Impulse Radio Ultra Wideband Wireless Transmission of Dopamine Concentration Levels Recorded by Fast-Scan Cyclic Voltammetry Ali Ebrazeh, Student Member, IEEE, Bardia Bozorgzadeh, Student Member, IEEE, and Pedram Mohseni, Senior Member, IEEE 

Abstract—This paper demonstrates the feasibility of utilizing impulse radio ultra wideband (IR-UWB) signaling technique for reliable, wireless transmission of dopamine concentration levels recorded by fast-scan cyclic voltammetry (FSCV) at a carbon-fiber microelectrode (CFM) to address the problem of elevated data rates in high-channel-count neurochemical monitoring. Utilizing an FSCV-sensing chip fabricated in AMS 0.35µm 2P/4M CMOS, a 3–5-GHz, IR-UWB transceiver (TRX) chip fabricated in TSMC 90nm 1P/9M RF CMOS, and two off-chip, miniature, UWB antennae, wireless transfer of pseudo-random binary sequence (PRBS) data at 50Mbps over a distance of <1m is first shown with bit-error rates (BER) < 10-3. Further, IR-UWB wireless transmission of dopamine concentration levels prerecorded with FSCV at a CFM during flow injection analysis (FIA) is also demonstrated with transmitter (TX) power dissipation of only ~4.4µW from 1.2V, representing two orders of magnitude reduction in TX power consumption compared to that of a conventional frequency-shift-keyed (FSK) link operating at ~433MHz.

I. INTRODUCTION Fast-scan cyclic voltammetry (FSCV) at a carbon-fiber microelectrode (CFM) is recognized as the preferred choice for real-time monitoring of endogenous neurotransmitters in behaving animals due to its exquisite temporal, spatial, and chemical resolution [1]. Indeed, this measurement modality provided the first monitoring of a behaviorally associated change in neurotransmitter levels with subsecond temporal resolution at a brain-implanted, micron-sized probe in an awake animal. Recently, advances in microfabrication and micro-electro-mechanical-system (MEMS) technologies have brought about high-site-density microelectrode arrays, enabling distributed recording of neurotransmitters in different brain regions concurrently [2]. However, in comparison, technological advances in neurochemicalrecording instruments for high-channel-count measurements in behaving subjects have considerably lagged behind. Since physical restraint of the subject via traditional tethering to monitoring equipment is a concern for altering behavior and a noise source, as well as limits the use of important experimental paradigms involving enriched environments and long-term recording, wireless solutions have been developed to address these problems. However, This work was supported by the National Science Foundation (NSF) CAREER Award DBI-0844957, National Institute on Biomedical Imaging and Bioengineering (NIBIB) Award EB-014539, and National Institute on Drug Abuse (NIDA) Award DA036331. A. Ebrazeh, B. Bozorgzadeh, and P. Mohseni are with the Electrical Engineering and Computer Science Department, Case Western Reserve University, Cleveland, OH 44106 USA (email: [email protected]).

978-1-4244-9270-1/15/$31.00 ©2015 IEEE

for neurochemical measurements with high temporal and spatial resolution, it is exceedingly difficult to support the resulting high data rates for wireless transmission within a reasonable power budget, especially if such measurements are conducted via conventional wireless links. Garris et al. have previously reported a backpack device for wireless transmission of FSCV at a CFM that comprises commercial off-the-shelf (COTS) components and employs Bluetooth digital telemetry for bidirectional communication between the backpack device and a home-base unit [3]. The device has dimensions of 4 × 2 × 1.5cm3, weighs 17g, and supports one hour of single-channel recording. Roham et al. have reported a miniaturized device for wireless monitoring of extracellular dopamine levels in the brain of an ambulatory rat using FSCV at a CFM [4]. This single-channel device comprises integrated FSCV-sensing circuitry, weighs 2.3g including a single battery (3V), and employs a conventional frequency-shift-keyed (FSK) transmitter (TX) operating at ~433MHz with a power dissipation of ~1mW for a transmission range of 1m. It is worth noting that Roham’s device would have had a data rate of ~23Mbps prior to any encoding, if the number of channels had been increased to 32, necessitating innovative solutions for simultaneous, multichannel, wireless operation. To address the problem of data rate in high-channelcount neurochemical monitoring, this paper demonstrates proof-of-concept of utilizing impulse radio ultra wideband (IR-UWB) signaling technique for wireless transmission of dopamine concentration levels recorded by FSCV at a CFM, and compares the results with those obtained by using a conventional FSK TX as the uplink. II. FSCV SYSTEM DESIGN REQUIREMENTS FSCV at a CFM for dopamine recording is shown in Fig. 1. The CFM potential is linearly swept every 100ms from -0.4V to 1.3V at 400V/s, resulting in scan duration of 8.5ms. In the positive sweep, dopamine (DA) oxidizes to dopamine-ortho-quinone (DOQ), which is reduced back to dopamine in the negative sweep. The total current from this electrochemical reaction includes both background and faradaic currents (Panel A). The latter, which is proportional to dopamine concentration, is obtained by subtracting prerecorded background current from the total current (Panel B). Background-subtracted faradaic current plotted versus CFM potential creates the background-subtracted cyclic voltammogram, which serves as a chemical signature to identify the analyte (Panel C). Dynamic information is also obtained by plotting peak faradaic current measured at the

7103

dopamine oxidative potential in each voltammogram versus time in successive FSCV scans (Panel D). Phasic dopamine transients vary in a range of 40nM to 1µM [1]. Further, typical FSCV systems in neurochemistry exhibit dopamine sensitivity of 10–50nA/µM, depending on the voltage limits, repetition frequency, and sweep rate of the FSCV scan. Hence, the total input noise current of the FSCV system should be <100pArms for a minimum signal-to-noise ratio (SNR) of four in chemical detection. This input noise current level should be achieved within a large dynamic range of, e.g., 900nA to accommodate the ever-present large background current, which depends on the FSCV scan parameters and total exposed surface area of the CFM, requiring current resolution of >12b for FSCV recording. Since phasic dopamine transients also vary on a timescale of subsecond to seconds, the input current signals of an FSCV system, resulting from electrochemical transduction, have low frequency components. Hence, oversampling deltasigma modulators (M) have been the preferred choice for an FSCV-recording front-end [5], as they strike a tradeoff between conversion speed and resolution. Further, targeting an input signal bandwidth of around 5kHz, we have found that a 3rd-order ΔΣM with sampling frequency of ~600kHz offers the most favorable tradeoff among sampling frequency, power consumption, and system complexity due to stability concerns with higher-order designs.

Fig. 1. Fundamentals of FSCV at a CFM for dopamine recording.

III. IR-UWB WIRELESS COMMUNICATIONS FSCV system design requirements, as discussed in Section II, have presented a bottleneck for simultaneous, wireless, high-channel-count FSCV monitoring due to elevated data rates that would emerge under such a scenario. For example, with a 3rd-order ΔΣM with 1b quantization, the serial output bit-stream would have raw data rates ranging from ~ 0.6Mbps (1 ch.) to 19.2Mbps (32 ch.), if decimation and filtering were to be performed in software at the external receiver (RX) side for implementation flexibility. In fact, data rate would further increase due to encoding (e.g., 2 for Manchester encoding) prior to wireless transmission, and can become prohibitively high for commercial transceiver electronics and traditional communication protocols. Fig. 2 compares the data rate versus transmission range for several wireless communication technologies. As can be seen, for short-range communication (<10m) at elevated data rates (10’s–100’s of Mbps), UWB would be a more suitable option compared to other strategies such as Zigbee or Bluetooth. In particular, as opposed to more established multiband orthogonal frequency-division multiplexing (OFDM) UWB, IR-UWB signaling techniques offer the added value of much simpler TX architectures that readily lend themselves to low-power, small-area, silicon implementation, which is especially advantageous for biomedical applications [7]. IR-UWB systems use baseband, short-duration (ps to ns) pulses with a low duty cycle that have correspondingly wide spectra to enable high-data-rate communications [8].

Fig. 2. Comparison of data rate versus transmission range for several wireless standards available for biomedical applications (adapted from [6]).

A common feature of high-performance IR-UWB systems is their ability to have large control over the pulse power spectral density (PSD), or equivalently, its shape in the time domain. We have developed an all-digital IR-UWB TX in 90nm RF CMOS based on a waveform synthesis technique, which affords great flexibility in reconfiguring the shape of the UWB pulse waveform in time domain and its corresponding PSD in frequency domain [9]. Fig. 3 depicts a 10-lobe and a 6-lobe UWB pulse with Gaussian envelope generated directly in the time domain by our all-digital TX along with their corresponding FCC-compliant PSD. IV. EXPERIMENTAL SETUP To showcase the utility of IR-UWB signaling technique in an experimental recording paradigm, dopamine concentration levels obtained in flow injection analysis (FIA) were wirelessly recorded with both an IR-UWB transceiver (TRX) operating in 3–5GHz and an FSK TRX operating at ~433MHz. Fig. 4 depicts the simplified architecture of the two CMOS-integrated systems forming the experimental setup along with their corresponding die micrographs. The FSCV-sensing chip, fabricated in AMS 0.35µm 2P/4M CMOS [5], incorporates an oversampling M with 1b

7104

quantization and clocked at 625kHz, an FSCV waveform generator, a clock generator from external reference, and a wireless FSK TX. The M features a power-saving mode that reduces the sensing power dissipation to 9.3µW at 2.5V via duty-cycling the M operation. The serial data at the M output is sent out at ~433MHz with the on-chip FSK TX that dissipates 405µW from 2.5V for received signal power of -52dBm by the FSK RX placed at a distance of 1m. Hence, the total power dissipation of the FSCV-sensing chip is heavily dominated by that of the FSK TX. The IR-UWB TRX, fabricated in TSMC 90nm 1P/9M RF CMOS [9], operates in three channels within 3–5GHz. The all-digital TX integrates a highly reconfigurable pulse generator and a timing generator for pulse modulation and phase scrambling, and generates FCC-compliant UWB pulses directly in the time domain without using a mixer/local oscillator for pulse upconversion or an off-chip filter for pulse shaping. The high flexibility in digitally reconfiguring the temporal waveform of the UWB pulse affords a channelized operation, with PSD center frequencies of the channels set to ~3.4, 4.1, and 4.65GHz. The RX utilizes a non-coherent, energy-detecting architecture for low-power operation and low system complexity. It employs a low-noise amplifier (LNA) and two RF amplifiers, followed by a passive self-mixer for detection of the received signal energy, a comparator for digital quantization, and back-end digital circuitry for baseband clock and data synchronization [9]. The RF front-end provides up to 37dB of adjustable gain and is designed with programmable center frequency and bandwidth to tune the RX to 3.5, 4, and 4.5GHz for channel selection and robustness against the out-of-band noise/interference.

Fig. 3. Two UWB pulses with Gaussian envelope and their corresponding FCC-compliant PSD.

Fig. 4. Architecture of the FSCV-sensing and IR-UWB TRX chips for wireless recording of dopamine concentration levels. Die micrographs of the two integrated systems are also shown.

V. MEASUREMENT RESULTS A. IR-UWB Wireless Transmission of PRBS Data The objective of this test was to demonstrate the IR-UWB TRX utility for robust, high-data-rate telemetry. To that end, a TX and RX on two separate die were interfaced with two identical off-chip antennae. Specifically, an UWB chip antenna (ANT1085, TDK) with an operation frequency range of 3.1–5.2GHz and a typical gain of 2dBi was used. Featuring a foot-print of only 10 × 8.5 × 1.2mm3, this small antenna can yield a TRX system with small form-factor that is ultimately suitable for implantation or wearability. A clock signal of 50MHz and 16b pseudo-random binary sequence (PRBS) data were used to OOK-modulate the TX with one UWB pulse per each data bit of “1”, effectively achieving a data rate of 50Mbps. The TX was programmed to generate ~350-mVpp UWB pulses with PSD center frequency of ~3.4GHz, consuming 14pJ/pulse from 1.2V. Fig. 5 shows wireless transfer of the 16b, 50-Mbps PRBS data over a distance of <1m. The top plot shows the 16b PRBS data, the middle plot shows the OOK-modulated IR-UWB TX output, and the bottom plot shows the wirelessly received PRBS data at the IR-UWB RX output. The measured bit-error rate (BER) was <10-8 at 0.5m and <7  10-5 at 1m, showing reliable, high-data-rate telemetry.

Fig. 5. IR-UWB wireless transmission of 16b, 50-Mbps PRBS data at <1m. Top and bottom plots show the transmitted and received PRBS data, respectively. Middle plot depicts the UWB pulses at the TX output after OOK modulation, with one UWB pulse representing each data bit of “1”.

B. IR-UWB Wireless Transmission of FSCV Data For this test, the FSCV-sensing chip was interfaced with a CFM working electrode (WE) positioned in the inlet of a flow cell reservoir, with a chlorided silver wire (Ag/AgCl) placed at the bottom of the buffer-filled reservoir as the reference electrode (RE). All measurements were collected in buffer containing 150mM NaCl and 15mM TRIS (pH = 7.4). A 5-second, 1-µM bolus of dopamine was injected to the flowing stream entering the reservoir inlet via a loop injector driven by a pneumatic actuator. The FSCV-sensing chip

7105

applied the FSCV waveform (-0.4–1.3V, 400V/s, 10Hz) to the CFM WE and digitized the total current from the electrochemical transduction at 625kHz with 1b quantization, resulting in a 625-kbps serial data bit-stream. Fig. 6 shows the IR-UWB wireless transfer of a 16b-portion of this bitstream. The order of the plots is identical to that in Fig. 5, but note that the time axis is in microseconds, as opposed to nanoseconds in Fig. 5. Fig. 7 shows wirelessly measured background-subtracted cyclic voltammogram of dopamine. The peak current at nearly 670mV during the forward sweep corresponds to dopamine oxidation to its quinone, whereas the peak current at nearly -280mV in the reverse sweep corresponds to reduction of the electroformed quinone back to dopamine. Finally, Fig. 8 shows the dynamic plot (i.e., backgroundsubtracted dopamine current versus time) measured by the two wireless links after 15 seconds of 400-V/s, 10-Hz FSCV. The rise and fall time instances correspond to when bolus injection into the flow cell was turned ON and OFF, respectively. The time offset was due to an inherent lag with FIA, as the analyte was injected distal to its measurement. As seen in Figs. 7 and 8, the measured results with the two wireless links were in excellent agreement. While the FSK TX dissipated ~400µW, the IR-UWB TX dissipated only ~4.4µW (for a balanced serial data bit-stream with equal numbers of “1” and “0”), significantly reducing the transmission power dissipation. VI. CONCLUSION For neurochemical measurements with high temporal, spatial, and chemical resolution, FSCV at a CFM with an oversampling M front-end has been the preferred choice. However, due to oversampling in such FSCV systems and the resulting elevated data rates, wireless, high-channel-count neurochemical monitoring within a reasonable power budget is becoming exceedingly more difficult. This paper demonstrates proof-of-concept of utilizing an IR-UWB TRX for wireless transmission of dopamine concentration levels, and compares the results with those obtained by using a conventional FSK link. By virtue of providing a much higher data rate than a conventional FSK TX at significantly lower power dissipation, the IR-UWB TRX presents an excellent solution for wireless, high-channel-count neurochemical sensing with FSCV at a CFM. REFERENCES [1] [2]

[3] [4]

D. L. Robinson, et al., “Monitoring rapid chemical communication in the brain,” Chem. Rev., vol. 108, pp. 2554-2584, 2008. Y. Song, et al., “The dual-mode microelectrode arrays and integrated detection system for neuroelectrical and neurochemical detection,” in Proc. IEEE-EMBS Int. Conf. Biomed. Health Informatics, pp. 228-231, Hong Kong and Shenzhen, China, January 2-7, 2012. P. A. Garris, et al., “Wireless transmission of fast-scan cyclic voltammetry at a carbon-fiber microelectrode: proof of principle,” J. Neurosci. Meth., vol. 140, pp. 103-115, 2004. M. Roham, et al., “A miniaturized device for wireless FSCV monitoring of dopamine in an ambulatory subject,” in Proc. 32nd Annu. Int. IEEE Eng. Med. Biol. Conf. (EMBC’10), pp. 5322-5325, Buenos Aires, Argentina, August 31-September 4, 2010.

Fig. 6. IR-UWB wireless transmission of a 16b-portion of the 625-kbps serial data bit stream at the output of the FSCV-sensing front-end. Plot order is identical to that in Fig. 5.

Fig. 7. Wireless recording of background-subtracted cyclic voltammogram of dopamine using a conventional FSK link and the IR-UWB wireless link.

Fig. 8. Wireless recording of background-subtracted, 1-µM-dopamine current after 15 seconds of 400-V/s, 10-Hz FSCV. Data have been recorded using a conventional FSK link and the IR-UWB wireless link. [5]

[6] [7] [8] [9]

7106

B. Bozorgzadeh, et al., “A neurochemical pattern generator SoC with switched-electrode management for single-chip electrical stimulation and 9.3µW, 78pArms, 400V/s FSCV sensing,” IEEE J. Solid-State Circuits, vol. 49, no. 4, pp. 881-895, April 2014. Wireless Connectivity for Medical Applications. Source: Texas Instruments, Inc. A. Ebrazeh and P. Mohseni, “30pJ/b, 67Mbps, centimeter-to-meter range data telemetry with an IR-UWB wireless link,” IEEE Trans. Biomed. Circuits and Systems, vol. 9, no. 3, pp. 362-369, June 2015. F. Nekoogar, Ultra-Wideband Communications: Fundamentals and Applications. NJ, USA: Prentice Hall Press, 2005. A. Ebrazeh and P. Mohseni, “A 14pJ/pulse-TX, 0.18nJ/b-RX, 100Mbps, channelized, IR-UWB transceiver for centimeter-to-meter range biotelemetry,” in Proc. IEEE Custom Integr. Circ. Conf. (CICC), San Jose, CA, September 15-17, 2014.