This presentation will show you how RAD’s solutions can make your cellular backhaul network as cost-effective and efficient as possible.

We’ll start with a look at the current status of the cellular market. From there, we’ll move on to common backhaul applications. We’ll finish with a wide overview of the new trends, solutions and technologies geared for backhaul.

Let’s begin with a broad look at today’s cellular market.

GSM has been in use for over 15 years. By conservative estimates, GSM and GPRS will retain their dominant role for at least the next 5 years. On the other hand, 3G networks are becoming a reality in Europe, Asia-Pacific and other areas.

CDMA is following a path similar to that of GSM, and is expected to retain its market share accordingly.

Here we have a breakdown of the global cellular end user market. We see that ¾ of the world’s 1.5 billion subscribers are served by GSM networks, while most of the remainder are served by CDMA networks.

In keeping with these statistics, the material we cover from this point on will hold for GSM and CDMA networks.

Here, we see that the main driver for new infrastructure investment is the provisioning of lucrative next-generation services. GSM is here to stay, but 3G is the major growth engine for infrastructure.

In this slide, we see the UMTS Forum’s data on the recent surge in implementation of 3G networks. A ramping up has taken place since 2003 in the western and eastern hemispheres, earning 3G services an increasingly important place in service offerings in key markets.

Let’s have a look now at the key trends in the cellular space, keeping in mind our emphasis on backhaul efficiency.

First, we’ll see what’s happening in the transport network. Cellular operators are shifting from voice- to data-centric services. The statistical nature of data traffic requires them to redefine their transport infrastructure accordingly. For example, they have to support Quality of Service, or QoS, for each service. Higher bandwidth requirements stem from the high speeds of data links and needs in the backhaul network.

Next, we note that operators are looking for ways to generate funds with which to deploy 3G networks – one way is to redefine and leverage their existing 2G and 2.5G GSM and GPRS networks.

As we saw earlier, 3G is now the main driver of infrastructure investment. Operators are exploring new transport technologies that will increase the efficiency of their 3G deployments. They tend to implement 3G networks slowly and in stages.

In this slide, we’ll go over major trends on the operations side.

With the continuity of GSM/GPRS/CDMA and the emergence of 3G, we’re starting to see co-location of equipment servicing each of the 3 generations – 2G, 2.5G, and 3G. This stems from the need to handle multiple types of traffic and multiple technologies, in order to provide a rich bundle of services.

We all know how competitive the telecom and data services market is, and how each operator must do the utmost to reduce OPEX and maintain quality. QoS and up-time are crucial – hence the need for powerful and efficient remote management of all network elements.

An increasingly popular way to leverage cellular infrastructure is the provisioning of alternative services, such as WiFi, to provide data services like Internet surfing over the existing cellular transmission network.

We’ll now shift our focus to our central theme – cellular backhauling.

Backhauling is the connection of remote nodes to the network core. In the case of cellular networks, this means the connection of the remote base stations to the base station controller, and from there to the core network. Some operators have their own backhaul infrastructure, while others lease infrastructure from incumbent or alternative carriers. As we see in this slide, backhaul accounts for a major portion of the operator’s operational and/or capital expenses. This is the reason operators are so interested in reducing outlay on backhaul – it accounts for many millions of dollars in expenditures per year.

From here on, we’re going to show how RAD can help operators significantly cut their backhaul costs while increasing efficiency.

We’ll start with four examples of backhaul solutions that RAD has provided first- and second-generation cellular providers over the years. These solutions have been implemented extensively around the world, and have proven to cut costs and increase efficiency.

The first example is TDM grooming. This building block of cellular networking involves utilizing bandwidth in the most efficient way possible. Base stations using E1/T1 loops that are not fully populated are in fact not making the most of their resources. By implementing RAD’s one-to-zero groomer at a given base station, these time slots are filled by traffic from other base stations. This reduces the number of E1/T1 links necessary to connect multiple base stations to the base station controller.

RAD was a pioneer in grooming products that are deployed in equipment cabinets at base stations. These systems are highly redundant, ensuring a high level of reliability.

Grooming frees up bandwidth, so the obvious question is what to do with it? There are several possibilities. Let’s assume a base station that does not utilize all of the bandwidth available in its E1/T1 connection; furthermore, there is additional equipment on-site that the operator would like to control remotely from the central site – for example, the microwave links that connect base stations to each other. The base station does not provide technicians with a link to the central servers, and there is a shortage of SCADA ports. In such a case, grooming frees up bandwidth on the E1/T1 connection for these uses, each of which would otherwise need a dedicated data link.

With RAD’s drop-and-insert access router, operators can fill the unpopulated time slots in the base station’s trunk with the data necessary for these uses, all the way to the radio link. It does this with PPP protocol.

Our drop-and-insert access router has been deployed successfully in many locations, and operators are now using it to generate additional revenue streams. For example, it enables Wi-fi hot spots for wireless Internet access near the base station.

Let’s look at another application – this time, transmission of compressed voice over TDM and IP transport networks. This solution is in the network core as opposed to the backhaul, but it is relevant here because of its benefits and status as one of RAD’s most popular offerings to cellular operators.

Compressed voice is transmitted from one base station to another, from a base station to a base station controller, and from there to a mobile switching center. Very little can be done to optimize the voice bandwidth. Signals at the mobile switching center, however, are uncompressed at 64 kbps. A simple solution can reduce the cost of transport, where dozens or hundreds of E1/T1 links connect with the mobile switching center in a mesh topology. By compressing the voice channels with RAD’s Vmux product, the operator can reduce the number of E1/T1 lines necessary. This usually saves millions of dollars per year in OPEX. With compression rates of 1:10, 1:12, or even higher, return on investment can be achieved in a matter of a few weeks or months.

This slide shows RAD’s solution for traffic from the base station to the base station controller, via the transport network.

The traffic in question here is already compressed. To make this backhaul application more cost-effective and efficient, RAD addresses the A-bis protocol via which traffic runs over the E1 links between the base station and the base station controller. A-bis has certain flaws – for example, it transmits idle channels and does not always provide silence suppression when not transmitting voice. RAD’s Vmux addresses these inefficiencies, optimizing backhaul links over various media and reducing associated expenses by up to 50% - and sometimes even more. This enables operators to use a single E1 line to connect between sites, instead of the 2 necessary without the Vmux.

In summary, we’ve just looked at 4 common applications in which RAD equipment is deployed and reducing backhaul costs up to 50% and even higher. First, we saw grooming. Second, we saw the integration of control and management data into the BTS traffic. Third, we saw compression of voice between MSCs. Finally, we saw the optimization of the A-bis links that connect between base stations and BSCs.

We’re now going to move on to the third and final part of our presentation - a wide overview of the latest trends, solutions, and technologies geared for backhaul.

We’ll focus on the leading technologies, topologies, and types of products that are being implemented by cellular operators around the world to reduce costs and increase efficiency in the backhaul network. We’ll keep in mind the migration path to future technologies and services.

Let’s start by defining the challenge.

On the left, we see the different types of radio access network interfaces, or RANs, that are in use. Included here are the various elements of the RAN – namely the base station, Node B, the BSC and the RSC. 2G and 2.5G radio access networks are based on TDM interfaces. Emerging 3G RANs utilize ATM according to 3GPP Release 99 protocol. We have also seen the introduction of Ethernet and IP RAN platforms for 3G and future 4G.

To the right, we see the backhaul network transmission technologies. PDH and SDH have been popular since the early days of cellular provision because of their natural compatibility with TDM, but migration to ATM and packet switched networks is becoming more common.

As the arrows illustrate, the easiest way to connect a particular type or generation of radio access network to its backhaul is via a dedicated transport network of the same type. This approach, although logical, is inefficient.

As the second set of arrows shows, there is a need to connect different generations of radio access networks to different generations of backhaul infrastructure. 3G ATM RANs should be able to connect to PDH and SDH backhaul, as should packet switched RANs. Ideally, you should be able to use one infrastructure to connect all technologies.

As the third set of arrows shows, RAD offers solutions that enable you to run any generation RAN with any generation backhaul network. We can run various combinations using inverse multiplexing, circuit emulation and other methods. Running TDM or ATM over packet networks presents some difficulties, but we’ll see RAD’s solution for this later in the presentation.

Obviously, running different generations of radio access networks with different generations of backhaul infrastructure is a complex issue. Let’s focus on each transmission scenario one at a time.

We’ll start with PDH and SDH, as PDH is the oldest technology in use and SDH is deployed widely around the world.

RAD’s solution for PDH/SDH cellular backhauling centers on aggregation with our Optimux optical TDM multiplexer. Aggregation, or multiplexing, has been common in cellular transmission for many years, especially where E1 ports operate with high density. Normally, aggregation takes place at or near the base station controllers or mobile switching centers. More bandwidth is required in the transport network for third-generation services, so aggregation is now carried out at remote sites as well – often between the base station and base station controller, and frequently near the radio site.

You can use the Optimux to unify all TDM and ATM traffic into an STM-1 stream, which is then fed to the SDH network for transport. Feeding the SDH add-drop multiplexers with higher bitrate streams is more cost-effective than using lower bitrate ports. This frees up space in the ADMs for more high-speed ports, which allows for growth in your subscriber base and usage without adding more ADMs.

As for OPEX, this solution requires less management outlay. It also provides the savings associated with the extension of SDH or end-to-end QoS and high reliability into the backhaul network.

There is a drawback to this solution – it transparently aggregates any ATM traffic into the SDH network. This means running IMA traffic into VC-12 containers. This is not an issue in landline networks, but it creates an obstacle in the cellular network – an STM-1 stream full of VC-12 containers with IMA traffic. Third-generation radio network controllers can accept standard STM-1 streams or E1 streams with UNI protocol, but not VC-12s.

We do, however, have a way to work around this problem. Let’s see how.

RAD’s ACE SDH-ATM converter converts this channelized SDH traffic back into ATM format. The same device terminates all IMA sessions. The ACE then passes the converted signal to the RNC in a format it can read, thereby solving the VC-12 problem.

This solution enables you to make much better use of SDH infrastructure, eliminating the need to utilize multiple E1 ports at the RNC, which is expensive.

More and more operators are looking for solutions to transmit Ethernet traffic over SDH. This scenario is mainly applicable to CDMA networks.

CDMA traffic generally originates from a base station utilizing an Ethernet connection. Here, we need to convert between the Ethernet and TDM worlds. In order to leverage the SDH/SONET infrastructure, the operator needs to implement converters like RAD’s RICi or Egate products, which provide Ethernet conversion, aggregation and inverse multiplexing.

In the red circle, we see how the RICi-E1/E3 converts Ethernet traffic to E1/E3 format before passing it to the SDH ADM. Below this, we see how in applications where different granularities are necessary, the RICi-8 performs inverse multiplexing on the Ethernet traffic and then passes several E1 channels to the ADM.

On the right side of the diagram, we see how the Egate-100 converts the channelized STM-1 stream back into Gigabit Ethernet format and passes it to the controller.

We’ve now completed the scenarios for PDH/SDH cellular backhauling. Let’s move on to solutions for ATM backhaul networks.

ATM transport for backhaul traffic has several advantages.

The first stages of UMTS networking are defined as ATM-based. As such, ATM is the ideal backhaul solution for UMTS. TDM can be transported across the network by circuit emulation. ATM is also very flexible vis-à-vis future migration – it can run IP traffic very efficiently.

Just as RAD offers products that aggregate Ethernet traffic over SDH, our ACE multiservice aggregator does the same for ATM networks. This benefits the owner of the infrastructure – whether the cellular operator or the service provider leasing infrastructure to the operator.

In today’s implementations, aggregation takes place in any of several places in the network before the traffic reaches the central site. The ACE aggregates multiple types of traffic, such as TDM, ATM and Ethernet in order to leverage today’s infrastructure for future expansion and migration.

As for CAPEX benefits, as we saw in the SDH application, aggregation reduces the number of low-speed ports and cards in the ATM switches – this lowers costs accordingly. Outlay for 16 E1 links is equivalent to that necessary for an STM-1 link, so the migration path for future expansion is clear.

OPEX is reduced as well – since ATM is based on statistical multiplexing, the bandwidth required for transporting a variety of services is reduced. Prime examples of these services are delay-sensitive 3G traffic like voice and video streaming, as well as Internet surfing. Some of our customers have implemented a design known as “overbooking”, in which bandwidth is not allocated based on the expected traffic at peak times – rather, a statistical model is used. This requires statistical elements in the network, such as RAD’s ACE multiservice aggregator, as we see in the diagram.

Implementing compact devices also invoke additional savings on operating expenses that come from end-to-end monitoring and QoS. Deployed throughout the network, the ACE provides these all the way to the last mile. This lets the operator know exactly what is happening in the network at any given moment.

In accordance with density, geographical issues, and other factors, ATM aggregation access solutions are typically deployed in ring topologies in central business districts, while daisy chains are used in outlying areas

In daisy chain deployments, most base stations are linked by fiber – but there is still a need to cascade the traffic and link each base station to all of the others. At a given base station, the ACE drops all local traffic and connects to the media that links to the other bases stations, up to the base station controller. It aggregates the high-density STM-1 traffic, E1 lines and Ethernet. As we noted before, this allows the operator to offer a wide portfolio of voice and data services – such as WiFi – in remote locations.

Let’s have a look now at DSL backhaul in the ATM-based cellular network.

DSL is commonly used to provide voice and Internet access over the last mile. Some cellular providers are now using DSL as a low-cost alternative solution for cellular backhaul.

In this slide, we see a way to leverage last mile copper for cellular backhaul. This is applicable where there is space available in the DSLAM.

RAD’s TDM Integrated Access Device moves the TDM traffic from the 2G cellular base station to the copper SHDSL line, which then connects to the DSLAM. Alternatively, the Integrated Access Device does the same with the ATM traffic from a remote 3G node. The DSLAM then aggregates the traffic for transport over the ATM network, ending at the 2G base station controller or 3G radio network controller.

The Integrated Access Device thus lets you use existing infrastructure to access radio sites, whether your cellular network is GSM/GPRS or UMTS based. Once again, the solution supports end-to-end monitoring and quality assurance.

In the context of a third-generation network, which naturally requires more bandwidth, RAD’s DSL Bonding product performs a similar role. Through a protocol called DSL bonding, this device transmits the traffic over multiple SHDSL copper links to the DSLAM. This meets the need for increased bandwidth where a single E1 link in the SHDSL network is not sufficient.

As we see in the case of Node B, RAD’s DSL Bonding device can support several E1 links with IMA traffic on the remote side.

So far, we’ve covered TDM and ATM networks backhauled with PDH, SDH, ATM and ATM with SHDSL. We’ll now have a look at solutions for packet switched backhaul networks.

The general trend is convergence of packet technology and next-generation cellular networks.

Packet switched networks have already been deployed around the world, with many more still in pre-deployment stages. Manufacturers are pushing the technology as the next-generation transport method. Many incumbent operators have deployed packet networks for metro Ethernet applications.

In addition, the leading standards bodies have already defined their next generation standards to run cellular traffic over packet networks. These are based on convergence between technology and future development of the cellular network.

Our second bullet point shows the major scenarios for cellular backhauling. Many of these have already been implemented, and we are likely to see the others deployed soon. The most popular is the creation of IP backbones within the core network, using voice soft switches. These operators would like to leverage the infrastructure to provide access to base stations over the packet backbone. We also expect to see alternative carriers providing access services over their existing Ethernet metro networks. Satellite connectivity is a good solution for remote areas with no access to other transport media – this is most common today in Asia. This type of solution is most effective with Ethernet-based modems and service. Finally, we see more and more radio solution vendors providing Ethernet or IP-based radio links, replacing TDM radio. This is a more cost-effective solution for point-to-point and point-to-multipoint transmission.

Handling legacy equipment is of high importance. This refers to traffic in the GSM environment, as well as traffic in the ATM environment – which will eventually become a legacy protocol run over any type of packet switched network.

Let’s focus up close on this issue.

On the left, we see an illustration of the “classic” approach to handling legacy environments. This has been commonly implemented for years in order to provide a variety of IP services – from Internet surfing to more sophisticated services. This solution consists of running Ethernet or IP over first or second layer transmission technologies, or even directly over the transmission media itself – for example, Ethernet or IP over fiber.

On the right, we see RAD’s approach to legacy networks, which is the opposite of the classic approach. It involves running the TDM or ATM traffic over one or two packet switched network layers.

There are significant benefits to this reverse approach. In this slide, we’ll see how this approach is applied in a typical 2G or 2.5G network.

RAD’s approach to legacy networks is based on an area in which RAD is a pioneer and has supplied tens of thousands of links in the world – TDMoIP. RAD developed its IPmux to create virtual circuits that carry TDM transmission across the packet network. This enables unmodified legacy services to be carried across the packet network, serving as an alternative to VoIP. This solution has many inherent advantages.

The TDMoIP protocol encapsulates TDM traffic, gives the capsules headers, and runs them transparently over the Ethernet or IP transport network. It serves as a tunnel over the packet network, connecting remote TDM sites. In cellular networks, the IPmux gateway draws the E1 trunk across the entire transmission network to the destination site, where another IPmux serves as a termination unit, reconverting the traffic back to its native format.

All signaling remains intact, as the tunneling is completely transparent. This is very different from VoIP, where gateways handle the traffic and distort the signaling.

We’ll now move to a solution for 3G ATM backhaul over a packet switched network.

In this diagram, we see a TDMoIP solution in place in a current-generation network. Let’s see how our solution can migrate with the network to third-generation technology and services.

As in the 2G and 2.5G application, our 3G solution is based on encapsulation and pseudowires. The same tunneling over the IP or MPLS network is used, only this time using the standards currently under development by the ITU and IETF. Here, it is the ATM traffic that is encapsulated and transmitted by pseudowire across the packet network. On the far end, the RAD termination unit reconverts the traffic back to its native TDM or ATM format.

As we all know, clock synchronization is critical in TDM transmission. RAD’s TDMoIP solution maintains the clock throughout the process, keeping timing transparent throughout all radio sites, base stations, and base station controllers.

There are 3 ways to implement this. First, you can install a GPS-based reference clock at each aggregation node. Those who do not want to rely on a satellite can use a dedicated TDM link for clock distribution. This requires a separate link through the TDM network from all sites, all the way to the aggregation node. This is reliable but costly. The third possibility is to distribute the clock directly through the packet network. This is a complex technical challenge that RAD has solved with a transparent solution that integrates with the traffic itself. RAD has successfully implemented this solution at customer sites, including clock recovery according to the stringent GSM and UMTS requirements.

In this presentation, we’ve seen that RAD can provide you with the solution you need to backhaul traffic over your cellular network with significantly reduced costs and increased efficiency. We have pioneering, field-proven products that fit any combination of radio access and backhaul transmission technologies. Our solutions are applicable for 2G and 2.5G networks, they provide a sound migration path to 3G and 4G networks, and they leave no technological obstacle unsolved.

A tailored RAD cellular backhaul solution is a wise investment that usually achieves return on investment, or ROI, in a very short time. Our products help you protect your current infrastructure investment, while providing a proven migration path. For example, RAD equipment supports the emerging HSDPA in UMTS networks.

Based on our technological innovation, extensive experience, and large installed base, we are confident in stating that we are your best partner for cellular backhauling.

Here is a quick-reference guide showing how RAD’s solutions are matched to backhaul applications in nearly any combination of infrastructures and technologies – whether 2G, 2.5G, or 3G.

For further information, you are welcome to consult with your local RAD representative.