Applications Need Devices, and Devices Need Deployments
IMS deployments have been announced, but the advanced services/applications are not widely available, and the terminals are not, as yet, fully aligned. In the meantime “naked SIP” applications delivered via softswitches on NGNs are generating revenues. The industry is divided on the right approach, but IMS is happening on a step-by-step basis. by Bob Emmerson
IMS is a complex, controversial topic. Complex from a tech- nology perspective and controversial because the industry as a whole–including handset makers, software developers, network operators, etc.–is expected to have revenues of around $1 trillion a year. And by 2011 around four billion people will be carrying a mobile phone.
Let’s start by considering the different aspects of the IMS proposition. On one level, it’s a technology designed to launch a panoply of new services/applications in a flexible way–primarily in a high quality-of-service (QoS) environment. To achieve this objective, you also need a service-de-livery framework.
On another level, it’s a flexible architecture that allows operators to reduce their operating costs and thereby compete with the new breed of IP-centric service providers. The architecture also allows different wireline and wireless networks to be logically unified, thereby enabling a common service set to be delivered to a wider market. This is particularly important in the United States, where consolidation has taken place. In this case, several disparate networks are involved.
Right now, IMS is moving out of the trial phase, and ven-dors/network operators are talking about commercial deployments. However, you can count the applications on the fingers of one hand, and as of yet there are no devices that are 100 percent IMS compatible. So are we really talking about extended trials?
I put that point to Todd Ashton, who is head of IMS Sales, Multimedia Solution at Ericsson ( www.ericsson.com). The company, which is headquartered in Stockholm, Sweden, is the
market leader. Ericsson has signed 39 contracts for commercial IMS launches and performed more than 80 trials. He started by defining a commercial IMS launch. “It includes a CSCF [call session control function] and HSS [home subscriber service] and will typically include one or more applications; for example, IP telephony, IP Centrex, push-to-talk, and presence.”
Then I asked about the devices, rather the lack of IMS-compliant phones, and here too it depends on the definition of compliance. “What we can do today is build the requisite functionality into programmable [open OS] smart phones… create a framework that mirrors the network.… As yet the mobile phones do not enable the full IMS vision–one exception being the devices that SoftBank specified in Japan…. You can regard today’s implementations as being extended trials, but they are commercial because operators are generating top-line revenue based on an IMS infrastructure.”
When one considers the breadth and depth of the full vision, it is clear that operators have to seed the market and implement IMS on a step-by-step basis. Moreover, the legacy network has to run in parallel in order to retain revenue streams until the switch can be thrown over.
These twin objectives can be realized because, as stated earlier, the IMS architecture is flexible, so key components such as the CSCF and HSS can be deployed at an early stage (see figure 1). The industry is therefore marketing this development as a commercial deployment because the glass is half full. Others would point to the limited number of applications; e.g., VCC, presence, IM, and video sharing, and say that it’s half empty since these applications and many more are already available in a regular softswitch environment. In addition, the operating costs of running a network based on softswitches is significantly lower than that based on IMS.
IMS vendors do not create IMS applications. They come from third-party software developers–companies such as Broadsoft, Sylantro, and Ubiquity. And they are designed to run in architectures other than IMS; e.g., all-IP networks
WWWWW.VWO.VNOMNAMGA.CGO.CMOM
References:
Archives