Analog front-end (AFE) design for sensor interfaces

High performance, low-power consumption, and precise signal processing for advanced applications

CSEM specializes in the design of analog front-end (AFE) circuits tailored to a wide range of sensing applications. Our AFEs are diligently engineered to meet the specific needs of different domains, ensuring high performance, low power consumption, and precise signal processing. The following sections highlight our expertise and portfolio of AFEs in three key application areas:

  • AFEs for wearable and implantable medical devices, providing reliable, accurate, and efficient performance in a variety of applications from wearables to implantable devices.
  • AFEs for audio applications, designed for ultra-low power, high-resolution, high-dynamic range, and low-latency audio applications from consumer electronics to professional audio equipment.
  • AFEs for electrochemical sensing applications, optimized for sensitivity, ion selectivity, and low power consumption applications from diagnostics to environmental monitoring applications.

AFEs for wearables and implantable medical devices

With over 20 years of experience in developing medical technologies—from sensors and SoCs to complete system designs—CSEM’s biomedical AFEs deliver state-of-the-art performance. We focus on:

  • Noise-optimized signal acquisition for unparalleled biomedical precision.
  • Ultra-low power and compact designs for minimally invasive wearables, portables, and implants.
  • Regulatory-compliant solutions that detect even the weakest biomedical signals without distortion.

Detecting electrophysiological signals is crucial to a variety of diagnostic applications. In neuroscience, EEG signals help us understand brain functions and disorders. In cardiology, ECG signals are used to diagnose and treat cardiac conditions. Additionally, in muscle physiology, these EMG signals enhance the understanding and diagnosis of muscle disorders. CSEM focuses on these domains to provide medical-grade signal acquisition and processing solutions.

Below are examples of CSEM’s AFE-integrated ASICs and SoCs.

Multi-channel ECG, EEG, and EMG Front Ends

  • 20 low-power, low-noise configurable sensing channels that can simultaneously acquire and process EMG (electromyography), EEG (electroencephalography), and ECG (electrocardiography) signals.
  • 1uV to 10 mV input signal amplitude in frequency range from sub-Hz to 10 KHz. With a dedicated data management unit for multiple-channel simultaneous ExG sensing and lossless data compression.
  • Very low noise (µVrms) within the signal bandwidth combined with high input impedance (GΩ-level) to support interfacing with dry electrodes.
  • Capacitive input to support large electrode offsets.
20 channel electromyogram (EMG) AFE system-on-chip architecture

Block diagram for 20-channel electromyogram (EMG) AFE system-on-chip architecture.

Die micrograph of a multi-channel electromyogram (EMG) AFE system-on-chip

Die micrograph of a 20-channel electromyogram (EMG) AFE system-on-chip.

AFEs multimodal and multichannel cooperative sensors

While traditional multi-lead ECG systems require a shielded cable for each electrode in a star configuration, CSEM’s patented cooperative sensor solution integrates an ASIC on each active electrode, allowing connectivity with just one unshielded cable. A central battery supplies all active electrodes through the same cable that carries all communication between the sensors. This topology using unshielded cabling enables cost-effective integration of active dry or gel electrodes with wearables, using conductive textiles or minimal wiring.

  • Patented CSEM readout SoC interface to support wearable multimodal and multichannel systems
  • Multichannel modal for EEG, ECG, EMG, bioimpedance, and sound recording for cardiac, respiratory, muscle, and brain monitoring are designed to operate with a single battery.
  • Multichannel ExG/EIT readouts connected over a 2–wire bus with unshielded cables to support garment integration and enhance patient comfort.
Concept of traditional star-connected biopotential acquisition system (left) vs. cooperative sensors (right).

Concept of traditional star-connected biopotential acquisition system (left) versus CSEM’s patented cooperative sensors (right)

cooperative sensors

CSEM’s patented cooperative sensors for electrocardiogram (ECG).

Silicon-proven chip implementations of cooperative sensor AFEs

Analog Front Ends for implants

At CSEM, we lead in developing advanced analog front ends (AFEs) that are crucial to invasive, minimally invasive, and implant applications. These AFEs are the backbone of medical devices, enabling precise signal acquisition and processing, which is essential for accurate diagnostics and effective treatments. Our AFEs ensure high-resolution signal capture, which is vital to detecting subtle physiological changes. They are also designed to operate with minimal power, extending the battery life of implantable devices. We focus on creating compact AFEs that fit seamlessly into small, implantable devices, reducing patient discomfort and improving integration.

For invasive applications, we have designed AFEs for s-EEG systems that deliver real-time data with remarkable precision, supporting critical decision-making in the surgical treatment of challenging epileptic seizures. For implant applications, we are working on AFEs for neural implants that facilitate advanced brain-machine interfaces, opening new possibilities for treating neurological disorders. At CSEM, our commitment to innovation in analog front ends is driving the future of medical technology, ensuring better patient outcomes and advancing healthcare solutions.

Silicon proven chip implementation of AFEs in ASICS for Epilepsy treatment and subsequent thermal ablation

This compact integrated chip, measuring 7.2 mm x 0,65 mm, is designed for high-precision invasive epilepsy treatment. It features an 18-channel Analog Front End (AFE) with Low Noise Amplifiers (LNA), Programmable Gain Amplifiers (PGA), and ADC drivers for precise analog-to-digital conversion.

The chip includes an ablation switch bank for controlling electrical signals during tissue ablation, 18 high-voltage switches for managing high-voltage signals, and AFE isolation to prevent interference and ensure signal integrity.

A digital controller coordinates the operations of various components, while the Power Management Unit (PMU) regulates power distribution. High-voltage clamps (1.8V/0V to 16V/-16V) protect against voltage spikes, and low-side AFE decoupling capacitors stabilize the power supply, reducing noise.

Additionally, the chip integrates oscillators to provide precise clock signals for timing and synchronization

CSEM's needle-like ASIC for stereo-electroencephalography (S-EEG)
© CSEM – Needle-like ASIC for stereo-electroencephalography (S-EEG)

AFEs for ultra-low power, high-resolution audio

CSEM’s audio codecs target a wide range of applications, covering the following key features:

  • Keyword spotting: wake-on noise, ultra-low power on systems, etc.
  • Voice band: hearing aids, speech processing, low latency, etc.
  • Professional audio: high-resolution and performance audio with an effective number of bits (ENOB) > 16.

A simplified block diagram of an audio transceiver includes receive and playback paths

  • The receive path features, on the one hand, a microphone bias (Micbias) block that biases the microphone appropriately and, on the other, a programmable gain amplifier for amplifying weak input signals before being digitized.
  • The programmable gain amplifier (PGA) is designed to function even with passive microphones thanks to its large dynamic range and programmability of the amplification factor.
  • The analog-to-digital converter (ADC) is chosen from our data converter portfolio of oversampled ADCs for high-performance audio and voice applications.
Audio receive and playback paths

The playback path supports wide PCM (Pulse Coded Modulated) digital words or PWM (Pulse Width Modulated) coded data as inputs. The digital inputs are converted into a suitable format without compromising the resolution or bandwidth by an oversampled digital-to-digital converter before being converted into analog. A filterless Class-D or Class-AB amplifier driver, depending on application requirements, is used to drive the speaker load efficiently.

A low-power energy measurement block that computes the energy in a predetermined frequency band is also available to detect keywords and other interesting audio signatures in order to either trigger an external system or to wake up/power down the audio codec.

Electrochemical front-ends for biosensing and environmental monitoring

CSEM powerful electrochemical analog front-end ASICs offer highly integrated solutions for electrochemical analysis. These ASICs support several electroanalytical methods to analyze a wide range of analytes, including pH measurements.

Key features:

  • Electrochemical methods: Chronoamperometry, cyclic voltammetry, open circuit potentiometry, linear sweep voltammetry, and more.
  • Analytes: Ions (Na+, K+, Cl-), enzymes (lactase), and metabolites (glucose, lactate).

A simplified block diagram of CSEM’s electrochemical front-end solution includes the following:

  • The reference and working electrodes are accurately biased by digital-to-analog (DAC) converters.
  • Meanwhile, the counter electrode is efficiently powered by a potentiostat amplifier.

The resulting current from the working electrode is then digitized using a high-performance and low-power analog-to-digital (ADC) converter. All the voltage and current references are internally generated. Additionally, an on-chip clock generator is designed to precisely clock both the digital and the on-chip data converters.

Block diagram of the electrochemical front end
Analog Front-End (AFE) chip for electrochemical sensing

How can we assist you with advanced AFE solutions?

A custom high-performance analog front end can be a differentiator for your sensory system.

Articles and references