Wireless transceiver system for two-way voice simultaneous transmission

The carrier frequency of FM signal is set to 48.5MHz, and the absolute value of relative error is not more than 1‰; the peak frequency deviation is not more than 25kHz; the antenna length is not more than 0.5m.

Through the FM wireless transceiver system to transmit any way voice signal A or B, the bandwidth of the voice signal is not more than 3400Hz. requires wireless communication distance of not less than 2m, demodulation output voice signal waveform without obvious distortion.

The wireless communication distance is not less than 2m, and the waveform of the demodulated output voice signal has no obvious distortion.

This device is a wireless transceiver system for two-way voice transmission. In this design, it is mainly composed of two parts: transmitting part is composed of LPF circuit, AM subcarrier amplitude modulation circuit, main FM circuit and power amplification circuit; receiving part is composed of mixing circuit, frequency selection circuit and power amplification circuit. The transmitting part mainly consists of LPF circuit, AM subcarrier amplitude modulation circuit, main FM circuit and power amplification circuit; the receiving part consists of mixing circuit, frequency selective amplification circuit, demodulation circuit, local oscillation source circuit, signal processing circuit, FPGA acquisition circuit, STM32 microcontroller processing and display circuit, which can realize the communication distance more than 2m, two-way voice simultaneous transmission without distortion, and fast frequency tracking when the external VCO value is changed manually.

The system is divided into two parts: transmitting and receiving, and the main difficulty lies in the two parts: simultaneous transmission of two signals without distortion and fast tracking.

The generation of wireless interference is diverse, the original dedicated radio system occupies the existing frequency resources, different operators network configuration is not appropriate, the transmitter itself set up problems, cell overlap, environment, electromagnetic compatibility (emc), etc., are the causes of radio frequency interference in wireless communication networks, coexistence interference between systems working in different frequencies, essentially due to the non-perfection of the transmitter and receiver caused The Usually, the active device transmits useful signals at the same time, due to the device itself and the filter out-of-band rejection limitations, outside its operating band will also generate spurious, harmonic, intermodulation and other useless signals, these signals fall into the operating band of other wireless systems, it will form interference to them.

A and B voice signal in two-way transmission, if not processed may produce two-way crosstalk phenomenon, the second is to solve the problem of frequency tracking when two-way transmission, in order to solve the interference phenomenon, the system uses the method of spectrum shifting so that a channel can be transmitted to different frequencies of voice two does not produce interference.

Solution 1: Time-division multiplexing: Time-division multiplexing (TDM, Time-division multiplexing) is to divide the time provided to the entire channel to transmit information into a number of time slices (referred to as time slots), and these time slots are allocated to each signal source to use. You can use the RF switch to switch the two voice signals quickly, but also to consider the capacitance in the circuit, you need to do a part of the phase shift circuit, etc. as the compensation of the switching frequency, etc.

Option 2: Spectrum shifting method, spectrum shifting refers to the transmitter side of the modulated signal from the low-frequency end to the high-frequency end, to facilitate antenna transmission or to achieve different signal sources, different systems of frequency division multiplexing.

It is proved that, by taking the second option, as long as the B signal is AM modulated, the frequency division multiplexing method can be used to realize multiple voice signal transmission on one channel, and then demodulation and filtering are done in the receiving part to restore the original signal.

In this system design, three microcontrollers and one FPGA are used as the control circuit.

In the transmitting part, one microcontroller is mainly used to generate 60khz subcarrier signal, and another microcontroller is used to control MB1504 to generate FM wave and transmit.

In the receiving part, the FPGA is mainly used to collect the square wave signal after the comparator, so as to reduce the burden of the microcontroller, and then communicate with STM32. When the VCO is manually adjusted in the transmitting part, a frequency offset will be generated at the carrier frequency of 48.5M, and the microcontroller can get a deviation amount to form a closed loop, and this deviation amount will be adjusted by PID to automatically change the AD9959 output according to the change of the local This deviation is adjusted by PID to automatically change the AD9959 output source frequency according to the change, so as to achieve frequency tracking.

Before conducting the test, ensure that the circuit connection of each part is correct.

(1) which if you want to achieve the system requirements for frequency tracking range as large as possible, the attenuation of the selected frequency band in the bandpass filter can not be too large, so as not to affect the amplitude modulation circuit.

(2) according to the actual situation of the circuit, you can get only A signal at 8mv, B signal at 200mv waveform output without significant distortion.

(3) in the debugging process often appear in the module alone can be used, but after combining the front and rear circuit, the effect is often not good, should consider impedance matching, isolation through the crossover and other issues.

Error analysis.

(1) in the debugging process other groups of signals will generate interference, so in the debugging process can avoid the 48.5MHz central band, using other frequency bands to do the test.

(2) There is a lot of interference from other electromagnetic signals in the air.

(3) There is no common ground in the circuit connection and improper test methods can introduce different degrees of interference, making the system unstable.

Shortcomings and improvements.

(1) Too many Dupont line test interfaces should be avoided in the circuit.

(2) The whole system should be fixed on a plank to reduce various chance factors.

(3) FPGA is used as the acquisition point of square wave in the system, and the FPGA acquisition speed is very fast, but it fails to control the AD9959 frequency correction with FPGA, and it must do communication with STM32 MCU so as to control the AD9959 to generate the corresponding frequency offset for feedback, which makes the speed of the whole system not matching enough to do fast frequency tracking.

(4) During signal demodulation, interference signals are generated, and the A and B signals interfere with each other, making the waveform distorted seriously in a certain frequency band.

(5) The receiving part has a slight distortion due to the demodulated waveform, and after the comparator, the FPGA is unable to collect the current frequency due to a part of other signals mixed in it, the solution is to connect a capacitor to ground in parallel with the input acquisition port of the microcontroller, but the value of this capacitor is very critical, mixed with large frequency signals then use a small capacitor, small frequency signals use a small capacitor.

(6) impedance matching problem, some op-amp output waveform is very good, but after accessing the next level of circuit, the amplitude is severely attenuated or waveform distortion, a large part of the problem comes from the lack of isolation or matching problem.

Due to the above unstable factors, the system only completes the basic part, the frequency tracking can be completed, but not up to the play part of the finger

1.2.1 Power supply module

Program I: high-voltage differential linear power supply, for example: 78XX series power supply, the output power ripple is small, but the input power must reach a certain pressure difference to work properly.

Option 2: low-voltage differential linear power supply, for example: TPS7A4700, the output power supply ripple is small, but the current output capacity is not large, can not be high current output.

Option 3: using TPS5430 chip switching power supply, current output capability, but the noise factor is large.

1.2.2 Frequency selection amplifier circuit

Option 1: the use of discrete components 3SK228 transistor frequency selection amplifier circuit, through peripheral components, adjust its frequency selection bandwidth, the use of resonant frequency selection for amplification, in a certain frequency range

Linearity is very high, but the circuit is discrete components, the circuit is easy to self-excitation, in the range of 100M gain less than 10dB

1.2.3 Filter design and comparison

Solution 1: For high frequency signals, LC filters should be used so that the attenuation is smaller and the performance is better, in addition to the design of the parameters of the bandpass filter in this system also uses part of the low frequency signal filter, if the RC filter, the effect is not good, the passband is very wide, can not meet the requirements of the topic