Application of Digital microphone and array sound pickup Technology

With the development of digital signal processing technology, more and more electronic products are using digital audio technology. Digital audio interfaces have become a trend of development. The ECM and MEMS digital microphones using pulse density modulation (PDM) interfaces are also born. Currently, ECM and MEMS digital microphones have become the mainstream for portable laptop sound pickup devices.

Digital ECM or MEMS microphones have irreplaceable advantages over traditional ECM microphones. First, mobile devices are evolving towards miniaturization and digital devices are in urgent need of digital pickups and technologies. Second, more and more functional units are included, such as laptops, which integrate Bluetooth and WiFi wireless functions, microphones are very close to these interference sources, and the requirement for device interference resistance is getting higher and higher. Third, the development of triple play requires Internet access, and video and voice communication can be performed simultaneously, this usually affects the environment noise and echo of mobile devices. Fourth, we hope to use SMT Welding for microphones to improve production efficiency. Digital microphones are suitable for SMT welding and can solve the noise produced by various RF interference on voice communication. The digital array microphone pickup Technology of fudi technology can suppress and eliminate echo and environmental noise during calls, the digital interface facilitates connection with the digital system.

Analog microphone and digital microphone

Microphone structure: the ECM analog microphone is usually composed of a diaphragm, a back plate, a junction-type FET (JFET), and a shielded housing. A vibrating film is a thin film coated with metal. The back plate is made of electret material and carries a charge after high voltage polarization. The two form a flat capacitor. When the sound causes vibration of the diaphragm, the distance between the two changes, resulting in voltage changes, complete the acoustic conversion. The junction FET is used for Impedance Transformation and amplification. Some high-sensitivity microphones use op-ops to improve the sensitivity of the microphone (see Figure 1a ). The ECM digital microphone is usually composed of a vibrating film, a back plate, a digital microphone chip, and a shielded housing. The digital microphone chip is mainly composed of a buffer level, a zoom-in level, a low-pass filter, and anti-modulus conversion. Buffer-level Impedance Transformation, amplification-level amplification of signals, low-pass filtering to filter out high-frequency signals, prevent mixing of Analog-to-analog conversion, analog-to-analog conversion to pulse density modulation (PDM) signal, usually using a one-bit sampled Delta-Σ modulus conversion (see Figure 1b ). The MEMS analog microphone consists of MEMS sensors, charging pumps, buffer amplifiers, and shielded housing. See Figure 1c,
A mems sensor consists of a diaphragm, a back plate, and a Bracket Made of a semiconductor process. A charging pump is used to add an appropriate polarization bias to the back plate. The buffer amplifier completes Impedance Transformation to enlarge the signal. The MEMS digital microphone consists of a MEMS sensor, a charging pump, a digital microphone chip, and a shielded housing. See figure 1D. To improve the anti-interference ability of the microphone, a small filter capacitor is added between the power supply and the ground of the microphone. Generally, the 10pf and 33pf are connected in parallel.

　　

 

Figure 1a ECM analog microphone Figure 1b ECM digital microphone

　　

 

Figure 1c MEMS analog microphone figure 1D MEMS digital microphone

Microphone offset circuit: a typical application of the microphone circuit in a mobile phone. Compare the differences between an ECM analog microphone, a MEMS analog microphone, and a digital microphone. Figure 2a shows the offset circuit of the ECM analog microphone. To reduce interference, the microphone circuit in the mobile phone uses Differential output. After the microphone power is filtered by the R5 resistor C9 capacitor, the R6 is used to provide the field effect tube inside the microphone. The differential output circuit consists of R6 and R9. C15, R6, R9, and microphone output impedance constitute a low-pass filter, used to filter out the high-frequency signal that exceeds the voice frequency band, to prevent the rear-Level Circuit Module installation and replacement of generation mixing. C13 and C17 are isolated from DC bias. R7 and R8 are used to prevent the impact of capacitors on the discharge of the chip input end. The remaining 33pf capacitors are used to filter out RF interference. The analog input from the microphone output to the Baseband Chip adopts differential wiring to reduce noise and RF interference (see Figure 2b ). The bias circuit of the MEMS microphone. After the microphone power is filtered by the R1 resistor C2 capacitor, it provides the buffer amplifier and charging pump circuit built into the MEMS microphone. The sound signals picked by MEMS are converted into analog electrical signals. After buffering and amplification, the output is filtered by a π filter consisting of C5, R2, and C6, and the pseudo-differential circuit is routed to the Baseband Chip. Figure 2C shows the offset circuit of the digital microphone. The microphone power supply is supplied to the microphone after simple filtering. Audio is converted into analog electrical signals that are amplified by internal buffering. Driven by the clock signal (SCL), the modulus is finally converted into one-bit PDM audio data, which is output from the data pin.

　　

 

Figure 2a ECM analog microphone Circuit

　　

 

Figure 2B MEMS analog microphone Circuit

　　

 

Figure 2C ECM/memes digital microphone Circuit

Comparison of various types of microphones: Table 1 lists the performance indicators and their respective advantages and disadvantages of ECM analog microphones, ECM digital microphones, MEMS analog microphones and MEMS digital microphones.

　　

The analog signal is converted into a PCM signal. According to the nequest principle, a fixed sampling frequency greater than twice is usually used to sample the analog signal. Module replacement. Each sampling point can be represented by multiple bits. The larger the number of BITs, the higher the sampling accuracy, the smaller the distortion, but the circuit will be complex, the high cost, is not suitable for low-cost digital microphone applications. 3B. A digital microphone uses a one-bit Delta-Σ Analog-to-analog converter to oversample analog signals (it can only be used for signals with limited bandwidth and is not suitable for broadband signals, such as video signals ), the sampling rate is provided by the external clock. Over-sampling can make the quantization noise stay away from the sampled audio signal. The closer the signal frequency fs, the smaller the noise. At the same time, the demand for antimixing filters is greatly reduced, and high accuracy can be achieved.

　　

 

Figure 3 PDM Signal

The digital microphone is usually composed of five pins, namely power supply (VDD), ground (Gnd), clock (CLK), data (DAT), and channel selection (L/R ). The digital microphone interface chip must provide microphone power (must match the system level) and external clock signals (1.024 ~ 3.074 MHz). After obtaining the clock signal, the digital microphone changes from power-saving to normal working. Extract audio signal oversampling and convert it into the data stream of pulse density modulation (PDM) (the more intense the signal amplitude changes, the more dense the pulse density) to the processing chip, the extraction filter inside the chip (decimator) low sample and low-pass filtering, which converts signals of high-frequency low-bit streams into PCM signals of low-frequency high-bit streams, and filters out quantization noise. The PDM interface can be mounted to two digital microphones to share the clock and data lines. The channel is used to select (L/R) the microphone of which channel is used for clock height and low time. Figure 4 shows the output signal of the digital microphone. When the clock is high, the L/r = 0 microphone (mic0) data line maintains a high impedance, transmission L/r = 1 microphone (mic1) data; when the clock is low, l/r = 1 microphone (mic1) data line to maintain high impedance, transfer L/r = 0 microphone (mic0) data.

　　

 

Figure 4 digital microphone output signal

Application of Digital microphone array on mobile phone platform

The two digital microphones use the same group of power supplies, and the power supply voltage is the same as that of the voice processing chip FM34-395 (see figure 5 ). The digital microphone array is configured as the main microphone (L/R grounded) and reference microphone (L/R powered) through the L/R pins ), the pickup signal is amplified by a digital microphone and converted into a PDM signal connected to the FM34-395 chip of the speech processing chip. The two-way microphone signals are replaced by 16-bit PCM signals after downsampling, which are amplified and filtered for processing.

　　

 

Figure 5 typical application of digital Microphone Array in MTK Mobile Phone platform

In hand-held mode, based on the difference of the near-end signal picked up by the digital microphone array, the steady-state and non-steady-state noise suppression and linear echo elimination are performed on the near-end voice. The processed signal passes through the PCM pin (txdp) send to the Baseband Chip (PCM in pin) as the uplink signal, send and generate a side tone to the receiver ). The downstream signal is sent to the FM34-395 Input Pin (rxdc) through the Baseband Chip (PCM out pin), after noise suppression, determine whether to enable a clear voice engine to brighten the voice of the receiver based on the noise information provided by the digital microphone array, so that users can still hear clearly in a noisy environment, it also serves as a reference signal for echo cancellation.

In hands-free mode, based on the echo signal difference picked up by the microphone array, the near-end speech with mixed acoustic echo is eliminated, and the processed speech is subjected to steady-state noise suppression, the output signal is sent to the Baseband Chip (PCM in pin) as the uplink signal through the PCM pin (txdp. The downstream signal is sent to the FM34-395 Input Pin (rxdc) through the Baseband Chip (PCM out pin), after noise suppression, determine whether to enable a clear voice engine to brighten the voice of the receiver based on the noise information provided by the digital microphone array, so that users can still hear clearly in a noisy environment, it also serves as a reference signal for echo cancellation.

Placement of the digital microphone array: The main microphone is placed at the front or bottom of the mobile phone, that is, close to the user's mouth as much as possible. The reference microphone is placed at the top or top of the mobile phone back, that is, close to the user's ears, in this way, in hand-held mode, there is enough difference between the signals picked by the two microphones in the near-end voice digital microphone array, while there is no difference in far-away noise, using a digital voice processor FM34-395 processing, you can achieve targeted distance pickup to suppress various environmental noise.

Application of Digital Microphone Array in laptop

Figure 6 shows the typical application of a digital Microphone Array in a laptop. The digital microphone array and the camera module are usually installed in the center of the laptop display, so that the user can transmit the sound source in the digital Microphone Array pickup bundle during video chats or calls, the noise on both sides is restrained outside the sound collection bundle to achieve clear speech communication. The digital microphone in the array can be placed at 5mm as a small digital microphone array, or 70 ~ 210mm wide array, configure the corresponding software according to the microphone.

　　

 

Figure 6 typical application of digital Microphone Array in laptop

Digital microphone pickup and conversion of signals to PDM format is connected to the digital microphone interface of the HD audio decoder (HD Audio Codec) Sound Card of the laptop using cables, downsampling is converted to two-way audio signals, it is delivered to samsoft, a small microphone array processing software located in the driver layer of the HD audio codecs to Implement noise suppression and acoustic echo cancellation ), far Field pick up (far field pick up) can be used to suppress keyboard noise.