Wimax: the next disruptive technology?
Fifteen years ago, the focus of the “true” service provider was on voice traffic and data offerings based on virtual circuits, implemented with a plethora of semi-compatible technologies slowly developed within the ITU organization: X.25, ISDN, Frame Relay and the all-encompassing ATM.
In the meantime, some relatively small companies (including Cisco, Wellfleet and 3Com) were producing so-called “routers” that supported two technologies nobody took seriously: Ethernet and IP.
Fast-forward to the present: everyone is using IP and Ethernet (now called Carrier Ethernet), and the technologies that service providers once so proudly offered to their customers are now considered legacy overhead that has to be transported across high-speed IP/MPLS networks.
A similar revolution might happen in the wireless space in the next few years. Traditional mobile operators are following the slow path from voice calls through GPRS and UMTS to LTE. On the other hand, a new entry in the market, WiMAX, might (if executed correctly) cause the same level of disruption that IP did a few years ago.
WiMAX has all the necessary credentials to repeat the IP/Ethernet Cinderella story: the base Layer-2 standard was developed inside IEEE; Intel (one of the original backers of Ethernet) is pushing WiMAX heavily because Intel has already developed the chipsets; WiMAX development is backed by companies that truly understand next-generation IP-based networks and the WiMAX Forum’s Network Working Group is leveraging existing IETF-developed technologies to their maximum extent.
Just to give you a few examples:
- The WiMAX modem subscriber station (SS) looks like an Ethernet bridge to customer-provided equipment (CPE).
- WiMAX supports multiple service flows between a subscriber station and the base station, providing various quality-of-service (QoS) classes. To simplify the QoS mappings, the service flows can be mapped to VLANs on the CPE-SS link.
- The backhaul in the Access Service Network (ASN) is implemented with Ethernet and IP.
- Base stations use IP to transport subscribers’ data to the ASN gateway.
- Similar to the ASN, the core Connectivity Service Network (CSN) is a pure-IP network.
- The ASN gateway uses the RADIUS protocol to authenticate clients and DHCP to give them IP addresses.
- Roaming is implemented with Mobile IP.
Based purely on its technology merits, WiMAX should be another success story… but, as we all know, the devil is in the (sales and marketing) details. It’s illusory to expect that the majority of traditional mobile operators will offer WiMAX service instead of waiting for LTE; they’ve invested too much in the existing 3G infrastructure to throw it away and replace it with a completely new concept. To succeed like IP did, WiMAX needs a critical mass of disruptive startup mobile operators, and it’s questionable whether we’ll ever see them. The explosive growth of IP was spurred significantly by the dot-com boom (its low price and excellent scalability didn’t hurt, either), and it’s questionable whether the same feat can be repeated in the current economic climate.
WiMAX is also facing a user base problem. It’s commonly used as an alternative access method in underdeveloped rural areas. While this is undoubtedly a huge market, it’s questionable whether these users can afford high-speed wireless service. Promoting WiMAX handsets makes less sense today; users are usually satisfied with the speeds provided by existing 3G networks (assuming that the network is performing well, of course). Once you’re able to view YouTube videos on your handset, you don’t need a much faster link… unless another high-bandwidth killer application is just around the corner.
The history of IT and data communications is full of unexpected disruptive newcomers. For example, 20 years ago almost nobody believed that IP would carry the majority of voice, video and data traffic worldwide, or that corporate servers would run Windows in a virtual machine on Intel-based hardware. Will WiMAX be able to uproot the traditional mobile networks that are slowly migrating toward LTE? Only time will tell.
WIMAX is just on the road to eventually having super high speed mobile broadband WITH COMPARABLE LATENCY to fixed line. Not 100% sure where it fits in the road.. I.e. a competitor of LTE or a step to LTE (I suspect the former) But it is beyond current HSDPA/HSUPA which you could call 3.5G perhaps.
Anyhow, the back-haul is becoming all IP already. Goes (very roughly)something like this.
Mobile-Air/Radio(WIMAX/HSDPA/EDGE/GPRS/HSUPA/LTE)--NodeB--(IP Backhaul)--RNC--(IP/SS7)---SGSN(Data)/MSS/MWG(Voice)--MPLS Backbone--Foreign networks (Like the internet)
The above does not include the evolution to R5/R6 IMS (converged fixed/mobile) since it hasn't exactly taken off.
The specifications evolve in releases from R99(3g)-R4(IP Voice Core network)-R5,R6 (IMS)-R7-R8 (LTE) etc .. They basically evolve different parts of the network. For example R4 is enhanced Voice core network over IP.
So apart from disrupting the mobile phones and potential new hardware/software in radio equipment there will be no disruption on the backbone networks (apart from upgrading to cope with increased capacity and general evolution of the Core voice and packet networks -- I guess this is a disruption but only in the same way fixed line networks have evolved from 7500 days to CRS. Same will be true of the mobile backbones now all on their way to CRS or equivlent ).
To carry the analogy further, it is the equivalent of turning you ADSL to FTTC (If I am not over egging it)
The actual protocols for handling both data and voice traffic over mobile networks are well defined in 3GPP specs (let me know if you want some ;-) )
If Wimax wins, most of the 3G infrastructure will go the same way as Frame Relay/ATM gear did.