Ripl Networks Introduces MLMesh, a Wireless Messaging System Enabling Battery-Powered Sensors to Talk IP

Share Article

Machine Learning Radio Adapter and Software Stack Delivers Edge-to-Cloud Encryption with Integration of Battery-Powered Sensors Into the IP WAN.

Ripl: IoT ShouldTalk IP
Ripl MLMesh is a Wireless Messaging System that transforms any IoT sensor into a wireless IP emitter and eliminates security risks, complexity and costs of having to run separate data networks for IoT.

Ripl™ Networks, the leading innovator in IP networking for low power things, today announced Ripl MLMesh™, a Wireless Messaging System that makes any IoT sensor a wireless IP emitter. Comprised of a radio adapter (Ripl-Edge™), border router (Ripl-X™) and software stack (Ripl-Lib™), the system resolves long-standing problems in enterprise IoT:

  • Eliminates the security risk, complexity and expense of running a separate type of data network for IoT due to non-IP protocols
  • Allows the enterprise to code, scale and manage IoT applications without any change to their IP-centric platforms, extending the ROI of their existing knowledge-base and infrastructure
  • Creates self-healing IoT networks that maximize the use of free wireless spectrum by leveraging machine learning at the edge
  • Generates better 3D location data using AI to enable asset and personnel tracking in fixed facilities and worksites including company events, firefighting and emergency response, without GPS dependencies
  • Enables impromptu networking for instant IoT deployments, with or without Internet connectivity, bypassing the need for 3rd party towers, gateways and service fees

The foundation of Ripl MLMesh is an ML powered software stack that coordinates guaranteed message delivery and symmetrical communications over an array of battery-powered radio options, including the Ripl-Edge. Ripl-based sensors enable data handling technology the enterprise demands such as TLS (transport layer security), blockchain and SD-WAN (software-defined networking) that’s absent in devices that do not use the ubiquitous IP protocol. Battery powered devices looking to live many years can’t take advantage of IP connectivity from WiFi and 4G/5G due to their high power consumption. Ripl creates for battery-powered sensors what WiFi does for laptops – making them a simple extension of the IP WAN.


Ripl MLMesh uses AI to learn about its radio frequency (RF) environment in order to coordinate and optimize radio messages across all nodes in the network. MLMesh resolves recurring issues such as loss-prone RF channels and collisions that occur in “spray-and-pray” type wireless architectures.

Ripl edge side code will allow integrators to reap the system benefits on existing battery-powered radios from other vendors, while recommending the software-hardware integrated Ripl-Edge for new deployments. Instead of protocol gateways that translate both addressing and cryptography schemes to IP, Ripl uses a low power IP mesh design using one or more border routers, licensing the necessary Ripl control software inside. Border routers are like WiFi routers but for battery-powered devices and enable Ripl nodes to seamlessly join and interact with the corporate WAN just like everything else. Ripl compliant radios typically use sub-ghz, spread spectrum protocols with up to 20 kilometers of range and battery life of up to 10 years.


IoT devices don’t need to travel extraordinary distances on separate networks when there is almost always an IP network close by. Ripl-enabled sensors can be deployed anywhere without the need for ethernet or power drops, and without requiring WiFi, Bluetooth or cellular connectivity, unless desired. MLMesh radios are inexpensive and support a meshed design where devices are closer together for self-healing resiliency and better battery life.

“CIOs don’t want a 2nd type of data network just for battery-powered IoT,” says Kerry Shih, Ripl founder. “If your devices don't talk IP you'll end up duplicating the security, coding and management costs when you really don’t have to.”


Ripl MLMesh first and foremost enables any sensor to function as an IP emitter, but the system also drives accurate, GPS-free location tracking born from AI perpetually trained in real world scenarios. The Achilles heel of RF-based tracking is that every environment is different and the typical set of heuristics create unreliable data sets. MLMesh allows integrators to train any environment by either walking or flying a drone through a worksite. The calibration creates an initial ground-truth model for the AI system which in turn creates a specifically-detailed understanding of what RF heuristics look like throughout the deployment site.

The trained AI system learns patterns that traditional calculations struggle with, especially with difficult locations i.e. near metal objects inside a building. MLMesh outputs continuous 3D metrics for any set of nodes and displays it in a real-time interface, all as a standard feature of the software library. As an example, a team could deploy MLMesh enabled sensors into an area using a drone and have an instant geo-fence to track personnel and assets without cellular or GPS connectivity.

The US Navy sees the value. It wants IoT sensors for everything from drone defense and fleet management to guest awareness on military bases. DoD Naval Surface Warfare Center’s Alan Jaeger confirms, “Battery powered sensors can solve several critical tasks for us and Ripl Networks provides a secured and integrated option.” Jaeger, an Office of Research and Technology Applications Manager further explains, “We need to track moving assets without relying on GPS using battery-powered trackers we could deploy literally anywhere. MLMesh is the only system I've seen that can do that.”


Analysts agree that the IoT market has been slow to reach its potential. Ripl believes integration difficulty born from non-IP protocols is a primary inhibitor. System integrators note that multi-million dollar deals are lost because CIO’s rightly don't want a separate data network for IoT projects. "Everything works better when everything's IP,” explained Bill Locker, CEO of Cisco Partner Sage Network & Communications. “We deploy Ripl because it leverages the value chain surrounding IP infrastructure including TLS security and SD-WAN performance. Ripl cuts out all the strange protocols of IoT that frankly customers don't care about and don't want to see competing as a second network."

This approach should be important to integrators building out secure industrial IoT (IIoT) and Industry 4.0 applications. “Industrial and manufacturing facilities have instrumentation data that is fed to IP networks in various ways. Ripl technology moves the packetization and security all the way out to the emitting device,” said Joe Weiss, IIoT security expert at Fortium Partners. "If it's going to be IP it needs to be IP from end-to-end. Protocol and encryption gateways present numerous and vulnerable intrusion vectors. Ripl eliminates their unnecessary threats to security. It’s a great IIoT solution for edge-to-cloud based security and we think it’s long overdue."

The Ripl MLMesh suite and hardware reference designs will be broadly available by Jan 1, 2021 based on the success of current projects. Pricing and packaging will be announced this autumn.

Ripl Networks accelerates IoT adoption by economically bringing sensors into the Enterprise’s IP realm. The Ripl Networks MLMesh system reduces the cost of implementing, administering and maintaining IoT projects and leverages the enterprise’s existing investment in IP knowledge and networks to do so. Ripl software, devices and reference designs remove threats to the industry achieving its vast potential by slashing dependencies on specialized expertise and extending the obvious and solid momentum of IPv6 for IoT. Visit or contact Ripl.

Ripl, Ripl-X, Ripl-Lib and MLMesh are trademarks of Ripl Networks Corporation. Other trademarks and registered trademarks are the property of their respective holders.

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Daniel Kehrer
(310) 454-1600
Email >
Visit website