or different frequencies (also termed ‘in-band’ and ‘out-band’ relays). While efficient support of out-band relays is important for some NR deployment scenarios, it is critically important to understand the requirements of in-band operation which imply tighter interworking with the access links operating on the same frequency to accommodate duplex constraints and avoid/mitigate interference. 

In addition, operating NR systems in mmWave spectrum presents some unique challenges including experiencing severe short-term blocking that cannot be readily mitigated by present RRC-based handover mechanisms due to the larger time-scales required for completion of the procedures compared to short-term blocking. Overcoming short-term blocking in mmWave systems may require fast L2-based switching between rTRPs, much like dynamic point selection, or modified L3-based solutions. The above described need to mitigate short-term blocking for NR operation in mmWave spectrum along with the desire for easier deployment of self-backhauled NR cells creates a need for the development of an integrated framework that allows fast switching of access and backhaul links. Over-the-air (OTA) coordination between rTRPs can also be considered to mitigate interference and support end-to-end route selection and optimization.

The benefits of integrated access and backhaul (IAB) are crucial during network rollout and the initial network growth phase. To leverage these benefits, IAB needs to be available when NR rollout occurs. Consequently, postponing IAB-related work to a later stage may have adverse impact on the timely deployment of NR access.


There is also an interesting presentation on this topic from Interdigital on the 5G Crosshaul group here. I found the following points worth noting:

  • This will create a new type of interference (access-backhaul interference) to mitigate and will require sophisticated (complex) scheduling of the channel resources (across two domains, access and backhaul).
  • One of the main drivers is Small cells densification calling for cost-effective and low latency backhauling
  • The goal would be to maximize efficiency through joint optimization/integration of access and backhaul resources
  • The existing approach of Fronthaul using CPRI will not scale for 5G, self-backhaul may be an alternative in the shape of wireless fronthaul

Let me know what you think.

Related Links:



10 comments:

Unknown said...

Great insight

Ramses Rodríguez Ferguson (via 3GPP 5G Standards Linkedin Groups) said...

I think the self-backhauling concept (as integral to 5G NR) will be more than just a 2nd choice particularly in regions where dense fiber availability (both for backhaul and fronthaul) will always be prohibitive.

Whereas mmWave frequencies - for the short expected inter-Cell distances - will allow ample spectrum availability for backhaul without impacting the radio access (where there will also be sub-6GHz access-bearing spectrum capability).

Fatih Pazarlıoğlu (Linkedin) said...

I believe that 5g will be driven by massive small cells or unique smart integrated radio equipments which combine traditional antenna,rru and bbu's itself. Backhaul is an important deployment cost for the small cells.If backhaul access interference issue can be eliminated in this innovative solution , this will be important milestones for the 5g backhaul problems.

Dr. Mehdi Bennis (Linkedin) said...

interesting topic, this was looked at back last year, in this award-winning paper:
Joint In-Band Backhauling and Interference Mitigation in 5G Heterogeneous Networks

https://arxiv.org/pdf/1604.02750.pdf

Fatih Pazarlıoğlu (Linkedin) said...

This may also be evaluated for FTTx modem backhaul as WTTx.The important challange is the capacity management of radio carrier spectrums between MBB and FBB users.I know one vendor has an WTTx solution which can also be used in LTE right know.

Miguel Patiño González (Linkedin) said...

These schemes recall kind of Ad-hoc architecture between BS. If this was done, it should be self-recoverable and self-configurable. Critical part is balancing access/backhaul traffic ratio. Definitely, more Machine Learning and AI in general will be required to achieve it.

Srikrishna Komatineni (Linkedin) said...

applications can be for both FTTx. WTTx or even for IoT. Bottom line is the self-organization as Miguel mentioned. If this is achievable, then we are gonna see massive change in the way we design, operate & optimize the xhaul. Plenty of opportunities for ML + SW.

Stephen Sedio (via 3GPP 5G Standards Linkedin Groups) said...

My son's company is a rural WISP (wireless internet service provider). He started with a few access points, and used customers as access points for down stream customers until the number of customers justified a dedicated access point / back haul. It speed deployment with little to no impact on usability.

In addition, 5G is not a long reach technology. the number of access points required for coverage lends itself to self back hauling. Simplifying installation, and can add redundancy**.

As mentioned, the problems are latency, interference, and bandwidth. Latency will limit the number of hops, bandwidth, as the number of users grow. As this becomes an issue, the access point can be upgraded point to point wireless back haul, or fiber.

**If one fiber is cut, the system can find a path to another good fiber. Even if every access point is connected to fiber, I recommend self back hauling as a fail-safe, slightly higher latency is better than no connectivity.

Rajan Mahajan said...

There are many challenges in 5G technology. It's a very informative and useful blog. thanks for sharing this.

Unknown said...

Technically it is a viable and desirable solution but there is the old issue of spectrum allocation/costs. So long the industry has been working on mmW and still the spectrum is not harmonized and available at a reasonable price on many countries, as an example!!