TDD for Femtos, FDD for Macros

Almost a year back I’d blogged on the opportunity that TD-LTE is creating for network operators to architect & deploy non-interfering, high capacity, heterogeneous mobile networks.

Since then, the idea has gained good momentum in the mobile industry. From Hi3G Access AB’s LTE rollout plans in Scandinavia to Polish operator Aero’s LTE plans, we are witnessing operators adopting a hybrid spectrum acquisition strategy (i.e., both FDD & TDD) and rolling out hybrid LTE networks. This is a clear departure from 3G where TD-SCDMA deployments were mostly limited to China. The recent Light Reading article on LTE’s Kissing Cousins is a good compilation of TD-LTE’s adoption outside China. TD-LTE is gaining momentum in Europe but, most importantly, it is creating an unprecedented opportunity for operators to roll out heterogeneous networks.

In my recent article from MWC 2011, ‘Mobile data redefines operator battleground’, I discussed the need for operators to abandon the old ‘coverage’ centric strategy and adopt a new ‘capacity’ centric strategy as the majority of traffic on their networks has changed from voice to data. Operators need to architect and roll out 4G LTE networks that are designed for capacity – delivering the highest real-world data rates to users (i.e., not theoretical peak rates) and significantly lowering the cost/bit.

Heterogeneous networks hold the promise and small cells – femto, metro, pico – are the essential tools to deliver on that promise. However, whenever talking about two-layer network deployment (macros for coverage, femtos for capacity) the issue of interference takes center stage. This question has been at the center of the Femto industry for years and despite many white papers, numerous algorithms, various rollouts, and a wide range of products in marketplace, the issue just doesn’t seem to die. And it will not, because quite simply the laws of physics dictate that whenever two transmitters send signals over the same spectrum, they are bound to interfere. Irrespective of how clever the algorithm, there is always that next use case that will break that algorithm. Operators can choose: either continue to drink the same “smart-RRM-algorithm” Kool-Aid that many in the Femto industry continue to serve, or else pursue a robust, simple, clean spectrum acquisition strategy that paves the way for non-interfering, high capacity, cost-effective, heterogeneous networks.

Operators can develop their spectrum acquisition strategies in line with the broader objective of deploying capacity-centric networks. All major operators around the world are already acquiring wide bands of FDD spectrum for their 4G LTE networks. Augmenting their paired wideband FDD spectrum with smaller unpaired TDD spectrum creates the perfect combination that enables mass rollout of non-interfering, high capacity, cost-effective, heterogeneous networks. While no spectrum is cheap, relatively speaking TDD is cheaper and easier to find. Simultaneously leveraging FDD & TDD creates an unprecedented opportunity even for those operators who missed the boat and could not get premium bands (like 700MHz in US) for their LTE networks. A cost-effective LTE rollout even at 2.6GHz can be realized by acquiring small amounts of TDD spectrum and rolling out a heterogeneous network.

“TDD for femtos, FDD for macros” takes the interference issue completely off the table. Allocating spectrum in proportion to cell capacity – FDD (larger, paired) for macros where user density and cell capacity requirements are much higher while TDD (smaller, unpaired) for small cells where user density is much lower – will deliver great QoE to end users. More so for dense metrocells – lamppost-like deployments – the same TDD spectrum can be reused for over-the-air backhaul which is another essential element of heterogeneous networks. And not to forget LTE-A Relays. With device chipset providers already developing multi-mode (TDD & FDD) chipsets, the device ecosystem is already falling in place to support these multi-mode, non-interfering, high capacity heterogeneous networks.

Within a year of my blog, progress has already been made on this front, and the evidence is clear – smaller operators already see the value and are pursuing TDD to achieve differentiation and scale for their networks. Larger Tier1 operators should take note of this and should better align their spectrum acquisition strategy with this network rollout strategy. The opportunity is simple: with small incremental investment in TDD spectrum acquisition, operators can truly deploy heterogeneous networks that:

  • Requires less Macrocells, which are very expensive
  • Have lots of Femtocells and Metrocells, which are cheap and can offload Macros
  • Aggressively lower the cost/bit for the entire network
  • Deliver truly real-world realizable high data rates to users on their networks (a better QoE)
  • Create differentiation, accelerate mobile data adoption – and network ROI!

Twitter icon