If you could not attend live to the webinar from CEVA “Lear how to design a LTE-based M2M Asset Tracker SoC”, you have a second chance to access it remotely and to learn a lot. You will learn about CEVA’ Dragonfly platform 1 or 2, based on CEVA-XC8 or CEVA-XC5, and you will discover how mobile Machine 2 Machine (M2M) devices developed in the next years will use a combination of two technologies, cellular and positioning. Cellular for M2M like LTE or 3G is well known as well as positioning like Galileo Global Navigation Satellite System (GNSS), but the Time Difference On Arrival (TDOA) technology, is a novel positioning system based on existing 4G antennas.
The total cellular M2M connections, 243 million in 2014, are expected to reach 2 billion by 2020, with a 42% CAGR. But the M2M technology mix will change dramatically in the next 3 to 5 years: if the vast majority (60%) of WWAN modules were based on 2G in 2013, 3G technology will represent more than 50% in 2018 and 4G is expected to reach almost 70% in 2022. If the industrial smart meters dominates the M2M today, by 2018 the consumer market will dominate the industrial segment, and we expect about 500 M2M Mu/year to need both cellular and positioning technologies by 2020. Specifically the following market segments: asset tracking for containers, trucks, cars, appliance, kids, pets or bikes and smartwatch, smart grid, smart farms or smart plants.
If you take a look at the many cellular or positioning protocols on the above picture, you realize that you need to select a solution providing maximum flexibility by enabling multiple standards support. Moreover, this solution should be able to efficiently handle multiple standards concurrently. It should be low power (mobile M2M) and offer a cost low enough (we often hear about a $5 limit per module) to allow a wide development. All these requirements push for using a DSP based platform (flexibility) so the designer can optimize HW-SW partitioning (low cost and power efficiency) while handling multiple standard concurrently. That’s the description of CEVA’ Dragonfly reference platform, which is a SoC prototyping board with Radios, DSP and FPGA, going with a SW development platform with SW tools, RTOS, communication libraries and BSP.
CEVA has run a demo during MWC 2016, using Dragonfly to control drone navigation over LTE Cat-0, using satellite emulation for GPS:
As we can see on the picture, developing an asset tracker system (assuming the drone the asset in this case) requires the existence of an ecosystem. That’s why the two presentations from Galileo Satellite Navigation Ltd. of their GNSS solution and Nestwave of their Indoor/Outdoor positioning were very welcomed.
I think the satellite positioning technology is well-known today, even if it’s still a challenge to provide accurate positioning at low power cost and with acceptable complexity, so I will focus on OTDOA technology offered by Nestwave with their CellNav positioning system. CellNav doesn’t use satellites, but is based on existing 4G antennas. The base stations transmit a few subframes, the Positioning Reference Signals (PRS) and the mobile listens to the signals and compute their time on arrival. Similar to GPS, triangulation can be made and the mobile positioning determined with 5 to 30 m accuracy, expected to be enhanced to 1-5 m in the future.
In fact, it’s an extra option for geolocation, software only and low cost. Nestwave expect CellNav to enable new applications that need low power, always-on, fast time to first fix and ubiquitous indoor/outdoor geolocation. Pretty elegant solution from a company created in 2014!
You will get the complete picture, including GNSS or OTDOA computation or hardware description of CEVA Dragonfly, by going HERE to attend remotely to this webinar.
Eric Esteve from IPNEST