Traditional wireless infrastructure relies on a monolithic basestation chassis sitting at the foot of the tower feeding signals back and forth to passive antennas mounted on top of the tower.
The connection between the basestation chassis and passive antenna components is via a coaxial cable as shown conceptually in Figure 1.
So only half of the signal power transmitted by the basestation chassis is received by the antennas.
To solve this problem designers use a distributed basestation architecture where radio cards (that host power amplifiers) are removed from the basestation chassis and mounted directly on the towers adjacent to the antennas. These radio cards are called remote radio heads (RRHs).
Common Public Radio Interface (CPRI) is one of the most commonly used protocols to transfer low-power modulated baseband signals from the channel cards to the radio. Signal loss in a fibre link is negligible for low power signal when compared to transferring a high power signal over a coaxial cable.
Despite the cost benefits, the adoption of a distributed baseastation architecture has been gradual. Mounting RRH on top of the tower results in installation (higher weight and wind loading), maintenance and reliability concerns.
Improvement in technology, higher integration, remote field-programmability and control, as well as size and weight reductions in the equipment are helping to overcome these hurdles.
In addition, the potential of distributed basestation architecture has opened a way to address the network capacity problem faced by wireless operators.
Increasing coverage and capacity
Wireless network operators have to continuously add capacity to the network to meet perpetual demand for higher data rates. Limited availability of new spectrum is a big hurdle in meeting growing demand for network capacity. Improved communication technology and use of multiple transmit and receive antennas (diversity, spatial multiplexing, beam forming) such as in LTE and LTE-Advanced technology and increasing use of Wi-Fi offload provides ways to increase network capacity.
Also, active antenna systems (AAS), an incarnation of RRH that has the same tower footprint as the existing antenna, supports multiple active antenna elements for beam forming to help improve coverage and capacity.
However, boosting network capacity with improved technology may not be sufficient. Use of cells with smaller radius serving smaller number of users appears to be necessary to scale network capacity, particularly in dense urban areas where capacity crunch is most severe. New sites need to be located to roll out new basestations on an on-going basis.
Basestation and RRH considerations
The acquisition of a new basestation site (a tower and adjacent space to place the base station cabinet) is becoming an expensive and lengthy process due to arduous complexities of attaining approvals from local/ municipal governments. A major part of the cost in installing a base station is in acquiring real estate and costs associated with the civil works. In a typical installation, electronic equipment constitutes less than 20% of the total cost of installing a base station. Besides scarcity of finding space for cell towers in congested urban/sub-urban areas, visual impact of a tower is making matters worse and causing increasing hurdles in getting city approvals.
The distributed base station architecture allows the location of basestation cabinets miles away from RRHs, thereby providing the opportunity to mount smaller RRH units on the side of buildings, electric poles, and so on.
Distributed basestation architecture deployment does have a shortcoming in scaling the front-haul network or connectivity between the basestations and remote radio heads.
Typically, optical fibre is laid out to connect RRH to basestation chassis. Technology changes need to happen to allow sharing of a common media by incorporating class of service prioritisation to maintain appropriate levels of deterministic latency and end-to-end delay management.
The rise of small cells
While distributed basestation architectures have been making a gradual progress in providing a viable solution for network capacity, small cells have arisen as an alternative approach.
No industry standard definition exists for small cells. In general terms, small cell is a complete basestation in a small and compact form factor enclosure that can be easily deployed on light poles and the side of the buildings.
The transmit power of a small cell is in the range of 250mW to 5W, with cell range varying from 50m to 5 km.
A small cell radius of 100’s of metres supporting 32/64+ users is likely to be most common in dense urban settings.
A small cell is low cost and is expected to be in the range of 1/10th of the cost of a typical macrocell.
A cluster of 5 to 10 or higher numbers of small cells are expected to exist within a macro cell as an underlay to enhance network capacity, primarily expected to cater to data services.
A small cell gateway is expected to control small cells to work cooperatively with neighbours and mother macro cell. Load balancing and inter cell interference management are expected to be crucial parameters controlled by the small cell gateway.
While small cells have a lot of benefits and are generating considerable interest within the industry, there are considerable challenges for wider deployments. Small cell backhaul is a major hurdle.
Based on the deployment scenarios, ability to choose a backhauling approach from among a toolkit of backhaul solutions is critical to ease deployment. This creates considerable challenges in integrating backhaul into small cell.
An approach using two boxes, one for small cell and another for backhaul, may not be viable in the long run. Small cell radio access network (RAN) creates an additional layer in the mobile backhaul hierarchy, further stressing already constrained mobile backhaul access network.
Another significant barrier for small cells is the need for an industry standard and a truly interoperable framework to ascertain coexistence of a small cell with neighbouring cells.
Small cells are expected to work as well behaved self-optimising network entities. Inter cell interference coordination and dynamic load balancing become issues if not properly managed and controlled, and this can lead to poor overall network performance.
Despite all the challenges, small cell technology holds tremendous promise in providing a good solution to manage perpetual growth in demand for network capacity.
Small cell deployment numbers and velocity of deployment, in summary, would depend on how quickly the industry comes up with ways to address small cell challenges and the extent of distributed basestation architecture adoption and penetration.
Remote radio heads and active antenna systems, being an extension of the existing macrocells, are equally attractive in many situations.
There is no doubt that both distributed basestation architectures and small cells will be deployed and coexist in increasing network capacity.
Harpinder Matharu is senior product manager at Xilinx
