Creating smart cities with connected and smart devices

Semtech Corp. recently announced a collaboration with SAS, a leader in data analytics, and the Town of Cary, NC, to help improve the town’s quality of life, which included improving flood predictions, minimizing its environmental footprint, and reducing waste. The goal of the collaboration is to develop, deploy, and use real-time insights from data and analytics generated across multiple use cases to solve specific challenges.

Together, Semtech’s LoRa technology and SAS’s real-time sensing capabilities and advanced analytics are helping cities and organizations better predict and prepare for challenges that range from natural disasters to sustainability. Semtech’s LoRa technology provides the communication layer for the LoRaWAN low-power wide-area networking protocol, which is maintained by the LoRa Alliance.

The collaboration with the Town of Cary is yielding a unique Center of Excellence to solve real-world problems through the deployment of edge-to-cloud internet-of-things (IoT) solutions, leveraging Semtech’s expertise in LoRaWAN connectivity and SAS’s data analytics built ON Microsoft Azure. This is expected to be an example of how cities and towns can leverage IoT data and analytics for a smarter and safer planet, said Semtech.

Prior to the partnership, the Town of Cary had already started adopting IoT solutions for a variety of services, which included working with SAS on flood prediction. However, Cary faced hundreds of siloed solutions that needed to be streamlined and more responsive to the town’s needs, which meant building a scalable digital infrastructure.

The team’s answer to the challenge is the implementation of end-to-end solutions using LoRaWAN connectivity with SAS’s IoT analytics platform. A LoRaWAN infrastructure is made up of sensors, gateways, and base stations. The LoRaWAN protocol uses LoRa technology as the physical layer that launches data packets from the transmitting antenna to the receiving antenna.

LoRa and LoRaWAN have been designed to work together and marry long-range transmission with the lowest possible energy consumption, said Richard Lansdowne, senior director for business development, Semtech.

This technology is perfect for small packets of data such as sensor readings and the lower radio signal is incredibly robust to interference, he said. In addition, LoRa and LoRaWAN-enabled devices stay connected on battery power for many years using far less power than Wi-Fi or Bluetooth, for example, he added.

“The type of data captured includes water levels, rainfall levels, temperatures, pressures, and all of those things, and at this point, it’s data,” said Lansdowne, “but when the analytics is added that SAS brings it becomes insights and predictions that you can base decisions on or even decisions that are made automatically.”

Creating smart cities with connected and smart devices

LoRa and LoRaWAN applications. Click for a larger image. (Source: Semtech)

The challenge

The Town of Cary had invested in a lot of different projects and ended up with hundreds of siloed solutions with a gazillion logins, apps, and user management access, said Lansdowne. The town’s vision is to streamline all of these solutions and to use its rainfall and river data to predict floods ahead of time.

Here is where the town entered – what Lansdowne calls – the maze of complexity, where hours and weeks and maybe months are spent looking at the array of sensors and solutions available. This is typically followed by pilot purgatory, he explained, where you start working on different pilot projects such as smart parking or smart streetlights or waste management, ending up with siloed solutions each with its own connectivity – login, application, dashboard, and sensors – and analytical solution with different partners.

“This was not good for Cary and it’s not good for any city,” said Lansdowne. “SAS is [headquartered] in Cary, so they asked SAS for help to start building analytical models that can pull some of these data points together.”

In turn, SAS said its solutions needed to be able to predict things like floods, crop yields, energy usage, etc., and needed a sensing partner to help. At the time, SAS started to look at the LoRa ecosystem, which is about 500 companies today, and the company entered its own ‘maze of complexity,’ said Lansdowne, and decided it needed to work with a company at the center of the LoRa ecosystem.

“SAS are residents of the Town of Carey so between the three of us we thought what a great opportunity to bring these technology ingredients together to solve this maze of complexity and remove that pilot purgatory to provide a great outcome for Cary,” said Lansdowne.

One of the first projects the team worked on was flood predictions, which was initially started by SAS building hydrographic models before Semtech joined the partnership. However, SAS had a problem with the sensing engine. Some of the creeks were monitored with cellular-based river level monitors, powered by tiny solar panels and a battery, but after days of bad weather they would go offline due to a lack of sunlight to charge the batteries.

This is when SAS decided to look at LoRa technology, which uses very little power and enables data communication over a long range.

“LoRa as a technology is really built for very sleepy devices. Devices that wake up, make a measurement, and go back to sleep. Because they spend most of their life asleep they can last for an extremely long time on a battery,” said Lansdowne.

“LoRa has enabled a new breed of sensors creating data feeds from the edge that in many cases, we’re not previously feasible”, said Lansdowne,” and that critical data at the edge powers the SAS analytical models to capture value from the insights.”

LoRa-based, battery-operated sensors, which are easily installed, can be used for a variety of solutions, such as determining if parking spots are occupied; water-level, rainfall, and flow-rate monitoring for flood predictions, asset tracking, and waste management.

Architecturally the solution is a repeatable framework, which means it is secure, standardized, and scalable. “That’s how we avoid 300 siloed solutions. You just add the new sensors, drop in a new analytical model, and you get different insights,” said Lansdowne.

Once the data is in the cloud it’s ingested into the SAS analytics system at which point the LoRaWAN data is combined with data feeds from other sources, he said. The analytics create real insights and deliver solutions to the problem and instead of just being connected, they are becoming smart, he added.

Creating smart cities with connected and smart devices

An example of a Semtech and SAS IoT solution stack. Click for a larger image. (Source: Semtech)

A lot of IoT projects in Cary are still in the early stages with the team tackling one or two at a time. Some of those include smart parking, waste and wastewater management, utility metering, asset tracking, irrigation, and smart buildings, which covers a range of applications from occupation density to HVAC optimization.

Cary had an initial list of projects but “once you dive a little deeper into the things that LoRa does, it opens up the possibilities of what you can do and what is now feasible,” said Lansdowne.

For example, if an asset tracker needs to be charged once a week or once a month “you start to tradeoff the benefit versus the logistical nightmare of working out what batteries need charging, and where, and on what piece of equipment,” said Lansdowne.

However, if you can deploy the asset tracker with a battery that will last six to eight years on the device, and the device has a predicted life of six years, you’ve jumped over that logistics barrier, and it’s a very different equation, he added.

A key point Lansdowne makes is that data without analytics has very limited value and if there isn’t any data there isn’t any analytics. “The two need to live together and to go from data to insights is an order of magnitude in greater value.”

There is also a significant difference between something that’s connected and something that’s smart, he said. “If something is connected you’ve got data and if you use that data through analytics, you can create insights that allow you to predict and to take action based on those insights ahead of time, so that you can have a better outcome. That’s when things start to become truly smart.”

about Semtech