Linear-Phase Delay Filters for Ultra-Low-Power Signal Processing in Neural Recording Implants

Gosselin, Benoît; Sawan, Mohamad; Kerhervé, Eric

doi:10.1109/tbcas.2010.2045756

Cited by 43 publications

(26 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this method, only the APs are detected and transferred out of the body. Since the duty cycle of this method is in the range of 2% to 20%, the data can be compressed by a maximum factor of 50 [28,30]. This method is applicable in both digital and analog domains.…”

Section: Neural Recording Architecturesmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Channel Neural Recording Implants: A Review

Noshahr

Nabavi

Sawan

2020

Sensors

Self Cite

View full text Add to dashboard Cite

The recently growing progress in neuroscience research and relevant achievements, as well as advancements in the fabrication process, have increased the demand for neural interfacing systems. Brain-machine interfaces (BMIs) have been revealed to be a promising method for the diagnosis and treatment of neurological disorders and the restoration of sensory and motor function. Neural recording implants, as a part of BMI, are capable of capturing brain signals, and amplifying, digitizing, and transferring them outside of the body with a transmitter. The main challenges of designing such implants are minimizing power consumption and the silicon area. In this paper, multi-channel neural recording implants are surveyed. After presenting various neural-signal features, we investigate main available neural recording circuit and system architectures. The fundamental blocks of available architectures, such as neural amplifiers, analog to digital converters (ADCs) and compression blocks, are explored. We cover the various topologies of neural amplifiers, provide a comparison, and probe their design challenges. To achieve a relatively high SNR at the output of the neural amplifier, noise reduction techniques are discussed. Also, to transfer neural signals outside of the body, they are digitized using data converters, then in most cases, the data compression is applied to mitigate power consumption. We present the various dedicated ADC structures, as well as an overview of main data compression methods.Sensors 2020, 20, 904 2 of 29 an invasive method, records proper signals. However, the disadvantage of this is not being able to record deep in the brain and they average the activity of thousands of neurons [11]. Invasive techniques have been utilized by some BMIs. This is because recording single action potentials from neurons in distributed, functionally-linked ensembles are necessary for the most accurate readout of neural activities [7]. Therefore, increasing the spatial resolution and the number of electrodes are essential for developing BMI.The implementation of a neural recording implant is multi-disciplinary, as it involves the various scientific fields such as electronics, medical, materials, electrodes, and system integration. Increasing the number of electrodes and, consequently, the number of channels (in the range of thousands), creates new challenges for neural recording in the various fields mentioned. Microelectrode technology is not appropriate for these large-scale recordings [12]. Recently, Neuralink Company has built arrays of small and flexible electrodes (3072 electrodes per array), which have enabled thousands of channel recordings [13]. In the microelectronics field, large-scale recordings create many challenges with regards to decreasing the power consumption and chip area.In the design of the neural recording implants, the two constraints, the power consumption and chip area, should be addressed. Implantable circuits should consume very low power to avoid any damage to the surrounding tissue due to gen...

show abstract

Section: Neural Recording Architecturesmentioning

confidence: 99%

“…Another signal compression method, which is suitable for extracellular recordings, is presented in the literature [28,30]. This compression method is based on the sparsity of APs in the time domain.…”

Section: Data Compressionmentioning

confidence: 99%

Multi-Channel Neural Recording Implants: A Review

Noshahr

Nabavi

Sawan

2020

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Benchmarking with the works [14], [15], [25], [26] targeting a similar bandwidth in Table IV, both the GC and non-GC LPFs exhibit superior performances in terms of the Figure-of-Merit (FoM) defined as [27] (10) is the filter power consumption, is the order, is the cutoff frequency (exception for the center frequency in [25]), and is the dynamic range. Lower FOM figures indicate better filter performance.…”

Section: Experimental Verificationmentioning

confidence: 99%

15-nW Biopotential LPFs in 0.35-<formula formulatype="inline"> <tex Notation="TeX">$\mu{\rm m}$</tex></formula> CMOS Using Subthreshold-Source-Follower Biquads With and Without Gain Compensation

Zhang

Vai

Mak

et al. 2013

IEEE Trans. Biomed. Circuits Syst.

View full text Add to dashboard Cite

Most biopotential readout front-ends rely on the g m- C lowpass filter (LPF) for forefront signal conditioning. A small g m realizes a large time constant ( τ = C / g m) suitable for ultra-low-cutoff filtering, saving both power and area. Yet, the noise and linearity can be compromised, given that each g m cell can involve one or several noisy and nonlinear V- I conversions originated from the active devices. This paper proposes the subthreshold-source-follower (SSF) Biquad as a prospective alternative. It features: 1) a very small number of active devices reducing the noise and nonlinearity footsteps; 2) No explicit feedback in differential implementation, and 3) extension of filter order by cascading. This paper presents an in-depth treatment of SSF Biquad in the nW-power regime, analyzing its power and area tradeoffs with gain, linearity and noise. A gain-compensation (GC) scheme addressing the gain-loss problem of NMOS-based SSF Biquad due to the body effect is also proposed. Two 100-Hz 4th-order Butterworth LPFs using the SSF Biquads with and without GC were fabricated in 0.35- μm CMOS. Measurement results show that the non-GC (GC) LPF can achieve a DC gain of -3.7 dB (0 dB), an input-referred noise of 36 μV rms (29 μV rms ), a HD3@60 Hz of -55.2 dB ( - 60.7 dB) and a die size of 0.11 mm² (0.08 mm²). Both LPFs draw 15 nW at 3 V. The achieved figure-of-merits (FoMs) are favorably comparable with the state-of-the-art.

show abstract

“…LFP occupy frequency band ranging from a few mHz up to around 300 Hz. LFP have higher amplitudes than SPK and can be as large as 5 mV [4]. Most data reduction schemes aim to detect and extract occurrences of SPK which correspond to a fraction of the raw data [3].…”

Section: Introductionmentioning

confidence: 99%

“…Most data reduction schemes aim to detect and extract occurrences of SPK which correspond to a fraction of the raw data [3]. SPK can also be classified and sorted onchip to achieve very high data reduction ratios by only transmitting time stamps [4]. Applying a data reduction scheme requires to separate the recorded broadband neural signals into a SPK band and a LFP band prior to signal detection and extraction.…”

Section: Introductionmentioning

confidence: 99%

A low-power current-reuse dual-band analog front-end for multi-channel neural signal recording

Sepehrian

Gosselin

2014

2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

View full text Add to dashboard Cite

Thoroughly studying the brain activity of freely moving subjects requires miniature data acquisition systems to measure and wirelessly transmit neural signals in real time. In this application, it is mandatory to simultaneously record the bioelectrical activity of a large number of neurons to gain a better knowledge of brain functions. However, due to limitations in transferring the entire raw data to a remote base station, employing dedicated data reduction techniques to extract the relevant part of neural signals is critical to decrease the amount of data to transfer. In this work, we present a new dual-band neural amplifier to separate the neuronal spike signals (SPK) and the local field potential (LFP) simultaneously in the analog domain, immediately after the pre-amplification stage. By separating these two bands right after the pre-amplification stage, it is possible to process LFP and SPK separately. As a result, the required dynamic range of the entire channel, which is determined by the signal-to-noise ratio of the SPK signal of larger bandwidth, can be relaxed. In this design, a new current-reuse low-power low-noise amplifier and a new dual-band filter that separates SPK and LFP while saving capacitors and pseudo resistors. A four-channel dual-band (SPK, LFP) analog front-end capable of simultaneously separating SPK and LFP is implemented in a TSMC 0.18 μm technology. Simulation results present a total power consumption per channel of 3.1 μw for an input referred noise of 3.28 μV and a NEF for 2.07. The cutoff frequency of the LFP band is fc=280 Hz, and fL=725 Hz and fL=11.2 KHz for SPK, with 36 dB gain for LFP band 46 dB gain for SPK band.

show abstract

Linear-Phase Delay Filters for Ultra-Low-Power Signal Processing in Neural Recording Implants

Cited by 43 publications

References 39 publications

Multi-Channel Neural Recording Implants: A Review

Multi-Channel Neural Recording Implants: A Review

15-nW Biopotential LPFs in 0.35-<formula formulatype="inline"> <tex Notation="TeX">$\mu{\rm m}$</tex></formula> CMOS Using Subthreshold-Source-Follower Biquads With and Without Gain Compensation

A low-power current-reuse dual-band analog front-end for multi-channel neural signal recording

Contact Info

Product

Resources

About