The discussion about how 4G is different or enabling compared to 3G should start with a few guideposts.
It's important to note that 4G is built upon advancements in both wireless and wired networks. The link technology of Wimax and LTE is different than 2G and3G because it uses MIMO-OFDMA rather than CDMA/W-CDMA. OFDMA (Orthogonal Frequency Domain Multiple Access) works in the frequency domain, which makes it more efficient and easier to take advantage of evolution of smart and distributed antenna technologies.
The most authoritative definition of real 4G comes from the ITU. Both 3GPP LTE-Advanced and IEEE 802.16m (Wimax2) were last month officially granted admittance by the ITU's Radiocomms sector into the IMT-Advanced family.
The standard sets out stringent requirements, covering spectral efficiency for various MIMO configurations and deployment scenarios. Also included are latency and jitter, channel size and aggregation, and other criteria that impact the delivery of a wide range of communications from low-duty cycle M2M monitoring to applications that support real-time video and voice. The standard pushed the envelope of performance with the result now showing up in commercial trials that demonstrate sub-50ms end-to-end latency and high bandwidths.
Wider focus
What is also important is that 4G network standards are being developed to incorporate "intelligent" or "smart networking" methodologies. In fact, the ITU describes IMT-Advanced in terms of ICT, which is more comprehensive than simply describing this as the 4th generation of wireless networks.
Where do the innovations, improvements in performance, and enabling of new types of services come from that can make 4G much different than 3G? Both 3G and 4G have much in common in terms of technologies, product designs and manufacturing methodologies, and in the evolution of commercial markets.
LTE refers more to the evolution of commercial markets than the strict evolution of wireless technologies. 3G devices and prior equipment can't be used on the same frequency bands as LTE networks and vice versa. Newer equipment is often based on SDR/SCR (software defined/configurable radio) platforms that can be software upgraded from 3G to LTE or Wimax (Wimax being unlikely).
Much of the innovation comes from the combining of benefits and new capabilities that stem from convergence of wireless broadband with wired networks and computing methods. This impacts both the highly visible consumer device level and the various levels of the network and computing environment.
I leave the discussion of device and applications innovations as this is more a step-wise improvement over 3-3.9G. The new networks will deliver much lower latency and jitter, resulting in better performance for video conferencing and other real-time streaming applications than 3G.
However, even that gets blurred because HSPA+ adopts, even at a high cost and as a dead-end strategy, many of the MIMO and self-forming network (smart network), technologies. Users will not see a startling difference in speed between networks. However, 4G provides another 10- to 20-year roadmap for improvements.
Major advances will be made over the next 20 years or so in smart distributed WBB networks (SDWN). What that represents is the evolution of smart storage, routing and computing both from central servers, mostly the model of today, to more dynamic distributed ICT topologies. The evolution pairs with that of systems on a chip, distributed processing, smart DRM and other advances.
What is different about 4G is that it is an ICT rather than strictly a wireless platform. There are distinctions that will help lead to greater use of microcell and multiple-node aggregate base stations that are more easily deployed and help deliver greater capabilities at lower cost per bit. Overall, that is compelling.
Robert Syputa is a senior analyst and advisor at Maravedis