 |
|
|
Communications systems need numbers, like fish need water. When numbers are plentiful, the need can be ignored, and those that assign them can easily be undervalued. However, if the pool dries up, carriers can be left gasping and wishing they had paid attention earlier. Right now, the levels of most numbering resources that wireless carriers need are at dangerously low levels. At least wireless carriers will not be lonely as they flop around looking for a solution. Landline telecom carriers and internet providers have very similar problems!
The average wireless phone consumer is probably only vividly aware of their phone number (known as a mobile directory number or MDN). But the ESN (Electronic Serial Number), MIN (Mobile Identification Number) and IMSI (International Mobile Subscriber Identity) also play critical roles (or, in the case of IMSI, soon will). Without them, wireless phones cannot be uniquely identified, which would leave carriers unable to validate, provide customized service, or bill correctly.
Why do wireless phones need so many identifiers? The need for a phone number is obvious. The ESN is used to identify the phone rather than the current subscription, and was designed to prevent fraud (with limited success). The MIN is used to identify the mobile within the wireless network (i.e. in communications between the mobile and the base station, and within the wireless backbone network). In most systems today, the MIN is based on the MDN, but that causes major headaches with number portability, international roaming and will result in exhaustion of the resource if not rectified. The IMSI is the mobile identifier in GSM systems, and will be the successor to the MIN in TDMA and CDMA systems. It has the least problems of any numbering resource.
The potential exhaustion of North American phone numbers has been well publicized, with pagers and wireless phones often being blamed. Yet, the problem lies almost totally with the landline carriers (or their regulators). Landline carriers have traditionally allocated blocks of 10,000 numbers to each rate center, which is the basis for distance-sensitive billing. Many of these areas are quite small and have fewer than 10,000 people living within their boundaries. If the rate center concept is followed, a rate center with 5,000 people in it could consume 200,000 numbers if 20 carriers (including competitive LECs and wireless carriers) offered service and respected rate center boundaries. Wireless carriers have traditionally offered extended local calling areas, and consequently can use numbering resources more efficiently by allocating one block of 10,000 numbers to customers based in several different rate centers. The FCC has favored number pooling as a conservation measure, which will only work if all carriers conform to rate center boundaries. If wireless carriers do this, they would probably waste vastly more numbering resources than would be saved by other carriers (such as the incumbent LEC) that currently conform to rate center boundaries. The FCC appears to have resigned itself to an arranged marriage with number pooling, rather than get involved in a messy fight to obtain its preferred partner, rate center consolidation, from the other suitor, state regulators. Caught in a wild wedding shootout, wireless carriers cannot win. If they resist pooling, they will be accused of being uncooperative, but if they succumb to the pressure to participate in pooling arrangements, they will incur enormous costs and drain the North American numbering pool far more quickly.
Obtaining a unique MIN in a wireless phone has usually been pretty easy for North American carriers – use the MDN. This has several major drawbacks:
Number portability and international roaming are forcing the separation of the MDN and MIN in North America. Although this will be difficult, there will be long-term benefits. Other countries with TDMA, CDMA or AMPS systems have already had to address this issue by migrating to the IRM (International Roaming MIN). An IRM is a MIN that begins with the digit 0 or 1 and therefore cannot be a valid North American MDN-based MIN. This provides a unique identifier, but unfortunately this portion of the resource is quite small. While North American carriers wastefully allocate the 8 billion MIN space that they occupy, all other cellular carriers around the world are forced to share 2 billion numbers. Consequently, the IRM resource is more than half used after only a couple of years of formal administration by IFAST (International Forum on AMPS Standards Technology). If North American carriers fully separate MINs from MDNs, literally billions of unique MIN codes will become available.
The successor to the MIN is the IMSI. Since it was designed for a system that was designed for use in multiple countries (GSM), it included a country code, which means that every country has a unique set of identifiers with no ongoing coordination required. The 3-digit Mobile Country Code (MCC) is more than enough for the world, and each country can allocate up to 1,000 Mobile Network Codes (or more if, as with the United States, multiple MCCs are allocated). Since the IMSI is 15 digits long, this provides each carrier with one billion globally unique numbers. If it was not for the painful transition that will be required, wireless carriers would probably all be using IMSI today. The transition to IMSI has started, with the development of standards for TDMA and CDMA equipment and for the TIA/EIA-41 backbone network. Wide-scale implementations are expected this year.
The ESN is still an important identifier, even though its use as an anti-fraud tool is limited. While one phone may be associated with several phone numbers – MINs and IMSIs over its lifetime – the ESN is designed to be unchangeable. While most uses of the ESN simply treat it as a 32-bit number (allowing over 4 billion unique values), they are not allocated in this way. The ESN is divided into an 8-bit manufacturer code and a 24-bit serial number. Thus, there are only 256 unique manufacturer codes, of which more than half have been currently allocated, and the rest are expected to be allocated in the next 5 to 10 years if nothing is done. One solution is to simply make the manufacturer’s code longer, and allocate ESNs in blocks smaller than the 17 million that are given out to each manufacturer now. This would dramatically increase the number of manufacturer codes that are available. Alternatively, the concept of a manufacturer code could be dropped entirely, allowing serial numbers to be allocated in blocks of any size. More drastic solutions are to expand the ESN to a larger number (which has the nasty side effect of breaking just about all software used in the wireless industry) or to question whether ESNs have to continue to be unique at all. Even if an ESN was in use by 2 or 3 mobiles around the world, they would still have the same modest value for fighting fraud.
The real challenge in designing a numbering plan is being able to predict the future. The oldest numbering plan discussed here, the North American dialable numbering plan (NANP) was designed in the 1940s. It is hard to imagine that its designers could have foreseen the massive expansion in telecommunications systems over the following 60 years. The MIN and ESN were designed in the 1980s, in the infancy of cellular communications, and nobody predicted the huge numbers of mobiles that would be in use 20 years later, or even that virtually every country in the world would have at least one wireless phone system. IMSI is the newest identifier of all, and so far seems to be faring the best. But, 60 years from now when all toasters (and perhaps even some fish) have a wireless communications device built in, will my successor writing on this topic be complaining that a thousand trillion numbers just is nowhere near enough?
© – Copyright