Types of output interfaces of MEMS microphones

The output signal of ECM (Electret Condenser Microphones) is almost all analog, and most of them are single-ended. However, in contrast to MEMS Mic, there are two types of output interfaces: "Analog" and "Digital". Analog is also divided “Single-ended” and “Differential” output. As for the digital which contains “PDM”, “I2S” and “TDM” output. PDM occupies the largest proportion.

The following introduces the MEMS Microphone output interface categories:

1. Analog Single-ended: This is a commonly used format, and the price is usually lower than the Differential type. The sensitivity is usually in the range of -38dBV ~ 40dBV. All the audio codec microphone inputs support this signal and make it the most popular one. But when designing the PCB circuit for this kind of microphone, noise can be avoided referring to Fig. 1

2. Analog differential-ended: This type is not common, and the price is usually higher than Single-ended. It is a Differential Output, therefore the sensitivity will also increase 6dBV, the sensitivity is usually set around -32dBV, and it can also increase the AOP (Acoustic Overload Point). Most audio codec microphone inputs also support this signal. Fig. 2 helps noise reduction, and Common mode noise could be eliminated by using differential circuitry Fig 2-1

Fig.2 Analog MEMS Mic differential ended

Fig.2-1 Differential Common Mode

3. PDM (Pulse Density Modulation): It is the main output interface of a digital MEMS microphone.  The Data BUS is shared by the L/R channel so it only needs 4 lines to complete the stereo L/R channel transmission, it is also good for the PCB layout design of portable devices. And it is digital to require a clock signal (usually 1MHz to 4MHz), so the audio bandwidth can meet the application of 20Hz to 20KHz, which is widely used in consumer electronics, automotive, and IoT devices. Fig.3

Fig.3 Digital MEMS Mic PDM Interface

4. I2S (Integrated Interchip Sound): This is a digital audio serial bus standard format developed by Philips for the transmission of audio data between digital audio devices. The bus is specifically used for data transmission between audio devices, widely used in various multimedia systems. Most entry-level MCUs don't support PDM interfaces, but basic I2S interfaces. Therefore, some microphone manufacturers still launch I2S MEMES microphones for applications without an audio codec. Fig.4

Fig.4 Digital MEMS Mic I2S Interface

5. TDM (Time-division multiplexing): It divides the usage time of the transmission medium several fixed time slots. Each time slot occupies a short period (for example, 20ms), and this short time is considered as a virtual channel. When data are communicating, the two interconnected parties will be set to transmit data on a certain time slot, and for a certain time it has the right to use the transmission intermediary; and for a longer time, the entire transmission intermediary is the same as having multiple connections and they are sending data in the same time. And because the smart speaker needs to perform "Far-Field Voice Pickup", so when the Beamforming microphone can no longer meet the criterion, the 6 to 8 pcs microphone array has been developed. An increase of microphones means more channels are needed (ex. PDM and I2S) thus in order to reduce the number and cost of microphone digital interfaces, TDM time, division multiplexing is used to transmit signals in turn to achieve low-cost multi-channel transmission.

Fig.5 Digital MEMS Mic TDM Interface

6. Conclusion: The above is an introduction to various microphone output formats. The correct microphone specifications can be selected according to product application requirements to meet the design criteria of high performance and low cost.