Base Band Processor for Sub 1-GHz (gateway for cc1310/cc1350)

Farshad Firouzi,

If you want to learn more, please visit our website HUAXUN.

1) I was wondering how many gateways (in how many channels) I need to buy and what model is suitable?

• you did not mention anything about range, range is an important variable in setting up a system. A basic rule of thumb, would be about 500 nodes per gateway is a good target.

2) in LORAWAN, I know that it has a maximum payload of 222 bytes (including 13 bytes header) and even for only that on the fastest rate (spreading factor 7) the nodes would have to wait about 35 seconds between those packets to comply with the 1% legal duty cycle regulations. So my WSN might not be compatible with LoRa no matter how many gateways you have (unless I increase the sensor readout interval). I was wondering what is the case in CC/cc.

• In the stack we have publised, we support 2 data rates, 5kbps and 50kbps. 5kbps is our long range mode. For the 1% dutycycle requirement, why do you not consider the 868MHz band or 915MHz (in the US). There you are allowed to go to 100% dutycycle is your system support LBT (Listen before talk) or Frequency hopping for the US market. Our stack supports both these features.

3) Semtech has a a Base Band Processor (SX) for Data Concentrator. I would like to know, TI has a similar processor, or I need to design a gateway with multiple cc/cc (for each channel)?

• We have the same concept.

4) In 433MHz, how many channel I can use with CC/CC? Do I need to assign a static channel to each gateway, or sensor nodes are smart can can send to different frequency channel if one of them is busy?

• I cannot remember the exact number but its approximately 10-15 channels.

5) Assume some sensors are in the range of two gateways (and each gateway can listen to all the frequency channels), what happens in this situation? Shall I statically assign each to one gateway or it would be better to rely on a dynamic approach and send the sensor data to all the nearby gateways?

• Each sensor has joined a specific gateway.

6) Finally, is there any wireless device classes (e.g. class A, B, C) similar to LORA in cc/cc?

• We do not define classes of devices.

7) Is there any official/commercial gateway for cc/cc (for STAR network using 15.4)?

• No.

8) Finally is there any excel file for some bandwidth calculations (similar to LORA) with all the parameters like this one:

One of the earliest uses of 5G will be fixed wireless access (FWA), which promises to deliver gigabit internet speeds. FWA can be delivered to homes, apartments and businesses in a fraction of the time and cost of traditional cable/fiber installations. As with any technological advance, FWA brings new design hurdles and technology decisions. Let’s dig into five things to consider when designing FWA systems:

• The choice of frequency spectrum: millimeter wave (mmWave) or sub-6 GHz
• Achieving higher data rates with antenna arrays
• All-digital or hybrid beamforming
• Power amplifier (PA) technology choices: silicon germanium (SiGe) or gallium nitride (GaN)
• Choosing components from today’s RF front-end (RFFE) product portfolios

#1: Spectrum choice: mmWave or sub-6 GHz

The first decision is whether to use mmWave or sub-6 GHz frequencies for FWA:

• mmWave. These higher frequencies offer a large amount of contiguous spectrum available at low cost. mmWave supports component carriers up to 400 MHz wide and enables gigabit data rates. The challenge is path loss due to obstacles like vegetation, buildings and interference. However, don’t assume FWA is useful only in clear line-of-sight settings between the base station and the home — FWA can actually perform very well in both urban and suburban settings. It’s true that vegetation and interference are challenging, but these can be overcome with antenna arrays that provide high gain.
• Sub-6 GHz. This lower-frequency spectrum helps overcome the problems caused by obstructions, but at a cost. Only 100 MHz of contiguous spectrum is available, so data rates are lower.

 
Efficient use of frequency range (sub-6 GHz or mmWave) is critical to scaling deployments. The choice for any situation will depend on balancing the goals of speed and coverage.
 

#2: Achieving higher data rates with antenna arrays

An FWA system will also need to employ active antenna systems (AAS) and massive MIMO (multiple input/multiple output) to deliver gigabit service.

Want more information on Baseband board? Feel free to contact us.

• AAS provides many directional antenna beams. These beams are redirected in less than a microsecond, enabling beamforming that offsets the greater path loss associated with high frequencies.
• Massive MIMO uses arrays of dozens, hundreds or even thousands of antennas, allowing simultaneous transmission of single or many data streams to each user. The results are improved capacity, reliability, high data rates and low latency. Beamforming also enables less inter-cell interference and better signal coverage.

Learn more about AAS and massive MIMO: How Carrier Networks Will Enable 5G
 

#3: All-digital or hybrid beamforming

A third element to consider is the type of beamforming to employ — all-digital or hybrid.

All-digital approach

Hybrid approach

An alternative is hybrid beamforming, where the precoding and combining are done in both baseband and RF front-end module (FEM) areas. By reducing the total number of RF chains and analog-to-digital and digital-to-analog converters, hybrid beamforming achieves similar performance to digital beamforming while saving power and reducing complexity.

Another advantage of the hybrid approach is the ability to meet both a suburban fixed or limited scan range (<20º) and dense urban deployments with wide scan ranges in both azimuth (~120°) and elevation (~90°).

The bottom line: all-digital and hybrid approaches both have advantages and disadvantages. We believe the hybrid approach is more appealing and doable today, but new products on the horizon could make the all-digital approach equally appealing in the future.
 

#4: PA technology choices: SiGe or GaN

To achieve 75 dBm EIRP with a uniform rectangular array, the PA power output required per channel reduces as the number of elements increases (i.e., the beamforming gain increases). As shown in the below figure, as the array size gets very large (>512 active elements), the output power per element becomes small enough to use a SiGe PA, which could then be integrated into the core beamformer RFIC.

As you can see from the table below, a SiGe PA can achieve 65 dBm EIRP using  active channels. However, by using GaN technology for the front end, the same EIRP can be achieved with 16x fewer channels.

A GaN FWA front end provides other benefits:

• Lower total power dissipation. To ensure an accurate comparison, the GaN power dissipation includes an extra 19.2 watts, to account for the 128 beamformer branches needed to feed the front ends. As shown in the following figure, at the target EIRP of 65 dBm, GaN provides a lower total power dissipation (127 Pdiss) than SiGe. This is better for tower-mounted system designs.

• Better reliability. GaN is more reliable than SiGe, with >107 hours MTTF at 200°C junction temperature. SiGe’s junction temperature limit is around 130°C.
• Reduced size and complexity. GaN’s high power capabilities reduces array elements and size, which simplifies assembly and reduces overall system size.

 
The takeaway: In wireless infrastructure applications, reliability is imperative because equipment must last for at least 10 years. For FWA, GaN is a better choice than SiGe for reliability, cost, lower power dissipation and array size.
 

#5: Choosing from today’s RF technology

The last consideration is selecting product solutions that are being used in real-world applications. Several RF companies are positioned to support the development of sub-6 GHz and cmWave/mmWave FWA infrastructure. Qorvo, for instance, is already supplying products for many Tier 1 and Tier 2 supplier field trials. Across the RF industry, examples of products for FWA include:

• Sub-6 GHz products: Dual-channel switch/LNA modules and integrated Doherty PA modules
• cmWave/mmWave: Integrated transmit and receive modules

 
Additionally, in the 5G infrastructure space, several things are a must:

• Integration
• Meeting passive cooling requirements at high temperatures

 
To support these trends, Qorvo has created integrated transmit and receive modules for cmWave/mmWave, as well as integrated GaN FEMs. These integrated modules include a PA, switch and LNA, and have high gain to drive the core beamformer RFICs. To meet the infrastructure passive-cooling specification, we use GaN-on-SiC to support the higher junction temperature.

For more information on Qorvo solutions for FWA, click on the images below or visit our 5G Infrastructure page, where you'll find product details and interactive block diagrams.

Are you interested in learning more about NOKIA Baseband Board? Contact us today to secure an expert consultation!

FWA is approaching — fast