The big worry, however, is the backbone network. Ishikawa notes that the increased traffic to the trunk network lines in the last one year due to increased use of smartphones itself is quite alarming—reiterating the need to strengthen the trunk networks to sustain the growth of the Internet.
Neeraj Arora, director-service provider vertical, Cisco IBSG, says, “Specifically from mobile network technology perspective, platforms today are often constrained. Challenges vary based on consumer- vs enterprise-specific service. While many consumer-specific mobile cloud apps require a high throughput, many enterprise M2M services (telematics, fixed telemetry, AMI, security, vending/POS) do not need high throughputs to function well. This is because they are only sending small amounts of data—often intermittently or even on an exception-only basis. However, they may have other requirements including optimised activation rates (ability of mobile packet gateways to scale out PDP context activations), stateful IP session failovers and IPv6 support, and higher QoS, which may pose challenges for mobile networks. To summarise, there is a need to optimise the mobile packet gateways for specific consumer or enterprise ubi-comp service requirement.”
A deeper look into the mobile communication challenges
There are many aspects that need to be looked into as we transition from the comparatively simple networks of today to a more complex ubi-comp world, network capacity being just one of them. Some other issues that need attention are the interfaces and protocol definitions for communication between significant number of heterogeneous elements and a fundamentally new network architecture requiring universal broadband access at the edge nodes.
“The existing mobile Internet net-work will fall short of ubi-comp requirements in the areas of local multipoint connectivity, type of traffic supported, bandwidth needed, fault management and recovery actions, and the spectrum required,” asserts Dr Suresh Borkar, a trend-watcher, consultant and communications expert who teaches at the Electrical and Computer Engineering Department of the Illinois Institute of Technology.
[stextbox id=”info” caption=”With so many chips, how to manage the e-waste?”]
If there is going to be a chip in every object, what will happen to the e-waste generated during the manufacture of these chips, and when these objects (and the chips in them) reach end of their life?
“Of course, there have been concerns. But when you consider that many of the IC chips are essentially silicon chemicals, you will realise ordinary stones and sand have silicon composition too,” quips Ishikawa.
So the issue is more about the materials used for packaging the ICs. While ceramic-based packing is quite clean, it is the resin mould based packaging that is a bit suspicious in this regard and needs to be tracked.
“But, in the case of very tiny chips used for IoT applications without much packaging (they are often sealed into sheets or some type of planar structure), the issue may not be that serious,” Ishikawa says. He also points out that the tiny chips attached to the objects might actually be beneficial with regards to e-waste—as the chip will help trace the larger object in which it is embedded and organise its proper disposal!
Tushar Abraham, a former CDAC-ian, ubiquitous computing enthusiast and specialist in ZigBee wireless communication, says, “The number of devices in the mobile, personal computing and appliances space is anyway seeing an explosion. Ubi-comp will only integrate these devices in a more meaningful manner. Either way, India is not currently prepared to deal with the huge amount of e-waste that is generated. An effective e-waste policy will have to be put in place to deal with the current and future usage patterns of its 1.2 billion population. I believe the Pollution Control Board is working on such a policy. However, we still do not have companies that have the technology to retrieve valuable metals and other materials from this e-waste. Therefore it is imperative that stringent penalties are imposed on polluters. Tax incentives and FDI must be allowed to bring in cutting-edge technology into the country to meet the ever-growing demand.”
[/stextbox]
Borkar, explains each challenge in detail: Grid architectural concepts will entail processing distributed through the constellation of locally-embedded processors that work in parallel, and dealing with issues of synchronisation and communication. Significant attention will be needed on coordinated multiprocessing—not only between tightly-coupled processors but also multi-computing for loosely-coupled entities. Common languages will have to be developed for cooperative computing, data sharing, coherence and synchronisation.
The capacity requirements of a network are inherently dependent on the type of traffic carried through the various segments from access to backbone. Ubiquitous computing paradigm will be completely different from the current pattern of the traffic based on the use of PC or wireless smartphone like environment.
