Charting the path to RAN virtualization

Charting the path to RAN virtualization

C-RAN means

  • Centralized : Centralized processing resource pool supports 10-10,000 cells
  • Collaborative : Multicell joint scheduling and interference management
  • Cloud
  • Clean


  • H/W accelerator for L1 function FFT/IFFT channel encoding/decoding
  • “Recent Progress on C-RAN Centralisation and Cloudification” Chih-Lin I
  • CPRI compression
  • RAN and core within the shared NFV platform -> closer to each other and remove the sharp demarcation line that has separated the two.


  • One fiber connection provides 6Gbps
  • 9.8Gbps per carrier(FA). Doubled for bi-direction.
  • 3 carrier = 9.8G23 = 58.8Gbps
  • CPRI compression ration 50%
  • WDM(Wavelength-Division Multiplexing) : fewer finer connection compared to dark fiber which is native frontal solution that has the best performance
  • Three way for C-RAN/v-RAN adoption
    • Reduce frontal requirements by using functional split – enable some functions to be virtualised while the others remain distributed
    • Improve the efficiency of the fronthaul – pack more traffic
    • Use wireless alternative where fiber is not unavailable or too expensive

Dark-Fiber alternative

  • OTN(Optical Transport Network)
  • PON(Passive Optical Network)
  • CPRI over Ethernet(CoE)
    *outdoor small cell frontal in dense urban areas

    • higher latency than Dark-fiber
  • Wireless
    • outdoor small cell frontal
    • Suitable for short distance and relatively low capacity requirement

CPRI alternative

  • CPRI : 0.4ms latency, very low jitter
    • But not standard-based
    • ETRI’s ORI – Standardized CPRI for interoperability
  • Ethernet : should be carefully managed end-to-end to meet the CPRI’s performance

Split funtionalities

Functional split to reduce fronthaul requirement
PHY is L1 function which requires FFT/IFFT


Need 20-50x bandwidth than PHY-MAC
eICIC and CoMP can be implemented effectively, so the C-RAN can manage interference effectively.


  • Coordinated transmission can not implemented in the pooled BBU as there are L1 functions instantiated at the cell sited(PHY)
  • Macro-only network, this can be acceptable trade-off

Central scheduler based scheduling can be another option. MAC-level coordination among cells has lower latency and bandwidth requirements and ethernet can be used for this.

Functions in each layer

  • L3
    • Control RRC
  • L2
    • PDCP
    • RLC
    • Transport MAC
  • L1
    • CoMP
    • eICIC
    • Channel decoding/coding
    • Quantization/De-
    • Resource Block mapping
    • Sampling/De-
    • Modulatio/De-
    • FFT/iFFT

Cost reduction with vRAN

  • 30% in capex – mainly from equipment and installation cost
  • 54% in opex – lower site rental
  • Resource utilization
  • Mobility management
  • Robust interference management

Leveraging uneven traffic distribution

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s