The Internet of Things (IoT) is no longer a buzzword; today the average American owns 4 or more internet-connected devices including smartphones, tablets, PC’s, and laptops, and with trendy wearables like Google Glass, iWatch, and more on the rise—we’ve really got connections. But without reliable connectivity, a smartphone wouldn’t be as “smart” and a smartwatch would be quite dumb. Imagine your iPhone without 4G LTE, your laptop without WiFi, and Beats By Dre headsets without Bluetooth—they’d be 80%+ useless.
With the new wave of IoT driving interaction over individuality, better networking, and more bandwidth are required more than ever, not just for one device but among multiple, heterogeneous devices. This is where connectivity gets ugly. It’s no longer about troubleshooting a single annoying “limited access” triangle on your laptop or the red ‘X’ on your smartphone; it’s about getting your increasing collection of gadgets to discover, connect to, and sync with each other seamlessly, socializing in an “always on” mode. That’s easier said than done.
Ideally, all devices should be able to communicate with each other under one talking umbrella. And why not? They’re all aiming at the internet or the cloud, after all. But the reality is that different devices operate on different frequencies with different bandwidth requirements. All these differences have led to diverging network infrastructures from cellular networks such as 2G/3G/4G LTE, WiFi, to short-range connections such as ZigBee and Bluetooth and more. Telling manufacturers to stick with one protocol is like telling them to create identical devices—not conducive to market diversity and not likely to happen soon.
Yet the incentive to develop better connectivity solutions is high, with IoT promising gold for the world:
- Cardiologists can monitor patients remotely via connected temperature and humidity sensors and heart monitoring tattoos.
- Department of Energy and public utility companies can utilize smart meters and environmental sensors to create smart grids, smart cities, and data-driven predictions to yield smart alternative energy solutions and resource conservation.
- Data from traffic infrastructures and connected vehicles will enable transportation departments to lessen traffic congestion and increase safety and passengers to monitor their homes while out and about.
- Average consumers can use a smart device to unlock their smart home’s front door and turn on its lights and AC (and maybe the coffee maker), all in one go.
But in order to make these concepts a reality, devices and machines need a conducive environment to talk seamlessly. This requires an anytime, anywhere under any domain connection with undetectable interference. To meet demands, one of two things can be done:
- Uprooting current networking infrastructures and creating brand new IoT-centralized solutions
- Applying better networking management to existing protocols
In terms of practicality and cost-effectiveness, the latter is a winner. Thus for engineers and service providers, facilitating the interaction-driven world zeroes in on “how to manage” rather than “what else to create.” Einstein agrees:
To raise new questions, new possibilities, to regard old problems from a new angle, requires creative imagination and marks real advance in science.
Here’s what will mark a real advancement in science for IoT - self-organizing networks.
Self-organizing networks (SON) apply optimization techniques to existing networking protocols to enhance connectivity and ensure interoperability across different domains. Enabling solutions like configuration customization, self-healing and learning functionalities, and network orchestration and monitoring can help operators micro- and macro-manage their networks at a top tier.
This is especially suitable for IoT as software implementation and over-the-network upgrades are much more practical than hardwired solutions. When coverage includes billions of devices in millions of locations, door-to-door troubleshooting is a no-go.
Turning away from that, University researchers present more practical ways of handling IoT through SON:
Prompt neighbor discovery
Prompt discovery of devices is needed for a seamless device-to-device or machine-to-device interaction. If you want to access your refrigerator content through your phone, it’s got to be able to sense it first. And you don’t want to see that swirling "loading" icon for too long, if at all.
Solid path establishment
Devices need to establish solid connections with each other to transmit data without loss. This bi-directional data transmission is what’ll make connected devices useful to you. If you’re away from home, you won’t know what’s for dinner until your fridge starts babbling to your phone (unless you’re cooking, in which case you may still want to know how many onions and tomatoes are left for portion planning).
SON brings in clustering mechanisms to allow heterogeneous devices of a mesh network to connect to centralized data collectors, ruling out domain-specific connectivity problems, improving network scalability, and solidifying connection paths.
Medium access control (MAC) to avoid data traffic jams
MAC helps to minimize signal collisions for high data traffic situations. Does it matter? Oh yeah. When you’re stuck in heavy traffic in your connected vehicle and are syncing classical music to the car’s infotainment, you probably don’t want frequency collisions from the vehicle tailgating you enjoying their flow of heavy-metal and rock-n-roll. To this end, SON’s solutions have consumers’ backs:
- Time Division Multiple Access (TDMA): allowing devices to access channels in dedicated time slots.
- Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA): allowing devices to connect based on prior channel detection to see if it’s available or not; this is best used with less time-demanding transmissions.
Automated recovery management
When all our appliances are communicating with all our personal belongings, it’s likely that most people will have developed an over-reliance on IoT communication. Can you leave your smartphone for a week? We thought not. And sometime in the future, many won’t be able to bear being disconnected for more than half a minute. Luckily, self-organizing networks introduce self-recovery capabilities for failed network connections and service disruptions:
- Cell Outage: detection of errors and automatically switching to new network parameters
- A wide array of software enabling recovery solutions like self-recovery of network elements (NE)
Battery-operated IoT devices require efficient power consumption. Tie this back to the question of how long you’ll be able to go without connections in the near future. SON introduces data forwarding algorithms to optimize energy usage, such as:
- Decentralizing algorithms to balance load and make transmission routing more efficient
- Self-Organized Things (SoT) frameworks to minimize human interaction, automating procedures, and improving performance for optimal energy usage
In many aspects, SON is key to meeting complex networking demands of interacting IoT. Plenty of big-name industries have realized and seized this next-generation solution, like Nokia, for example. Not only do they enjoy SON’s automatic configuration, optimization, and self-recovery properties in their own workflow, but they are also offering a white boxed iSON solution for outside vendors to indulge in as well. In a recent interview, Nokia’s Chief Marketing Officer Barry French explained why:
We really want to make technology invisible and seamless, working for you in the background. Networks are constantly evolving. These [self-organizing]networks are smart enough to put capacity where it’s needed when it needs to be there.
The invisibility factor and white box approach can transition us into a simpler and finer era of technology for sure, where people can interact with devices without specialized knowledge or even being aware of it. That’s real seamless, indeed.
Other leading SON developers include Cisco Systems, Amdocs, Ericsson, Huawei Technologies, Eden Rocks, and Cellwize. Considering self-organization can automate network performance by more than 50%, with market forecasts reaching $6,500 million within the decade, early adopters to these solutions will be making a smart choice in the right direction. The place we’re headed for is beautifully described by Peter Hartwell of HP:
With a trillon sensors embedded in the environment—all connected by computing systems, software and services—it will be possible to hear the heartbeat of the Earth, impacting human interaction with the globe as profoundly as the Internet has revolutionized communications.
Are you eagerly awaiting the day that the IoT really comes into its own? Do you think that we really will become as connected as predicted? your thoughts and comments below...