Today, most of the world still communicates wirelessly over 3G and 2G networks. That’s expected to change as more cellular traffic moves to 4G (for fourth generation) LTE (Long Term Evolution) networks. Meanwhile, the majority of cellular subscribers use pre-paid, rather than post-paid or subscription-based mobile services, and that number continues to rise.
These two realities — the necessary coexistence of older cellular technologies and the new 4G LTE, and the growth of pre-paid services — are headed for a collision. Operators can’t share information between 3G and 4G networks about pre-paid subscribers without considerable expense or pain, but a new protocol developed by a Tier 1 mobile operator and a provider of signaling technologies, promises to tackle some of the incompatibilities, ensuring interoperable, pre-paid subscriber roaming between 3G and 4G LTE networks, and easing the transition to 4G and beyond.
4G LTE is the next iteration of wireless broadband technology, and it’s capable of download speeds four to five times faster than the 3G networks, thanks to faster processing and increased capacity. It provides an enriched Quality of Experience (QoE) for users through higher peak data rates, lower latency and higher spectral efficiency. As such, it can support the needs of today’s bandwidth-hogging Internet apps. Developed by the Third Generation Partnership Project (3GPP), an industry trade group, LTE is a radical shift from its predecessors 2G and 3G, which use the legacy SS7 telephony signaling protocols.
The Diameter protocol has replaced RADIUS as the standard for all aspects of policy, charging and subscription management in the 4G Evolved Packet Core (EPC). Diameter is based on RADIUS but has been enhanced to support failover, transmission security, reliable transport, agent support, server initiated messages, capability negotiation, peer discovery and configuration and is expandable with command codes (CC) and attribute-value pairs (AVP).
The base Diameter protocols have been defined by the Internet Engineering Task Force (IETF) and these have been extended using additional CCs and AVPs by the 3GPP and 3GPP2 to support interfaces between the various 4G network elements and applications. LTE is Diameter-centric and its upper layers use TCP/IP. That means all traffic, whether it is data, voice, video or messaging, can be carried over an all-IP network.
The shift from 3G (and 2G, since in some places in the world that’s still offered) is already underway, but a complete transition is still years off. According to an ABI Research report issued in September 2013, just 2.9 percent of cellular subscriptions worldwide currently have access to 4G LTE. But the report shows that adoption will surge in the coming years and by the end of 2018, about two-thirds of the total cellular traffic being broadcast around the world will be 4G LTE. The ABI report notes that by 2016, 3G will become the minority carrier network.
Until 4G LTE is fully entrenched, all operators — whether they have only 3G, a mesh of 3G and 4G, or only 4G, will have to coexist in order to support all subscribers around the world. 4G LTE was designed to interwork with 3G and other networks, many of which have been there for a long time and which have their own unique technology foundations. But integrating the various technologies into the single LTE umbrella is a daunting task. One challenge is addressing the scenarios that roamers are more likely to encounter. Trouble will arise whenever a pre-paid service subscriber has to roam from one type of cellular network to another. That’s because the mechanisms that authenticate subscribers and track their data usage on the network are incompatible.
Keep in mind that while the uptick of 4G is ongoing, so is the uptick of pre-paid services around the world. Outside of the United States, about 90 percent of people use pre-paid cellular services. Much of the United States does use post-paid, subscription services, but that’s changing too. Pre-paid subscriptions grew to over 100 million by June 2012, increasing by 12 percent from the previous year while post-paid subscriptions remained flat. Some estimate the ratio of pre-paid to post-paid in the United States to be approaching 50/50.
For subscribers using post-paid, subscription-based mobile services, interworking functions between legacy networks and the new 4G LTE networks are less of an issue. For example, authentication protocols are all that’s needed when roaming is used and a pre-paid subscriber has to leave his or her 4G LTE-based home network and gets handed off to a 3G-based network. And postpaid data-usage charging works the same in LTE as it does in 3G, because usage is totaled at the end of the billing period. But that’s not the case for pre-paid services, where the usage paid for must be debited in real time across the home and roamed networks.
The CAMEL Application Part (CAP) is a signaling protocol within SS7 that enables prepaid services in 3G. It is not supported in LTE. And there is no standard yet that manages data usage charges while roaming between 4G LTE and legacy SS7 networks. It’s a complex problem; the transaction has to be captured and held, and it all has to happen in real time.
Some of the protocol translations between LTE and SS7 have been defined by standards bodies, such as the 3GPP TS 29.305, a specification that enables authentication interworking between SS7’s Mobile Application Part (MAP) and the Diameter protocol defined for LTE. There also are workarounds. For large carriers and operators who currently have a 3G infrastructure and are overlaying a 4G LTE network on top, some are simply duplicating the subscriber information. It’s a costly solution and opens the door for error. Because there are two sets of data, things can get out of synch and errors can be introduced. But for operators with only 4G, that’s not a viable solution.
LTE signaling technologies provider Diametriq plans to introduce a standard to the 3GPP that ensures pre-paid subscriber roaming between LTE and 3G networks. Developed in conjunction with a Tier 1 operator, the Interworking Function (IWF) protocol enables roaming and handles the translation between SS7’s CAMEL and LTE’s Diameter, which facilitates charging events processing in 4G networks. IWF supports a number of use cases including the standardized MAP to/from Diameter (e.g., S6a, S6d, S13, S13’ to GSM MAP and ANSI-41 MAP) and be dynamically customized using script or plug-ins.
The IWF comprises a layered architecture with the following structure:
Protocol – Every transaction in an interworking scenario involves two underlying protocols, for example LTE’s Diameter and SS7’s Transaction Capabilities Application Part (TCAP), that operate independent of each other and comply with their separate protocol specifications involving parameter and message encoding and decoding, message routing and protocol management.
Transport Handler – This abstracts the usage of the corresponding underlying transport from the higher layers. Thus, there is one Transport Handler relevant to each protocol. Diameter is supported over TCP/SCTP and TCAP is supported over SIGTRAN/SS7.
Transaction Manager – This is a software function that manages the interworking transactions/sessions. It uses a State Machine in determining how to deal with an incoming request or response, and whether protocol conversion should be invoked.
Protocol Translator – This function performs the parameter and message mapping from one protocol to the other.
It takes a deep knowledge of SS7, IP and LTE signaling to make this work. That’s because bringing two “dissimilar” technologies under one hood poses challenges with respect to the mapping of routing parameters, and information carried by each protocol payload.
The Transaction Manager is the control center of the IWF protocol and is invoked when a protocol message arrives. It determines the protocol of the received message and whether the message is a request or a response. It also determines if it is the first message of a transaction, known as the “protocol at request origination.” For example, when a 2G/3G subscriber roams into an LTE area and switches on his device, the ensuing authentication and mobility management procedures cause specific Diameter transactions to be initiated toward the home network registry and the protocol at request origination is Diameter.
Vice versa, if an LTE subscriber roams into a 2G/3G area and switches on his device, the ensuing authentication and mobility management procedures cause SS7 MAP transactions to be initiated, and in this case the protocol at request origination is MAP. The Transaction Manager in IWF also can determine if it is a TCAP protocol (which is very complex), the type of TCAP message that has arrived and the TCAP transaction the TCAP message belongs to. Based on the message received, the Transaction Manager determines the unique context of the message.
Once all determinations are made (the protocol, whether it is a request or response, etc.) the Transaction Manager computes the IWF event that has occurred on a transaction; depending on the event it establishes context and then maps that context to IDs. That information is passed along to the IWF State Machine, which maps a possible incoming event and current state of the IWF context to a succeeding state and corresponding actions. By the way, the State Machine as defined in the IWF protocol is designed to operate unaware of any underlying protocols so it can be extended to support additional interworking such as Diameter-RADIUS.
The Protocol Translator is the knowledge engine of the IWF, and is invoked when the Transaction Manager determines a protocol conversion is required. Implementing protocol translation between disparate protocols requires knowledge and experience in the encoding and decoding of parameters and an understanding of the correlation between corresponding parameters of the respective protocols. How to implement the mapping requires algorithmic and programmatic skills.
The shift from SS7-based 3G and earlier cellular networks to 4G LTE Diameter-centric, IP-based networks is radical, and won’t happen overnight. Each operator won’t be dependent on how fast they get their own network upgraded; they’ll be dependent on how fast the world gets upgraded. For at least a decade, operators large and small will have to ensure that their services co-exist in this heterogeneous environment. Because of that, interworking will become increasingly important. The new IWF solves at least one critical problem, and it will ensure pre-paid subscriber roaming between legacy 3G networks and the faster, higher capacity 4G LTE networks.