5.8G wireless private network RF system based on WiMAX technology

As a wireless metropolitan area network technology, WiMAX has obvious advantages in bandwidth, coverage and data rate compared with the current 3G and WIFI, and has the characteristics of low maintenance cost, convenient installation and low construction cost. For such an emerging and promising wireless communication technology, there are still relatively few domestic institutions and enterprises that have carried out research and development. Basically, the products available on the market are foreign products with high prices. The research and development of wireless LAN products is of great significance to the development of wireless LAN in China. However, as a new technology, RF technology design is quite difficult, especially in the 5.8G frequency band, which puts forward high requirements for the linearity of the power amplifier, the noise index of the low noise amplifier and the phase noise index of the frequency source. The 5.8G wireless private network RF system designed by us based on WiMAX technology mainly adopts 0402 package device mounting, which is convenient for connection with the baseband circuit.

1 Working principle of RF receiving and transmitting system

The function of the RF receiving and transmitting system is to provide a wireless receiving and transmitting channel for baseband I and Q signals. In the transmitting time slot, the baseband I and Q signals are mixed twice, with the first intermediate frequency being fixed at 380 MHz and the second intermediate frequency local oscillator frequency being variable, so that the RF operates in the desired channel. The frequency-converted RF signal is filtered and power-amplified before being sent to the remote end by the antenna; in the receiving time slot, the signal received by the antenna from the remote end is low-noise amplified, mixed twice, and sent to the baseband processing board with equal amplitude under the control of the receiving AGC.

The RF module is directly connected to the baseband board through a 125-core socket. The signal connected to the baseband board is transmitted through the 125-core socket. The RF signal is installed on the PCB board through the MCX connector and directly output to the transceiver antenna.

2 RF transceiver module

2.1 Highly Integrated Chip

The intermediate frequency chip and radio frequency chip of the top company are selected as the highly integrated solution for this wireless radio frequency system. These chips are composed of IF and RF transceivers, supporting the 4.9-5.9 GHz air interface frequency band. The chipset can support the system to work in TDD mode through a complex I/Q interface. This highly integrated chip can reduce the space, which not only helps to simplify the design, but also saves the bill of materials (BOM) cost. The intermediate frequency chip has the characteristics of low noise and high linearity. It only needs one intermediate frequency filter, and also includes two intermediate frequency and radio frequency synthesizers, a high-speed digital variable gain amplifier, and its gain control range is up to 50 dB.

Function of the IF chip: Complete the up-conversion of the I/Q baseband signal to a fixed IF signal of 380 MHz in the transmitting time slot; complete the down-conversion of the received 380 MHz fixed IF signal to a zero IF I/Q baseband signal in the receiving time slot.

RF chip function: In the transmission time slot, the fixed intermediate frequency signal of 380 MHz is up-converted to the required RF channel frequency; in the reception time slot, the received RF signal is amplified and down-converted to a fixed intermediate frequency signal of 380 MHz.

2.2 Linear Power Amplifier

In digital microwave communication systems, the nonlinear distortion of the power amplifier has a great impact on the quality of signal transmission. In high-order QAM modulation systems, in critical cases, if the third-order intermodulation coefficient of the power amplifier deteriorates by 1 dB, the bit error rate will deteriorate by 80%. In this system, the power back-off method is used to improve the third-order intermodulation coefficient of the system. The power amplifier works 10 dB below the P1dB output power of the final power amplifier (considering the 1.6 dB insertion loss of the transceiver switch and the RF filter). When the maximum transmission power requirement of the system is 16 dBm, the P1dB of the final power amplifier should be at least 28 dBm. We use a high-efficiency linear power amplifier. When the output power is 21.5 dBm, the EVM is 3%. This indicator plays a decisive role in the transmitter's transmission constellation error indicator. The EVM indicator requirements of the entire transmission system are shown in Table 1.

2.3 Receive sensitivity and receive AGC control

Receiver sensitivity is an important technical indicator of the receiving system. For the 802.16d system, its receiver sensitivity can be calculated by the following formula:

Rss=-102+SNRrx+10Log((Fs*200)/256)

For 3.5 MHz bandwidth, Fs=3.5*8/7

SNRrx is the normalized signal-to-noise ratio requirement for system demodulation. For 64QAM-3/4, the normalized signal-to-noise ratio requirement is 24.4 dB.

By calculation Rss=-72.6 dBm.

The relationship between receiver sensitivity and noise figure is:

Rss=-174 dBm+101g BW+NF

Therefore, to meet the above-mentioned receiving sensitivity requirements, the NF of the receiver must be small enough under the condition of a certain bandwidth. The LNA noise coefficient we selected is better than 1.7dB, which fully meets the index requirements.

The receiving sensitivity requirements of SS are shown in Table 2 (BER≤10-6).

In communication systems, receivers must use automatic gain control (AGC) circuits to improve the dynamic control range of the receiver, so that strong input signals do not saturate the receiver and produce large distortion, and small signals do not go undetected by the receiver demodulator and are completely lost. In other words, the system is provided with appropriate gain and linearity to meet the requirements of receiving sensitivity.

In this system, the maximum normal receiving level is not less than -30 dBm, and it should be ensured that the receiving system will not be damaged when the input signal is 0 dBm; the minimum receiving level (i.e. the receiver sensitivity when the modulation mode is BPSK) is -91 dBm, and the AGC control range is 91 dB. The receiving gain control adopts a three-level gain control scheme: the first level is at the LNA, and both LNAs have a bypass switch function, which can set the current to 0 and achieve the minimum insertion loss. When a large signal is received, the bypass mode can adjust the dynamic receiving gain range to more than 26 dB; the second level is at the 5.8 GHz RF signal, and the RF chip contains a 15dB gain control range; the third level is at the 380 MHz intermediate frequency signal, and the intermediate frequency chip has a high-speed digital VGA control with a range of 50 dB.

2.4 Frequency Source

Since the RF local oscillator is not a perfect continuous wave single frequency source, but has phase noise, it will change the input signal at the output of the RF conversion stage. Since the phase of the digital signal carries information, the introduced phase change increases the bit error rate, and the high-order degree of modulation affects the degree of increase in the bit error rate. For the 64QAM modulation method, it is required to offset 1 kHz at the RF output frequency and the phase noise requirement is -88 dBc/Hz. At the same time, the frequency drift of the RF local oscillator (i.e., frequency stability) causes phase error in the demodulation process, resulting in a decrease in the effective signal amplitude and an increase in the bit error rate; the technical indicators of frequency stability depend on the modulation method used by the system and the user's requirements for communication quality. This system uses 64QAM modulation, requiring a frequency stability of ±1.5×10-6. The frequency stability and phase noise of the frequency source are another key technical indicator of the system. We plan to use loop phase-locked technology for frequency synthesis, and the reference source is generated by a high-stability crystal.

3 Test Results

The test was carried out in accordance with the requirements of the 802.16-based fixed broadband wireless access point-to-point RF technology. The AGILENT instrument E4438C was used as the signal source and the E4440C was used as the vector signal analyzer. The maximum output power of the system is 18 dBm, and the minimum power is less than -50 dBm. The channel bandwidth is 3.5 MHz, and the transmit EVM is better than -31 dB. See Figure 3 for specific indicators. The received signal level is -30 to -91 dBm, and the minimum received signal level is less than -91 dBm (BPSK modulation).

4 Conclusion

The 5.8 GHz wireless private network RF system equipment, together with the baseband processing equipment, was installed in two locations with a hop distance of about 10 km. The equipment worked stably and reliably despite various adverse weather conditions such as heavy rain and fog, and there was partial obstruction between the two locations. The bit error rate of each system was better than 10-7. It can transmit 10-50 Mb/s Ethernet data in both directions, and transmit movie images and sounds with good results.