Reality Check: Borrowing brilliance for the SIP journey

Editor’s Note: Welcome to our weekly Reality Check column. We’ve gathered a group of visionaries and veterans in the mobile industry to give their insights into the marketplace.
In my previous Reality Check columns I’ve written about the benefits of SIP and how they can transform mobile service provider networks and services. I’d like to end my series of articles with some advice for those embarking on the SIP journey.
SIP services can generate new revenues, and SIP architectures, including IMS, can dramatically lower costs and improve business models and competitive viability. However, this endeavor does have its challenges. These challenges relate to all the aspects of the evolution – across access, core and interconnect networks – including: IMS-based VoLTE, RCS and SIPo3G, fixed mobile convergence (FMC), VoIP interconnect and SIP session routing. The following aphorisms will provide wisdom for mobile service providers as they begin to navigate the evolutionary path to SIP-based services.
In IP, trust no one
IP networks open the door to amazing new services, but they also expose service providers to new threats. With voice and messaging migrating to SIP, service providers need to protect their service and application infrastructure against malicious attacks designed to cripple a network element by overloading it with service requests. In addition to purposeful attacks, non-malicious overloads (such as the flood of calls during “American Idol” voting) can also cause increases in call signaling rates that exceed what the service provider infrastructure can support, resulting in network conditions that are similar in effect to DoS attacks. Other threats include malware, eavesdropping, service fraud and identity theft. While the private, licensed spectrum access network for mobile service provider may give a sense of security, the IP services infrastructure can be vulnerable; using the Internet for FMC further heightens those risks.
Regardless of the type of attack, a successful attack can result in a variety of losses including SLA promises, customers, a reputation, and revenue. The solution to this daunting challenge includes network design that incorporate elements that can provide DoS attack prevention, signaling rate limiting, policing, and encryption.
IP addresses will always be a collection of heterogeneous schemes
In service delivery, SIP traverses multiple IP networks – multifaceted access technologies, Internet, backbones, data centers, transit carriers – and these networks make use of a wide range of IP addressing schemes and technologies: IPv4, IPv6, MPLS, overlapping private spaces, and NATs. Many service provider networks use private address spaces, so IP interconnection between service providers becomes impossible. While LTE and the EPC are IPv6, the core service networks of service providers will likely remain IPv4 for some time due to the cost and complexity migration. Mobile service providers will need to deliver services to a mix of IPv4 and IPv6 access networks including 3G, LTE, and Internet connected endpoints. SIP service networks must have a means of mediating these IP address incompatibilities at the edge, looking deep into the SIP messages to change all relevant IP addresses for compatibility and routing.
SIP is not the only protocol
While SIP is at the center of all next-generation services architectures including 3GPP IMS, ETSI TISPAN, and PacketCable, the reality is that we live in a multiprotocol signaling world. Converged service providers will need to deal with H.323 enterprise IP PBX environments or MGCP devices, and some transit and termination wholesale carriers still use of H.323. SIP also has many dialects, including the 3GPP IMS SIP profile, RFC 3261, SIP-I and SIP-T. Further, “standard SIP” offers multiple options for transport protocols – UDP, TCP and SCTP – multiple options for signaling security – TLS, IPSec and none – and multiple options for media security – SRTP, IPSec and none. There are also differences in how DMTF is handled and response codes may differ between networks. While all these different options for SIP provide tremendous choice and flexibility, they also guarantee incompatibility and a lack of interoperability between networks. Tools for SIP signaling normalization and interworking to H.323 environments are critical components to maximizing service and network reach and expediting time to market.
Codecs will never converge to a few
While the world is standardizing on SIP for new deployments, codecs will never converge to only a few. Within the fixed-line world, voice codecs comprise the numerous G.7xx series and video ITU H series codecs. In addition to codec variants, there are also options for frame sizes – 10, 20 or 30 milliseconds, for example, which add to the complexity. In mobile, an entirely different set of voice and video codecs are used that feature adaptive dynamic support for multiple bit rates to optimize bandwidth utilization over the radio access network. Lastly, new codecs are being developed to further improve quality while minimizing bandwidth utilization and wideband (HD voice) codecs are further expanding the codec menagerie. Every call between endpoints not supporting at least one common codec, especially for converged operators serving mobile and fixed line phones, will require transcoding.
The myth of unlimited bandwidth, QoS & signaling resources
LTE dramatically increases access bandwidth for subscribers. However, seemingly excessive speeds will be consumed, and an increase in bandwidth at one network point, moves the bandwidth bottleneck to another part of the network. Similarly, the servers delivering interactive communication services also have finite capacities for call handling – fat pipes do not alleviate signaling constraints. If a link or a server is at capacity, and more traffic – just one more call – is placed on the link, the quality of all active calls will deteriorate, not just that last call. The network must take into account both the bandwidth and signaling constraints for the admission of SIP services and be able to respond to quality degradations to assure service level agreements and high quality of experience.
Some sessions are more important than others
On the Internet all packets are equal and are delivered on a best-effort basis. Any packet has the same probability of getting dropped or delayed. In the voice world, mechanisms are needed to provide special handling for particular calls or sessions. In the presence of voice server overload, service providers need the ability to gracefully reject low value calls to support high-value services. Emergency calls also need special handling in terms of prioritization and possibly pre-emption if service resources are oversubscribed.
IP regulatory requirements will increase
Today’s regulations will need to exist in IP world. As traditional voice move to IP networks and new interactive communications arise, compliance with regulations that better society, such as lawful intercept and emergency calling, will be required.
Service provider business models will never be homogenous
With SIP and IMS, interactive communication services – including IM, presence, video and multimedia sessions – are potentially so rich that business models will never be homogeneous. How they will be charged and settled will be widely diverse. Sessions will be available with or without QoS support (or mid-session QoS could be added) and different revenue models could be available for each scenario. With IP interconnects, bill-and-keep business models are rising as they eliminate the cost and hassle of the administrative infras
tructure associated with per-call settlement or negotiated s
ettlements on termination costs. This contrasts with another model using negotiated settlement across the value chain featured in the GSMA IPX. Ultimately, service providers need flexible routing capabilities that take all these relationships and metrics into account. These diverse SIP services also demand accurate accounting for billing, traffic planning and settlement.
So as you can see, the SIP evolution does not change the broader technology landscape – it is still imperfect and presents challenges to SIP service delivery. Rest assured there are solutions out there. After all, this SIP evolution has been underway for a decade. My best advice to service providers just joining the evolution is to borrow brilliance from those that have gone before.
Kevin Mitchell is Director, Solutions Marketing at Acme Packet where he leads wireless solution marketing focusing on wireless access, interconnect and core session routing applications of Acme Packet’s Net-Net product family. In addition to developing marketing and event strategy, Mitchell serves as the chief spokesperson for Acme Packet’s solutions for femtocells, fixed mobile convergence, RCS, 4G voice, VoIP peering and core session routing. Mitchell joined Acme Packet in 2005 after 8 years with Infonetics Research where he was most recently a Principal Analyst, serving as an overall company director and lead analyst responsible for consulting, analysis, product development and overall strategic direction of multiple technology coverage areas.