• Phones
• Phone Finder
• News
• Carriers
• Android
• iOS
• Glossary

• Advertising
• About
• Contact
• Privacy
• Terms of Use

The Problem

What Problem? 

What if 1998 and 1999 came and went without a single news report on Y2K? Doubtless, there would have been mumblings on the web or at parties from the millions of programmers around the world, but imagine the millennium without media induced hysteria about computer systems grinding to a halt.

It's not that hard to imagine, because something not so different is happening right now. Engineers and programmers at handset manufacturers, base station equipment companies, backend service providers and carriers are all scrambling to fix a problem akin to Y2K before their clock strikes the proverbial midnight.

Like the Y2K bug, this problem is due to the poor planning of some short-sighted engineers and the efforts of those who came after them to provide backwards compatibility. Like Y2K, unless this problem is fixed systems will freeze and chaos, however minor, could ensue. But also like Y2K, it is likely that all systems will be upgraded before time runs out, and most people will look back and say "what was all the fuss about?" Only this time there's no fuss.

ESN History 

In the mid-1980s when cellphones were still analog, required battery packs and looked like more like briefcases than phones, no one could imagine the day when more than 4 billion cell phones were manufactured. In order to identify phones on a network, engineers developed a 32-bit code called the Electronic Serial Number (ESN). Initially this code was used for billing and to make sure the right call went to the right phone. Although 32 bits only allowed 4 billion unique numbers, engineers probably didn't consider ESN a short-term solution, but certainly they believed that technology would change before ESNs were exhausted.

However, as technology advanced from analog to TDMA or CDMA and then CDMA2000, the networks continued to use ESNs to identify phones. It made sense to use them in order to maintain compatibility with older networks, since many carriers upgraded their systems piece by piece. We still have dual- or tri-mode phones today. Once they upgraded to digital networks, carriers also started using ESNs to secure phone calls and eventually prevent fraud.

Unfortunately, the rapid growth of TDMA and CDMA outside the US, as well as short handset replacement cycles have nearly exhausted the supply of ESNs. In addition, early on in cell phone history, large blocks of serial numbers were distributed to manufacturers rather liberally, speeding the depletion of a limited supply.

The fact that it only took 20 years to use 4 billion serial numbers is even more amazing when you consider that GSM phones don't use ESNs. Since GSM was launched as an all-digital system with no need for backward compatibility, they use a different numbering system called IMEI. Probably since GSM came along later, and could learn from the mistakes of ESN, IMEI is almost twice as long, providing a significantly higher limit of unique codes.

As recently as late 2004, it was estimated ESNs would run out by mid-2005. Thanks to the efforts of the organization in charge of dispensing ESNs (the TIA) to recover issued but unused codes, most manufacturers now won't run out of them until sometime in 2006 or early 2007. Before the supply of ESNs is exhausted, manufacturers and carriers will need to be ready for ESN's successor, Mobile Equipment ID (MEID). But they will be rushing to do this much sooner than they expected.

All About MEID 

MEID is a 56-bit serial code meant to replace ESNs for identifying phones on networks as well as securing calls and other ESN functions. By using a 56-bit code, MEID will have a much longer lifespan. The TIA will also maximize MEID's lifespan with stingier rules about assigning blocks of codes. As an added bonus, since it is the same length as IMEI, it will be possible to assign a single code to a multi-radio device.

MEID was originally intended to be implemented when carriers upgraded their networks to CDMA 2000 EV-DV, Revision D. At that time there were supposed to be enough remaining ESNs to assign to TDMA phones on the few remaining networks while all future CDMA handsets would receive MEIDs since the upgraded networks would use MEIDs to identify the phones.

Recently, EV-DO Revision A has replaced EV-DV on carrier roadmaps. Since MEID compatibility was slated for EV-DV rollouts, the industry was caught in a lurch. ESNs were running out and no one had an MEID-compatible network on their roadmap. The industry needed to come up with a stop gap solution, or two.

What's the big deal? 

If ESNs ran out and there was no standard to replace them, manufacturers would literally have to shut down production of CDMA phones. Each phone is required to have a unique serial number burned into its chips in order to prevent fraud. Without a unique number, the phone can't be sold.

If a replacement for ESN, like MEID, was developed for new phones, but carriers didn't have systems in place to support this, they wouldn't be able to sell any new phones. At the very least they would only be able to offer handsets from those manufacturers who could still provide phones with ESNs.

No one can afford to let situations like these occur. That's why manufacturers, carriers, and the cellular industry as a whole are working to make sure they can transition to MEID and provide constant flow of new handsets.

The Solution

Plugging The Dyke 

First, the industry requires a way to let MEID phones work on old networks that use ESN. This is critical for manufacturers that will run out of ESNs before networks are ready to use MEID. The phones will be programmed with an MEID which they will use to generate a temporary ESN called pseudoESN (pESN). The problem is that the pESN will not be unique. There is a chance that more than one phone with the same pseudoESN is on the same network. There would be no way for the network to tell them apart. If this happens the 2 phones would probably get each other's SMS messages, at the least, or cancel out each others' service preventing both phones from working, in a worst case scenario. Engineers are working to minimize these problems before pESNs handsets are rolled out.

To genuinely solve this problem, the industry needs to come up with a way to recognize MEIDs on current networks. This effort is known as "MEID on CDMA2000". It consists of two parts. The first is getting handset manufacturers to comply with a little trick. In addition to ESNs, every handset has an additional set of codes that tell the network what the phone is capable of. One of those codes has previously gone unused and has always been set to "off". Manufacturers would set this code to "on" for all handsets with an MEID.

Base stations would then need to be upgraded to query phones entering a cell for this code, which they never cared about before. If a phone responded with the code "on", then the cell would address all traffic to the phone using MEID, if it was still "off," the cell would continue to use an ESN. This would assure that every phone on a network would have a unique identifier, and thus avoid mixing up transmissions (which the network types call data collisions).

MEID-equipped handsets will need to be able to generate a pseudoESN as well as comply with MEID for CDMA2000 in order to be compatible with CDMA networks for the foreseeable future.

The Switch 

Although the TIA claims they have enough ESNs to last roughly until 2007, different manufacturers have varying supplies left. While Kyocera is confident it will not need to switch from ESN to MEID until 2007, Nokia is looking to switch by the start of 2006.

Whether handset manufacturers wait until 2007 to make the switch or do it earlier, the networks want to be ready for MEID-equipped handsets before ESNs run out. Sprint is already prepared to accommodate MEID phones with the use of pseudoESNs, and is waiting for the CDMA Development Group (CDG) to ratify MEID for CDMA2000 so they can upgrade their equipment "as soon as the CDG gives us something to hang our hat on." Verizon, meanwhile, has internal systems that still rely on ESNs. They are on an aggressive timeline to ensure their systems are MEID-compatible by the time their hardware partners need to switch.

According to Lucent, which manufacturers CDMA base station hardware and is helping to develop the MEID for CDMA2000 standard, once the standard is ratified customers will need to install a software upgrade on their base stations in order to support MEID. They expect this upgrade to be ready by early 2006.

Once this system is in place, the CDG can devote its efforts to changing its development plans and coming up with a new revision of CDMA. A new revision will need to support both MEID and EV-DO natively.

Don't Panic! 

One manufacturer described the hurried switch to MEID by saying "it's like falling out of a fifty story building; just because you haven't hit the ground by the time you reach the 25th floor doesn't mean that everything's okay." The good news is emergency crews are rushing to the scene, and it sounds like there will be a big airbag at the bottom.

There is an emergency response plan in place. Carriers have to free their back-end systems of reliance on ESNs, then decide with their handset partners if they will support pESN handsets or skip straight to MEID. Equipment manufacturers must ratify MEID for CDMA2000 and then test and release software upgrades for their base stations. Handset manufacturers need to work with carrier partners to make sure their carrier partners are ready for when they will run out of ESNs and switch to MEID.

As with the Y2K fixes, it's critical support for MEID is complete before time runs out. If everything goes smoothly, major CDMA carriers expect to be ready in mid to late 2006. This should give them a few months of cushion before the TIA predicts all manufacturers will be forced to switch to MEID in 2007.

If the switch to support MEID closely parallels Y2K, then the deadline will come and go and most people will be none the wiser. But it's worth knowing that this is happening because it's best to be prepared. We knew all about Y2K, didn't we? The coordinated efforts of every company and organization involved should once again pay off in smooth transition. However just like Y2K, there's no guarantee everything will work until the deadline comes and the switch is made.

view article organized across multiple pages

This forum is closed.

This forum is closed.

Log In  or  Register

Forgot password

By registering, you agree to our Terms of Use.
We value your privacy and will never share your email address. (full privacy policy)