How to plan the frequency of transceivers

btty038

How to plan the frequency of transceivers [Copy link]

Today a friend complained to me that the technical indicators of the frequency conversion channel given by the user were too difficult, especially the frequency planning plan, which was so confusing that he had no idea where to start.

Based on my work experience, I will briefly explain frequency planning, focusing on the key points without going into details. Given my limited work experience, my views are for reference only. We welcome your comments.

Since I am editing this on a mobile phone, I will not give the formulas here. I hope you understand.

Generally, users will provide the frequency range of the RF, the frequency of the IF and the instantaneous bandwidth.

According to the frequency, the frequency conversion channel can be divided into several main types. It can be roughly divided into broadband transceiver; narrowband transceiver; broadband receiver; broadband transmitter; narrowband transmitter.

From the perspective of frequency band, common products are divided into Ka, Ku, X, C, S, L, etc.

Frequency division may also be affected from an environmental perspective. For example, in some product environments, users will not participate in the frequency division plan. When the surrounding frequencies are more complex, users may explicitly require the receiver's local oscillator leakage, intermediate frequency suppression, etc., which will affect the specific frequency planning.

However, no matter how the frequency plan is divided, the following considerations are essential:

1. In-band spurious emissions must meet index requirements;

2. Image frequency suppression must meet index requirements;

3. Out-of-band spurious emissions must meet index requirements;

4. Is there a better solution (more cost-effective)?

5. Whether the device can achieve frequency allocation;

6. The project can achieve spurious suppression;

7. The process can meet the requirements of product realization;

8. What is the quality of the product?

Of course, the details may include more than just the above.

If it is a narrowband transmission and reception, then consider whether the first mixing can meet the indicators under narrowband conditions. Because the first mixing solution has a small number of components, low product power consumption, simple link, and high cost performance. If the first mixing cannot be done, such as high link gain requirements, or some mixing spurs fall within the band and cannot meet the indicator requirements, basically the second mixing solution can meet the indicator requirements. The first intermediate frequency of the second mixing solution should be as low as possible, because there is a little advantage of low frequency, and packaged components can be selected to improve product quality. However, everything is relative. The bad news brought by low frequency is that the local oscillator leakage spurious is difficult to suppress, and the image frequency suppression is difficult. The two local oscillator mixing spurious may also fall near the intermediate frequency and are difficult to suppress.

If it is broadband transmission and reception, then image frequency suppression must be considered, and segment processing must be considered at the receiving front end; at the same time, the width feasibility of a local oscillator and the requirements for phase noise must be considered. There are many broadband frequency conversion schemes, each with its own advantages and disadvantages, and they must be compared and selected based on actual conditions. For example, 2-18G broadband reception can be divided into two segments for processing, 2-8G and 8-18G. The intermediate frequency corresponding to 2-8G can be selected to 10G, and the corresponding local oscillator range is 12-18G, which can be achieved with HMC733. The 8-18G signal can be converted to the intermediate frequency at one time (if the intermediate frequency is around 2G). If the intermediate frequency is low, such as 70MHz, then it cannot be solved by one frequency conversion. Of course, the intermediate frequency of this product will not be too low.

For IQ mixers, attention should be paid to image suppression and local oscillator leakage, as well as the problem of deterioration and superposition of flatness in different frequency bands if the bandwidth is relatively wide.

Some receivers have a low input frequency range, such as 30M~8G. In this case, you need to communicate with the user to directly sample at the low end of the frequency. How many G can the high end of the sampling frequency be? Generally, there will be no problem up to 500M.

Some receivers have a relatively wide intermediate frequency bandwidth, such as 1.2G as the center and a bandwidth of 1GHz. This makes the frequency division of the receiver more annoying, because it is difficult to suppress the local oscillator for up-conversion and difficult to suppress spurious signals for down-conversion. The frequency allocation relationship needs to be reasonably planned.

Some receivers require relatively high phase noise, so a compromise needs to be considered as to whether a very high intermediate frequency solution is needed.

Some receivers require instantaneous frequency measurement function, so there are special response plans to divide the receiving frequency into segments.

Well, that's all for today. In fact, there is much more to it than this. I'll go on this journey with you again when I have the chance.

It's just a word of experience-l from RF Life