Microwave Journal
www.microwavejournal.com/articles/5743-the-impact-of-mobile-multi-hop-relay-on-wimax-above-and-beyond

The Impact of Mobile Multi-Hop Relay on WiMAX: Above and Beyond

December 13, 2007

This article is in response to a recent question I received regarding 802.16j Mobile Multi-hop Relay (MMR) and 802.16m IMT-Advanced standards and commercial developments. I have discussed over the past few years the development of a “middle tier” of wireless network requirements. More recently I have renamed this SDWBBN, for “smart distributed WBB network.” The recent draft for IMT-Advanced is expected to serve from low-cost, low-bandwidth applications to high-bandwidth and high-value specialized applications over the next couple of years. This area has been rapidly developing within standards efforts and individual companies, as reflected by patent applications and submissions to standards groups.

But there is much more to MMR than just extending network coverage similar to Wi-Fi MESH or 3G’s use of femto cells. The basic construct of 802.16e/m OFDMA combined with adaptive modulation, adaptive power control, adaptive HARQ methods, adaptive sub-channelization methods, and an adaptive architectural approach for how functions are scaled and mapped to hardware and applications results in a more “adaptive framework” approach to wireless. The diversity of spectrum and scale-of-use demands as exemplified in requirements for IMT-Advanced are adding to the push to adopt WiMAX’s framework approach to standards development.

It’s outside the scope of this article to discuss ground-up technical developments of MMR. Suffice it to say that the major reason LTE is forced to shift to using OFDMA is that this provides the framework to deliver the adaptive multi-service environment needed over the next 15-20 years. Most of the advances in network performance will be delivered through harnessing the core wireless framework to develop high reuse factors and more adaptive use of spectrum that can be finely tuned to instantaneous user and network demands.

Regarding “self backhaul” and providing scalable backhaul, 802.16e—when combined with higher-level MIMO-AAS and communications processor harnessing of 802.16j functionality—has the foundation capability to deliver a distributed network with highly-adaptive self back haul. WiMAXe/.16e has built-in capabilities to monitor and estimate sub-channel signal characteristics and to dynamically adjust them according to user requirements for QoS, user requirements for bandwidth, and broader network demands. Self back haul can be implemented such that an adaptive number/bandwidth of sub-channels can be allocated as needed. This can also be self-configured and adaptive to meet changing needs or account for changes in ability to access other segments of wired/wireless backhaul.

For example, if an MMR node drops out because of power drop or some other reason or if the wired back haul is reduced or goes out to a remote station hop node, the remaining nodes can adjust allocation of their back haul between stations to other segments of back haul. This can be accomplished by adjusting both the route traffic being hopped through the network and how much spectrum is allocated via sub-channel banding for back haul. That already goes beyond the capabilities of Wi-Fi MESH in three substantial ways: (a) It is all done within the framework of the standard rather than through the use of proprietary MAC or algorithms. (b) It is able to use separate modulation methods for back haul, including higher order modulation and MIMO (4X4 will be available sometime next year ahead of being codified into a standard requirement). This equates to much higher localized bandwidth for local area networking and back haul or PtP long range connections. (c) The multi-hop capability has only a relatively minor impact on communications latency and jitter performance factors. This contrasts with Wi-Fi MESH, which results in about a 50% bandwidth hit with each node hop and increased latency due to having to share access. This is because it uses a contention-based MAC, limited channelization, and adaptive mechanisms. Proprietary MAC can counter some of this network degradation, but it is not built into the standards or the basic MAC protocols.

The basic idea of combined .16e, .16j, .16m, and developments within WiMAX Forum task groups including the network management group is to provide a flexible, adaptive framework upon which systems can evolve. Although codifying much of anything into a standards document or set of certification requirements requires constraining aspects of the designs, the overall framework is intended to allow evolution. For the overall development that includes 802.16j, this means a framework for evolution of the SDWN—Smart Distributed Wireless (BB) Network. In light of its scalability to all classes of service and networks (as pointed out in recent ITU documents), it may be appropriate to reduce the term SDWBBN to SDWN. It is quite conceivable that even very low cost user devices will incorporate smart networking capability. Some researchers have postulated that smart BB networks would include user devices and that shared processor, programs, and content under DRM offers significant advantage.

Development of SDWN capabilities will have dramatic impacts on performance and sales of wireless networks. This harnessing of spatial domain via SDWN is one of the most important areas of wireless development for the next several years.