Tuesday 5 August 2014

New types of HetNet's Cell coordination

Came across this HetNet Infographic from Ericsson here. They are proposing something interesting as can be seen in the picture above. From the infographic:

Macro for downlink, micro for uplink
In the imbalance area, the downlink signal from the macro is the strongest, because it transmits at a much higher power, whereas the uplink signal from the pico node is much stronger, because it is closer. This provides the user with significantly higher uplink speeds than would be possible with macro coverage alone.
So if we have a macro with an overlay of small cells then the Macro can be used for DL and Small Cells for UL. This scheme is a counterintuitive to what I would have thought. Since there is a higher requirement in DL as opposed to UL, the uplink could be received by Macro and the DL could be sent from pico node. The macro also has bigger antennas and can process weak signals from the UE.


Ericsson and the South Korean operator, SK Telecom recently also demonstrated 'Elastic Cell'. As per Telecom TV:
The telco has confirmed that Elastic Cell technology is based on the 3GPP Release 11 CoMP specification, but with improved scheduling, energy efficiency, and cost efficiency. SK Telecom says it has applied downlink CoMP since early 2012 and uplink CoMP in April 2014, and both technologies are proprietary technology. Because coordination between networks will still be very important in 5G technologies, SK Telecom expects that Elastic Cell will become a key enabler for 5G.

Another similar approach that is proposed by NTT Docomo is the 'Phantom Cell' concept as described here. Their proposal is to separate control and user planes. Macro used for signalling (C-plane) and Small cells in higher frequencies for data (U-plane)

Finally, we also have the SK Telecom's SUPER Cell concept and blogged here. There is a lot of cell splitting in this but again we have the main frequency (lower) being able to do both control and data while the higher frequency is only to do data. Sounds a bit like the Phantom Cell and 'New Carrier Type' as blogged here before.

Are there any other types of cell coordination being discussed. Do you have any opinion on them? Feel free to add comments.


  1. Prem Prakash (via LTE / LTE-A 3GPP Rel-10-11-12 Linkedin group)6 August 2014 at 07:29

    It Looks like the concept is based on the DVB-SH where the DL coverage was proposed by the satellite and UL was with Radio base station.

  2. Manoj Das (via HetNet Linkedin group)6 August 2014 at 10:10

    Thanks Zahid, this is amazing! WSPs will do "whatever it takes" to provide better user experience. All possible combinations are being tried. Some of my ex-coworkers are working (and accumulating Intellectual Properties) on integrating WiFi in to this mix and match of technologies.

  3. Ericsson have always favoured the operator/vendor-managed and coordinated approach that produces the highest spectral efficiency in the most congested areas (at the highest cost). A key question is how much of the time this is needed. The eCoMP technology requires high quality fibre to connect the cells under an eNodeB - so the example given is usually for a single vendor eNodeB with remote radio heads for the pico cells under that node, although the SK Telecom elastic cell example has similarities to distributed MIMO.
    For the Phantom Cell concept and new carrier type, this is again tight integration for the most demanding areas and I feel has a lot of good points to it. However 3GPP simulations of the New Carrier Type were inconclusive on its benefits in Rel-12, particularly as the new, nearly "cell-free" carrier wouldn't be accessible to UEs from previous releases