Recently, we’ve noticed there is confusion regarding GSM cellular technology, and how it relates to CDMA. Not a big surprise given the wealth of jargon, buzzwords and semi-informed “analysts” in this area, but worth straightening out.
First generation: analog
In the beginning was analog cellular mobile radio. This was the first generation (1G) of cellular, an improvement on the previous Mobile Phone System (MPS) that provided better coverage, more capacity and allowed mobility: the possibility of the user moving, and being handed off from one base station [antenna] to another without dropping the call.
Various flavors were deployed by operators in different countries, including AMPS in North America, TACS in England and NMT in Scandanavian countries. These 1G systems are Frequency-Division Multiple Access (FDMA), meaning that the spectrum [radio band] allocated to the operator is divided into smaller bands called channels, and channels are allocated to users.
Second generation: digital and the warring factions
Problems with capacity and data communication led us to second generation (2G) cellular. Two warring factions emerged, with radically different views on sharing spectrum amongst users: the TDMA faction and the CDMA faction.
TDMA, Time-Division Multiple Access, means sharing one radio channel amongst a number of users by taking turns, one after another, in time. In North America, systems conforming to the IS-136 standard implement eight time slots on 30 kHz channels, allowing three users (one time slot for each direction for each user, plus time slots for control information). In the Rest Of The World, systems conforming to the Global System for Mobile Communications (GSM) standard implement sixteen time slots on 200 kHz channels, allowing seven users. Modems transmit 1s and 0s that are digitized speech (or digitized silence), or data (or idle patterns) at 9.6 kb/s between the phones and the base station over the radio channel.
CDMA, Code-Division Multiple Access, means not dividing spectrum into narrow channels, and not implementing time-sharing on those channels, but instead having all users transmit in the same carrier [wide frequency band], all at the same time. However, instead of transmitting 1s and 0s directly, the users transmit codes to represent 1s or 0s, and only transmit when they have something to say. Codes are binary numbers – strings of 1s and 0s – chosen so that if some users transmit and some do not, the base station can determine which transmitted. In North America and in limited places in the Rest of The World, a solution from Qualcomm Incorporated called CDMAOne and standardized as IS-95 was deployed. Qualcomm has patents on several functions necessary for cellular CDMA.
GSM/TDMA is the most popular today
GSM/TDMA became far more popular than IS-95, and so the market for selling phones and collecting money from GSM users is currently the largest. But, GSM/TDMA assigns 9.6 kb/s time slots to users, and even with band-aids and add-ons like GPRS and HSCSD, bandwidth like that is not useful for data communication (email, text messaging, Internet browsing, Google maps and location-based advertising, video) in any meaningful way. Further, GSM/TDMA reserves bandwidth for users whether they have anything to transmit or not – a very wasteful and inefficient use of scarce radio spectrum.
Third generation broadband: the factions continue to disagree
So we needed a third generation (3G). Because of the 2G schism, it was desired to have a global standard for 3G mobile radio. A group called International Mobile Telecommunications 2000 (IMT-2000) was formed to come up with a single world standard – and failed. They produced a document that had five incompatible variations. The two serious variations both were CDMA, since it is the most flexible and most efficient way of sharing the radio spectrum.
The warring factions did not make peace, they just changed what they were arguing about.
The 2G CDMA faction supported the variation called IMT-Multi Carrier, known as CDMA2000, basically a software upgrade from IS-95. A version called 1X using 1.25 MHz carriers was immediately deployed. 1X Evolution – Data Optimized (1XEV-DO) allows high-bitrate data communications.
The GSM/TDMA faction supported the variation called IMT-Direct Spread, known as Wideband CDMA and now marketed as Universal Mobile Telecommunications System (UMTS), which uses 5 MHz carriers and allows high-bitrate data communications using technologies like High-Speed Packet Access (HSPA).
The CDMA tipping point in the GSM/UMTS faction
After numerous false starts, the tipping point between 2G and 3G in the GSM/UMTS camp was finally reached in the summer of 2007, when more new activations on GSM/UMTS carriers’ networks were UMTS (3G CDMA) instead of GSM (2G TDMA).
What to take away from this discussion
Want more ?
There is, of course, more to the story than this brief tutorial. This topic is covered in more detail in Teracom instructor-led courses, DVD video Computer-Based Training courses, and textbooks:
Course 101 Telecom, Datacom and Networking for Non-Engineers: Chapter 6
Course 120 Understanding Wireless: Chapters 8, 9 and 10
DVD 6 Understanding Wireless 1: Parts 2, 3 and 4
Telecom 101 textbook, 3rd edition: Sections 8.2 – 8.7 (21 pages)
One thought on “GSM vs. UMTS… and the CDMA tipping point finally reached”