Large IoT systems usually have more than one job and need to work with other systems to be effective. Simplifying all this is one of the main things enterprise IoT platforms are designed to do.
But it’s a moving target, so vendors need to keep adding new capabilities to their platforms. On Tuesday, Nokia announced updates to its Impact software platform to cover IoT applications including lighting, video analytics and parking management. There are also updates to accommodate new low-power networks.
Impact is one of several software platforms designed to make IoT into more than a bunch of disparate sensor networks or automation systems. One application, like smart street lighting, may make another one, like connected parking meters, more effective. Data is the key, and integration can make all data more valuable.
General Electric’s Predix is another take on this objective, as is PTC’s ThingWorx. One of Impact’s strengths is its device management system, based on technology Nokia already uses to manage more than 1.5 billion smartphones and other devices worldwide, said Frank Ploumen, chief technology officer of IoT platform and applications at Nokia. Those management tasks includes things like collecting diagnostic data and doing over-the-air software updates. But the Impact platform also includes security, networking and analytics.
Many carriers already use Impact for services they offer to enterprises. Nokia is now targeting other industries, including utilities, cars, health care and city infrastructure.
On Tuesday, it introduced pre-integrated applications for parking, street lighting, and automotive capabilities like predictive maintenance and geofencing.
The parking software will help cities roll out things like empty-space detection and time-sensitive variable pricing, Nokia says. The lighting component is designed to make street lights into platforms for many types of sensors and pay for it with savings from more efficient lighting. For cars, Impact can now help companies detect low batteries or speed-limit violations.
But those packages by themselves are less important than the ability to integrate them and do even more. For example, street lamps with cameras pointed at the curb can detect whether each parking space is occupied, and that’s typically cheaper and more reliable than gluing physical sensors to the pavement. The same camera can capture the movement of cars and pedestrians for a more complete picture of overall congestion.
The new video analytics features in Impact also play into this. Nokia used machine-learning algorithms developed by Bell Labs for a system that can glean the information an application needs without transmitting full-resolution video over a network or requiring humans to view it, Ploumen said.
Without even recognizing objects such as pedestrians or cars, the visual algorithms can detect movement in an area through pixel patterns. Over time, they can learn the typical patterns and detect movements that are out of the ordinary.
The algorithms can run on computing elements at the edge of the network, detect anomalies, and report on those, he said. The same technology could make connected security cameras more efficient.
Nokia is also setting up Impact to work with LoRa and NB-IoT, two network types that are emerging as ways to connect small, battery-powered devices. Both are alternatives to standard cellular, with low speeds but high efficiency and long range.
LoRa, an open-source protocol supported by the LoRa Alliance, is designed for use on unlicensed spectrum and lets enterprises set up their own networks. NB-IoT is a form of LTE that carriers are expected to use with licensed frequencies.