Approaches to Solving Multiradio Hardware Design Challenges

The combination of multi-radio and multi-protocol solutions can accommodate two or more radios running different multi-protocols simultaneously on the same or different spectrums.

Today, the market is driven to design products with multiple RF protocols in a box called a gateway. Wireless connectivity brings many different benefits, providing a better user experience, and different protocols provide complementary advantages. Each IoT device can communicate with the Internet through a different protocol, whether it is Zigbee, Bluetooth, Z-Wave or Sub-1ghz, or some proprietary protocol. Multi-protocol (wireless) gateways play a vital role in the IoT infrastructure because they collect data from the sensor field and push it to the Internet via Wi-Fi, cellular or other wired and wireless networks.

The combination of multi-radio and multi-protocol solutions can accommodate two or more radios running different multi-protocols on the same or different spectrum simultaneously. This approach provides more efficient and reliable data flow by utilizing different protocols. Therefore, end users can take full advantage of this as they are able to connect multiple devices running on different RF bands and protocols through a single unit.

Key Challenges in Designing Multiprotocol Compact RF Hardware

The emergence of multi-protocol hardware is a response to the recent popularity of multiple different communication protocols. As a result, OEMs face some key challenges when designing multi-radio hardware:

• Multi-radio hardware requires a significant amount of time in antenna selection, layout, simulation, memory estimation, enclosure design, materials, and field testing
• Controls precise impedance to reduce interference, return loss, and coexistence between co-located radios so it can comply with regulatory agencies such as FCC and CE
• Coexistence exists when two or more radios are operating simultaneously and sharing the same spectrum, which can cause mutual interference
• Measure performance parameters such as communication latency, range, efficiency, and reliability of multi-radio hardware
• Coexistence often affects the performance of devices, resulting in packet loss or data corruption, popping and crackling noises in the audio, and reduced operating range and coverage
• Regulatory compliance for different geographic regions will also be a challenge when multi-radio hardware emerges due to different standards applicability.

Things to consider when developing firmware for multiradio hardware

As IoT applications become more complex, the demand for storage capacity is also increasing. Let’s understand the challenges of hardware engineering solutions brought by memory and firmware:

• Developing user-friendly and flexible embedded applications requires complex state machines, strong power optimization, memory density, and CPU performance. RF SoCs or modules require more flash and RAM optimization for best performance
• Providing over-the-air (OTA) firmware update capabilities requires enough flash memory to store the bootloader and twice the size of the application firmware to buffer the new firmware while retaining the old one, making the product competitive in today’s booming IoT market
• Manage multiple radios with various network architectures in real time without losing performance
• Flash memory can also retain user configuration and security key content for years without power, as information in flash memory can be read and written thousands of times during the life of the product. In this case, information can be written to and read from RAM quickly, which allows the processor to operate at such high speeds and enable edge processing.

Based on the above challenges, let's take a small example, a complex wearable/sensor/automation application may require 128 kB RAM and 512 kB flash provided by the RF module. A relatively simple beacon application may only require 24 kB RAM and 192 kB flash.

To address the multi-radio hardware, memory, and firmware challenges described above, let’s look at how the OEM’s hardware expertise can be leveraged to address these challenges and help improve the overall performance of the product.

The standard approach starts with a product understanding of all RF requirements and other peripherals, lists all RF interface protocol frequencies, and then performs a task such as module or SoC selection, relative antenna selection, identifying the housing with materials, module placement, and antennas. The early stages of PCB design will be 2D floor planning, which helps to understand the actual location and parameters of all RF modules in detail:

• Module or SoC selection: Selection criteria include RF protocol, modulation technology, manufacturer, MCU/processor requirements based on driver code availability, RAM, Flash, OS, regulatory approvals, maximum Tx output power, receiver sensitivity, power supply, and data rate to provide significant performance
• Memory Budget: Memory budget is a very important parameter for any RF module and its definition or calculation is purely based on the RF stack size, number of devices supported, and application business logic. The memory related requirements should be well calculated and clear before finalizing the module/SoC.
• Antenna selection and placement: Antenna selection is the most important factor in multiradio hardware as the choice depends on frequency range, polarization, radiation pattern, gain, feed point impedance, standing wave ratio, and power handling capability for use cases such as range coverage and space. For example, chip antenna vs PCB trace antenna vs external antenna

To reduce the coexistence and interference between two co-frequency modules, antenna placement plays a vital role. In this case, the antennas should be placed perpendicular to each other. We are working on proper isolation between the two RFs as per regulatory requirements, and the isolation between the two antennas should be equal to or greater than 30 dB. Sometimes, due to space constraints, it is not possible to achieve such isolation, in which case the interference and radiation pattern of all RF antennas need to be considered. We also need to study the transmit power of the RF module or SoC based on the application requirements and performance.

• RF simulation: Simulation is an effective strategy to predict radio problems in the early stages of product development. There are a variety of simulation software, such as HFSS, CST, ADS, etc., which should be used appropriately according to the type of problem.
• Enclosure design and materials: For best results, try to finalize a plastic enclosure with a material that balances environmental, reliability, and RF performance needs. Sometimes, COTS enclosures may cause placement challenges due to the already predefined size and structure, while in custom designs, we have the flexibility of PCB placement and can get better isolation and antenna placement options. The choice of dielectric constant/dielectric constant of the enclosure material plays a crucial role in RF performance. At 2.4GHz, good plastic materials are PC, ABS, PC+ABS, PVC. Sometimes we also need to choose a metal enclosure for outdoor and industrial applications, in which case the choice of external antenna will be key.
• Design Verification Test: Design verification of the RF interface is very important. A set of verification test cases need to be defined, such as antenna matching impedance and return loss, RF output power measurement, receiving sensitivity, indoor and outdoor distance testing, and worst-case RF environment under heavy traffic scenarios, and monitored using an analyzer.