Earlier this week, the Bluetooth Special Interest Group (SIG) announced the Bluetooth Low Energy (BLE) standards have been extended to include mesh network features. It is clear that the Internet of Things (IoT) is the intended market. The SIG says:
Bluetooth Mesh is “ideally suited for building automation, sensor networks and other IoT solutions where tens, hundreds, or thousands of devices need to reliably and securely communicate with one another.”
Mesh networks are not new. It is a network topology in which each node relays data for the network. All mesh nodes cooperate in the distribution of data in the network. The IoT-purpose-built Zigbee—a low-power, low-bandwidth ad hoc network—is a mesh network. Dating to 2002, Aruba Networks was founded to build Wi-Fi mesh networks. In 2014, student protesters in Hong Kong used mobile app FireChat to turn the crowd’s smartphones into a Wi-Fi and Bluetooth mesh network so authorities could not interrupt protester’s coordinating conversations by blocking 3G and 4G network access. Bluetooth Mesh has some very desirable features:
- Mesh networks are self-healing. If a node fails, packets are routed around the failure, provided there are a least two nodes within range of one and other.
- Standards bring interoperability.
- The combination of BLE mesh networks and BLE beacons have very interesting applications in asset tracking and providing location-based context to augmented reality (AR) applications.
- Multiple silicon sources from established semiconductor makers such as Texas Instruments, Nordic and Cypress foster competition and lower prices.
Is this the right time for an innovative IoT industry to commit to one standard above the data link layer?
The implementation of the BLE mesh standard at the physical and data link layers in silicon modules and radio ICs could result in competitive prices that would drive the adoption of IoT. But it does not have to be implemented above layer 2. Implementing further up the stack is questionable when there is so much innovation ahead in the IoT and continued research into protocol independent mesh networks that could accelerate innovation.
The Bluetooth SIG’s BLE mesh specification is very comprehensive, carefully considered and well engineered. The use cases—smart buildings, sensor networks and industrial networks—are also carefully considered. It is unclear, though, at this point in time when most IoT applications are prototypes if all the requirements of the use cases under development now and in the next few years are incorporated in the specification, and if a standards body can keep pace with rapidly changing requirements of an emerging new market.
Is a standards body like the Bluetooth SIG the right format for innovation?
There is not a Github repository accompanying the release of the BLE mesh spec. An open-source project that captures the software contributions of companies and independent developers could introduce new and better approaches to the implementation of the specification. To influence the specification, innovators who may need a modification to the specification must join the SIG, join a Working Group, contribute the modification to the specification, and if an agreement is reached, contribute this modification to the specification.
If the modification is not adopted, the innovator cannot implement the modification separately without conflicting with the SIG’s Bluetooth Patent/Copyright License Agreement. The agreement says:
“Compliant Portion” means only those specific portions of products (hardware, software or combinations thereof) that: (i) implement and are compliant with the actual Bluetooth Specification and/or Foundation Specification, whichever the case may be, (ii) are qualified pursuant to the Bluetooth Qualification Process, (iii) are within the bounds of the Scope and (iv) meet the requirements set forth in the Compliance Requirements.
That means the innovator could not use the extremely valuable Bluetooth brand to market the product. There is a little wiggle room for an innovator to engineer around the compliance issue because the agreement allows the implementation of functions not required for compliance.
Open source could meet the mesh specification in the middle
The Bluetooth Patent/Copyright License Agreement is not an open source agreement like the Apache or MIT Licenses that foster the contributions of improvements and enhancements back to the project. Access to the specification is free, but using the valuable Bluetooth brand requires a membership fee. The Bluetooth SIG communities benefit less from the virtuous cycle of innovation compared to open source communities. It also means the functions needed to implement the specification but not included in the specification will be redundantly developed by SIG members.
The workmanship in the Bluetooth Mesh specification to enable products to interoperate, interconnect or communicate is very impressive. An accompanying open source initiative to the announcement of the specification could accelerate the adoption of Bluetooth Mesh and save the redundant R&D of its implementation and ensure that the Bluetooth Mesh specification keeps pace with fast changing requirements.