Editor’s Note: Welcome to our weekly Reader Forum section. In an attempt to broaden our interaction with our readers we have created this forum for those with something meaningful to say to the wireless industry. We want to keep this as open as possible, but we maintain some editorial control to keep it free of commercials or attacks. Please send along submissions for this section to our editors at: [email protected].
Cooperation with a competitor – even going so far as to share network resources – is not a scenario that typically would be used to describe the hotly competitive mobile network operator world. But in the face of increasing competition from over-the-top players and an ever-growing capacity crunch, the network sharing business model is growing in popularity among MNOs.
The pressure on MNOs to continuously increase network capacity is immense and unceasing. The plethora of new data services that mobile networks are expected to transport does not result in enough of a rise in revenues required for ample investment in the network expansion. Network sharing is one effective solution to address the capacity crunch.
The sharing of network infrastructure and equipment requires careful planning to establish a trustworthy arrangement between two or more MNOs. Insufficient attention to points of contention between the parties can lead to bad relations and even breakup. The secret behind good network-sharing relations between otherwise-competitors is simple: Keep it transparent, and keep it fair.
The network sharing practice can be particularly effective in the backhaul portion of mobile networks, especially as conventional macro cells evolve into super-sized macro sites that require exponentially more bandwidth for backhaul. It has become abundantly clear that in these new scenarios, a new breed of backhaul solutions with significant investment is required. Sharing is one of the favorite options for mobile operators to cost effectively manage the need to address these new capacity challenges.
Many of today’s mobile operators are successfully working together to run services such as synchronization, voice, mission-critical dispatch, software download, and video streaming over a single, shared, packet, microwave backhaul network. But these sharing solutions only work if the proper steps are taken to address important issues like scalability, trust provisioning and migration.
Consider this scenario: Two separate mobile network operators are each running aggregate capacity of around 50 megabits per second to their own 2G/3G HSPA sites. Each MNO is planning a national-scale LTE deployment to increase capacity of its site to 100 Mbps in the first phase, with a road map to reach 200 Mbps in the second phase soon thereafter. Looking further into the future of LTE-A and small cells, it’s realistic to imagine 400 Mbps to the tail. Once that happens, the consolidation of two MNO sites into a single shared site would require an aggregate capacity of 1 gigabit per second.
The tricky part is that operators need to plan for a future with so many uncertainties. Looking back not too long ago, with only voice planning, the transport capacity target range was several DS1/E1s with a relatively small variance between most sites. Today, this situation has changed dramatically. The dynamic behavior that network designers experience with 4G indicates a variance between sites of hundreds of megabits per second. Think about two or three different generations of base stations each with different clock domains and probably an LTE ENodeB with higher requirements to be shared, not to mention a TDD spectrum that also requires strict phase synchronization. The considerations are enormous.
When you calculate in all the additional variables – the super-sized macro cell, the number of transport services it needs to support, the expected densification of sites, the need for convergence and still more requirements – it is easy to understand that planners are facing order-of-magnitude types of variances in almost every network factor.
To address the trust factor, each operator involved in site and/or network sharing must commit to a certain level of network attentiveness marked by fairness, transparent management, accurate reporting, tracking of SLAs and policy enforcement. That is not always so straightforward for parties who have been competing with each other for the as many as 20 years. The only way to build and maintain the trust is to use the proper tools and mechanisms that afford to each participant complete transparency.
The ability to address migration is the third important component to an effective network sharing arrangement. When people consider the movement from separate to shared infrastructure between two MNOs, the first goal they have in mind is expanding their capacity for half the price: One site instead of two, one BTS instead of two, one leased line instead of two and so forth.
Upon deeper consideration comes the realization that since the site is serving more subscribers, it needs increased resiliency at the transport level. In addition, there is a need for better reporting tools. With these added costs, expected savings from merging sites and transport will translate to 20% to 30% improvement in total cost of ownership. It’s not a 50% reduction in costs, but it is still quite significant.
Service quality is also an important consideration in network sharing. In order for service level agreements between sharers to work, they need to be enforceable, measurable and visible at the service-granularity level by each operator. This must be applied both to the guaranteed and non-guaranteed traffic. The information is required for the traffic that passed and for the traffic that was dropped.
A critical element for achieving fairness is the shaping mechanism. Even though, in a sense, traffic policing can also do the job, it is known to have negative effects on TCP/IP-based traffic, the majority of the data carried over the mobile backhaul. Shaping makes sure that traffic is sent at the proper rate and that buffers are used to absorb bursts while maintaining fair use of the total available capacity.
In shared networks, traditional reporting and assurance tools, like quality of service, are insufficient for setting and tracking service level agreements between the operators and between the services of each operator. They are also incapable of properly enforcing the rules.
Traffic-engineering tools – like hierarchical QoS at every site – maximizes capacity cost-effectively and satisfies trust requirements. These elements hold the key to long-term, successful network-sharing arrangements.
Policy-based shaping and scheduling are key elements in an H-QoS implementation. For each service, operators need to be able to schedule packet delivery at the proper rate, absorb bursts and, when relevant, eliminate the TCP-window re-size that happens due to extensive packet loss.
Network sharing to address backhaul challenges is a complex task that requires mobile network operators to consider many variables and address careful planning issues that will help establish a trustworthy and mutually beneficial arrangement. If approached properly, cooperation with competitors in the form of backhaul network sharing can indeed be a cost-effective method for MNOs to continue to modernize and evolve their networks.
Ran Avital, VP of Strategic Marketing, Ceragon Networks, has more than 18 years of experience in bringing to market next generation backhaul telecommunications networks. Avital is an active member of a number of technology forums such as the MEF, NGMN and others.
