Abstract
This paper presents two low power design techniques used for successive approximation registers (SAR) analog-to-digital converter (ADC) for transmission of Physiological signal: Dual split switching; set and reset phase. Dual split switching is used in one sided charge scaling digital-to-analog converter (DAC) to edge of the switching energy by reducing the leakage in dual transmission gate. The set and reset phase defines the amplification and comparison phase of the comparator. The delay time of the comparator is profoundly reduced with folded cascoded pre amplifier and a regenerative latch. A Serial In Parallel Out (SIPO) N bit register and SAR are designed with negative edge triggered D Flip-Flop (DFF). For power optimization the supply voltage of SAR ADC is designed with 500 mV. The Variable threshold concept has been utilized in the entire design to operate the SOC with 500 mV supply voltage. The designed SAR ADC is capable of supporting the sampling rate of 1 Msps. The circuit is designed using standard UMC 180 nm technology. The simulation results show that the power consumption of the SAR ADC is 13.99 μW and achieved 68.54 dB SFDR with ENOB value 7.69 bits. The DNL (max) is + 0.9/− 0.82 LSB and INL 1.06/− 1.31 LSB.
Similar content being viewed by others
1 Introduction
In the recent years the development of biomedical applications in wireless communication has become very popular [1, 2]. The research on Electrocardiogram (ECG) and Electroencephalogram (EEG) acquisition boards are giving impeccable results in the biomedical applications [1, 2]. The role of bio signal acquisition board is to collect bio signals from the human body which are weak in amplitude and frequency and transmit in digital form after amplification. So, for continuous transmission of physiological signals, a battery life is required to prolong the work time and the device size should get reduced to be implanted on the human body. Most importantly, high data accuracy should be maintained [3,4,5,6,7,8,9,10]. While going towards advanced CMOS processes, the benefits are: feature size is edged off and operates with lower threshold voltages, but leakage currents will play dominant role in the power consumption. Therefore, besides accommodating a small supply voltage, we need to minimize the leakage power to achieve better power efficiency. While transmitting the physiological signals through wireless, high data accuracy should be required because the neural signals are weak in amplitude (< 100 μV) and weak in frequencies (few Hz to below 100 Hz) [7, 8]. Physiological signals are the pure analog signals which are captured from the human body with transducers. The successive approximation register (SAR) type analog-to-digital converter (ADC) is an essential part in the analog signal processing [11,12,13,14,15,16,17,18]. SAR ADCs are available from 8 to 18 bits resolution with sampling rates up to 50 Msps [11, 12]. So SAR ADC is well suited when compared to other ADC architectures for transmission of physiological signals [7, 8] as shown in the Fig. 1. The charge redistribution DAC with charge scaling single ended spilt capacitor array is designed with small capacitance values so that switching speed increases and area is reduced. DAC operates in two modes, one is sampling phase and second is conversion phase by additional controlling as proposed in references [14], This paper also proposes a novel comparator which is designed to operate in the sub threshold region to have very low power consumption. An adjustable body bias is arranged in the regenerative latch phase to operate the transistor in weak inversion region so that leakage current is reduced for 500 mV Supply voltage. Few optimization technologies are adopted in the ADC blocks to reduce power consumption [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30]. To implement the low power multi-channel signal processing system, a 0.5 V bulk-driven folded-cascoded operational trans conductance amplifier is proposed in references [31].The digital data from SAR ADC is stored in the proposed design of low power 16 × 16 SRAM array in references [32].
This paper is organized as follows: In Sect. 2, the proposed SAR ADC is described. Section 3 describes the detailed circuit design of the ADC’s blocks. Low power optimized techniques discussed in Sect. 4 The measurement results and comparison with previous works are shown in Sect. 5 followed by conclusion.
2 SAR architecture
As shown in Fig. 1 the architecture of SAR ADC resembles the general ADC architecture only, but the essential blocks which are used in SAR ADC is designed with low power optimized techniques. Such as low power Comparator and single ended switched capacitor and the SAR control logic, which controls the DAC switches based on the comparator output. In the overall power consumption of the system, Comparator and DAC plays the major role because of switching between Vref and Vin, so we opted dual transmission gate based switching method instead of static CMOS inverter. The additional privilege in the proposed design is Variable threshold CMOS technique (VTCMOS). It is adopted for operating transistor under sub threshold region while minimizing leakage current. The working of the ADC is as shown in the Fig. 2 the complete operation of the ADC divided into three stages. The first stage is sampling stage, Second stage is conversion stage and finally recycling stage. During the initial stage the bottom plates of all the capacitors are connected to ground. The upper plates of the capacitors are connected to the inverting terminal of the comparator. Non inverting input of the comparator is connected to the sampled output. Based on the comparator output and control signals of switches, sampled signal is converted into digital form during second and third stages. The input signal is sampled by a sample and hold circuit which was explained in the Sect. 3.2. A detailed design process of all the blocks are explained in the next sections.
3 Design details of SAR ADC
3.1 Comparator
The low power regenerative comparator circuit is the essential block in the ADC and many electronic devices. The main goal is to design a high speed and low power preamplifier regenerative comparator circuit [28, 29]. Figure 3 shows the schematic of proposed preamplifier regenerative comparator circuit. The operation of the comparator is divided into pre-amp phase (reset) stage and dynamic comparison phase (set) stage with additional low on resistance switches S1 and S2 with non-overlapping clock signals as shown in Fig. 3. During the reset phase switch S1 will be open and S2 will be closed then the single stage folded amplifier, amplifies the voltage difference between the two input signals with voltage gain 40.4 dB (1). During the set phase S1 will close and S2 opens during the time regenerative comparator circuit compares between two input signals during the opted sampling intervals. Because of this set and reset phases the nonlinear error and offset error at comparator are reduced during analog to digital conversion. When \(V_{in} > V_{ref}\) the output signal exhibits logic high means level 1 and \(V_{in} < V_{ref}\) the output signal exhibits logic low means level 0.
The preamplifier is a single stage operational trans conductance amplifier (OTA) [32] as shown in Fig. 3. The current scaling technique is implemented by the transistors M5–M6 with higher widths. Transistors Mb1, Mb2, Mb3 form a current mirror which delivers the differential currents. These differential currents are feed to output stage, for voltage to current conversion. Transistors M1 to M6 are operated in weak inversion region with appropriate W/L ratio. The drain to source current IDS (2) in weak inversion region depends on reverse saturation current IS, T is the ambient temperature, n is the inclination of the curve in weak inversion, K is the Boltzmann constant, q the charge of the electron or hole. The total power consumption of the comparator is 1.8 μW with 500 mV rail to rail supply voltage.
Current scaling is achieved by increasing the output impedances and thereby ensuring current scaling. The source degenerated current mirrors are formed by transistors Mc1, M5 and M6, set the currents in the regenerative circuit (the difference between the currents in M3–M4 and M5–M6). In order to save the power, the bias circuit is designed in such a way that the currents of transistors in the regenerative circuit are only a small portion of the input pair. The current scaling ratio between Mb1 and Mb2 is 7:1(2ID/14). The currents in the regenerative transistors are ID/7; this is 1/3rd of the differential input pair current. Using the bias circuit formed by Mb2 and Mc1, we set the current in the M5 and M6 to be 8ID/7 [31] (Table 1).
The Transmission gate based switch is used for switching between set and reset operations, but to avoid glitches during switching and without voltage drops while passing through the channel, on-resistance should be reduced. The parallel combination of one low and high on-resistance of the transistors results low on-resistance. So on-resistance of the transmission gate is reduced while connecting one low W: L ratio pMOS transistor with high W:L ratio pMOS transistor and similarly for nMOS transistor.
3.2 Sample and hold
A simple sample and hold circuit is designed with CMOS Transmission gate and a sampling capacitor. To avoid charge injection error, a dummy switch is arranged as shown in the Fig. 4. The dummy switch is driven by an inverted clock which absorbs the charge injection from sampling switch. The unwanted glitches are also eliminated by using this dummy switch. The minimization of DNL and INL for better ADC design can be possible by connecting a buffer at the end of the sample and hold circuit.
3.3 Digital to analog converter (DAC)
A 10 bit charge scaling capacitive DAC is designed with a combination of two 5 bit charge scaling sub-DACs with a scaling capacitor \(C_{s}\) as shown in the Fig. 5. The operation of the DAC with example of five levels of shifting is explained with the help of the Fig. 6. The series combination of scaling capacitor starts with LSB array and terminates with MSB array. The accuracy of the DAC depends upon the scaling capacitor \(C_{s}\) as it is the terminating capacitor between LSB and MSB array.
The area of the DAC is proportional to the size of the unit capacitor. From the simulation test results the unit capacitance whose value is 62.5 fF. The operational amplifier which is discussed in the previous section is utilized in the DAC as a buffer to increase the linearity and to decrease comparator offset error [31].Total capacitance on the right of the scaling capacitor 1.937 pF and left of the scaling capacitor is 2 pF. DAC plays almost 40% in the total power consumption of ADC because dynamic power consumption and switching event capacitances, charge up phase dissipates more heat in the circuit. Dual split switching minimizes the switching energy by reducing on resistance of the transmission gate that could be explained in later sections. The power analyses of all the modules are as shown in the Fig. 7.
The high performance of the DAC is defined from the calculations of differential nonlinearity (DNL) and integral nonlinearity (INL). In this paper, DAC design is carried out for low frequency applications such as Physiological signals so there is no much importance for DNL and INL. However, all the performance metrics are carried out for proposed ADC as shown in the simulation results. DNL is the difference between an actual step width and ideal value of 1 LSB (Least Significant Bit).
\(V_{FSR}\) is the full scale range and N is the resolution of the ADC. Resolution of proposed ADC is 10 bit and \(V_{FSR} = 100\,{\text{mV}}\), so 1 LSB value is 100 μV.
3.4 SAR control logic
The advantage of Successive approximation register is to give zero latency compared to remaining ADC architectures. The circuit schematic of SAR Control logic is considered from Ref. [29] but the basic cell which is used in SAR control is DFF, it is designed with low on-resistance transmission gate with VTCMOS technique explained in previous sections. The gate level circuit is shown in Fig. 8.
4 Proposed low power techniques
4.1 Variable threshold CMOS (VTCMOS)
In this paper, MOS transistor substrate bias voltage is dynamically varied to control the threshold voltage as shown in Fig. 9 and followed equations. The substrate voltages of nMOS and pMOS transistors are generated by the variable substrate bias control circuit. The transmission gate using VTCMOS technique is shown in Fig. 3 The circuit operates with low-power dissipation (due to low \(V_{DD}\)) and a high switching speed (due to a low \(V_{TH}\)). When the circuit is in the standby mode, the substrate bias control circuit generates a lower substrate bias voltage for the nMOS transistor and a higher substrate bias voltage for the pMOS transistor. As a result, the magnitudes of the threshold voltages (\(V_{THn}\) and \(V_{THp}\)) increase in the standby mode due to the body bias effectively. Since the sub threshold leakage current drops exponentially with increasing threshold voltage, the leakage power dissipation in the standby mode can be significantly reduced with this circuit design technique. However, with technology scaling, the effectiveness of the VTCMOS technique reduces the channel length becomes smaller [33].
where \(V_{1b} = V_{in} \frac{{R_{4} }}{{R_{3} + R_{4} }} \,Similary \,V_{2b} = V_{in} \frac{{R_{2} }}{{R_{1} + R_{2} }}\).
4.2 Dual split switching
Dual Spilt switching is done with 2:1 MUX design. In charge scaling DAC, 2:1 Mux is used to switch between \(V_{in } \;to\; V_{ref}\) and at the Sect. 3 in Preamplifier regenerative comparator design also dual spilt switching is used to switch between amplifier and comparator. The primary element used in the dual splitting design is a transmission gate [34]. In the proposed transmission gate, another set of NMOS and PMOS transistors are connected in parallel, so that the on resistance and sampling distortion are maintained as low as possible. The length and widths are in 2:1 ratio. As mentioned in previous section transmission gate is operated with VTCMOS technique refer Fig. 10.
5 Test simulation results
The schematic of ADC was designed in the UMC 180 nm CMOS process. Chip fabrication was not carried out, but Simulation test results are verified with the theoretical. The performance of ADC is characterised by differential nonlinearity(DNL) and integral nonlinearity(INL) calculations. The simulated DNL and INL are 0.9/− 0.82 LSB and 1.06/− 1.31 LSB respectively as shown in Figs. 11 and 12. The input signal is a sinusoidal signal with frequency 100 Hz sampled at 1 MS/s with effective number of bits 7.69 bit. The power supply voltage is 0.5 V. Figure 13 shows positive level shifting for positive analog cycle and negative level shifting for negative analog cycle and Fig. 14 shows the measured 10 bit DAC divides the input range up to 1024 levels. Figure 15 shows the regenerative comparator output. The simulated SNDR and SFDR and ENOB results are shown in the Fig. 16 and Tables 2, 3.
6 Conclusion
A 10 bit SAR ADC with a 100 mV Vref input range using dual split switching network in DAC is presented. A regenerative comparator operating in the sub threshold region with low offset is used, along with low power SAR logic with negative edge triggered flip-flop at a 500 mV supply voltage. The proposed ADC consumes 13.99 μW power. The ENOB around 7.69 bits.
References
Tsai T-H, Hong J-H, Wang L-H, Lee S-Y (2012) Low power analog integrated circuits for wireless ECG acquisition systems. IEEE Trans Inf Technol Biomed 16(5):907–917
Lee S-Y, Hong J-H, Hsieh C-H, Liang M-C, Chien S-YC, Lin K-H (2013) Low-power wireless ECG acquisition and classification system for body sensor networks. IEEE J Biomed Health Inform 19(1):236–246
Seo M, Jin D, Kim Y, Hwang S, Kim J, Ryu S (2018) A 18.5 nW 12-bit 1-kS/s reset-energy saving SAR ADC for bio-signal acquisition in 0.18-μm CMOS. IEEE Trans Circuits Syst I Regul Pap 65(11):3617–3627
Taherzadeh-Sani M, Lotfi R, Nabki F (2014) A 10-bit 110 kS/s 1.16 μW SA-ADC with a hybrid differential/single-ended DAC in 180-nm CMOS for multichannel biomedical applications. IEEE Trans Circuits Syst II Express Br 61(8):584–588
Van Rethy J, De Smedt M, Verhelst M, Gielen G (2013) Predictive sensing in analog-to-digital converters for biomedical applications. In: International symposium on signals, circuits and systems ISSCS2013, Iasi, pp 1–4
Zhu Z, Liang Y (2015) A 0.6-V 38-nW 9.4-ENOB 20-kS/s SAR ADC in 0.18-μm CMOS for medical implant devices. IEEE Trans Circuits Syst I Regul Pap 62(9):2167–2176
Li Y, Poon CCY, Zhang Y (2010) Analog integrated circuits design for processing physiological signals. IEEE Rev Biomed Eng 3:93–105
Sadollahi M, Hamashita K, Sobue K, Temes GC (2018) An 11-bit 250-nW 10-kS/s SAR ADC with doubled input range for biomedical applications. IEEE Trans Circuits Syst I Regul Pap 65(1):61–73
Ginsburg BP, Chandrakasan AP (2006) A 500MS/s 5b ADC in 65 nm CMOS. In: 2006 symposium on VLSI circuits, 2006. Digest of Technical Papers, pp 140–141
Hariprasath V, Guerber J, Lee S-H, Moon UK (2010) Merged capacitor switching based SAR ADC with highest switching energy-efficiency. Electron Lett 46(9):620–621
Liu C-C, Chang S-J, Huang G-Y, Lin Y-Z (2010) A 10-bit 50-MS/s SAR ADC with a monotonic capacitor switching procedure. IEEE J Solid State Circuits 45(4):731–740
Yoshioka M, Ishikawa K, Takayama T, Tsukamoto S (2010) A 10-b 50-MS/s 820-μW SAR ADC with on-chip digital calibration. IEEE Trans Biomed Circuits Syst 4(6):410–416
Pang W, Wang C, Chang Y, Chou N, Wang C (2009) A 10-bit 500-KS/s low power SAR ADC with splitting comparator for bio-medical applications. In: 2009 IEEE Asian solid-state circuits conference, Taipei, pp 149–152
Liou C, Hsieh C (2013) A 2.4-to-5.2fJ/conversion-step 10 b 0.5-to-4 MS/s SAR ADC with charge-average switching DAC in 90 nm CMOS. In: 2013 IEEE international solid-state circuits conference digest of technical papers, San Francisco, pp 280–281
Lee P, Lin J, Hsieh C (2016) A 0.4 V 1.94 fJ/conversion-step 10 bit 750 kS/s SAR ADC with input-range-adaptive switching. IEEE Trans Circuits Syst I Regul Pap 63(12):2149–2157
Zhu Y et al (2010) A 10-bit 100-MS/s reference-free SAR ADC in 90 nm CMOS. IEEE J Solid State Circuits 45(6):1111–1121
Kim W et al (2016) A 0.6 V 12 b 10 MS/s low-noise asynchronous SAR-assisted time-interleaved SAR (SATI-SAR) ADC. IEEE J Solid State Circuits 51(8):1826–1839
Tai H, Hu Y, Chen H, Chen H (2014) 11.2 A 0.85fJ/conversion-step 10 b 200 kS/s subranging SAR ADC in 40 nm CMOS. In: 2014 IEEE international solid-state circuits conference digest of technical papers (ISSCC), San Francisco, pp 196–197
Rahiminejad E, Saberi M, Lotfi R (2017) A power-efficient signal specific ADC for sensor-interface applications. IEEE Trans Circuits Syst II Exp Br 64(9):1032–1036
Hsieh, S-E, Hsieh C-C (2016) A 0.44 fJ/conversion-step 11b 600 KS/s SAR ADC with semi-resting DAC. In: 2016 IEEE symposium on VLSI circuits (VLSI-Circuits), pp 1–2
Hsieh S-E, Hsieh C-C (2018) A 0.4 V 13b 270 kS/S SAR-ISDM ADC with an opamp-less time-domain integrator. In: 2018 IEEE international solid-state circuits conference-(ISSCC), pp 240–242
Huang G, Chang S, Liu C, Lin Y (2012) A 1-µW 10-bit 200-kS/s SAR ADC with a bypass window for biomedical applications. IEEE J Solid State Circuits 47(11):2783–2795
Yaul FM, Chandrakasan AP (2014) A 10 bit SAR ADC with data-dependent energy reduction using LSB-first successive approximation. IEEE J Solid State Circuits 49(12):2825–2834
Kim H et al (2016) A delta-readout scheme for low-power CMOS image sensors with multi-column-parallel SAR ADCs. IEEE J Solid State Circuits 51(10):2262–2273
Wattanapanitch W, Fee M, Sarpeshkar R (2007) An energy-efficient micropower neural recording amplifier. IEEE Trans Biomed Circuits Syst 1(2):136–147
Zahal CPBAU, Manjula J (2016) Design of low power high speed SAR based 16-bit analog to digital converter: charge sharing approach. In: 2016 international conference on communication and signal processing (ICCSP), Melmaruvathur, pp 1506–1510
Naveen IG, Sonoli S (2016) Design and simulation of 10-bit SAR ADC for low power applications using 180 nm technology. In: 2016 international conference on electrical, electronics, communication, computer and optimization techniques (ICEECCOT), Mysuru, pp 331–335
Abidi A, Xu H (2014) Understanding the regenerative comparator circuit. In: Proceedings of the IEEE 2014 custom integrated circuits conference, San Jose, pp 1–8
Mohtashemi D, Green MM A low-power 8-GS/s comparator for high-speed analog-to-digital conversion in 0.13 μm CMOS technology. IEEE Trans Circuits Syst II Regul Pap
Pham HT, Vu Nguyen T, Pham-Nguyen L, Sakai H, Dao TT (2016) Design and simulation of a 6-bit successive-approximation ADC using modeled organic thin-film transistors. Hindawi Publishing Corporation, Active and Passive Electronic Components, vol 2016, Article ID 7201760
Kiran Kumar M, Noorbasha F (2017) A 0.5V bulk-driven operational trans conductance amplifier for detecting QRS complex in ECG signal. Int J Pure Appl Math 117(19):147–152
Kiran Kumar M, Noorbasha F, Rao KS (2017) Design of low power 16X16 SRAM array using GDI logic with dynamic threshold technique. ARPN J Eng Appl Sci 12(22):6571–6576
Wang T, Kawahito S (2016) A variable-threshold voltage technique to enhance the linearity of folding-integration/cyclic cascaded ADCs. In: 2016 IEEE international instrumentation and measurement technology conference proceedings, Taipei, pp 1–5
Dixit A, Khandelwal S, Akashe S (2014) Power optimization of 8:1 MUX using transmission gate logic (TGL) with power gating technique. Int J Comput Appl 99(5):37–42
Acknowledgements
The authors are highly thankful to the Chairman of ANURAG group of institutions Dr. P. Rajeshwar Reddy M. L. C for his constant encouragement and also providing all the necessary resources to carry out this work. They are also thankful to Dr. K. S. R. Krishna Prasad, Professor, NIT, Warangal and Dr. K. S. Rao, Director, Anurag Group of Institutions for their valuable suggestions during this work. The authors deeply express their gratitude Department of ECE, K L. University.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Mandrumaka, K.K., Noorbasha, F. A low power 10 bit SAR ADC with variable threshold technique for biomedical applications. SN Appl. Sci. 1, 918 (2019). https://doi.org/10.1007/s42452-019-0940-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s42452-019-0940-3