Author Affiliations
1University of Chinese Academy of Sciences, Beijing 100049, China2Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, Chinashow less
Fig. 1. (Color online) Performance comparison of state-of-the-art DACs: SFDR@1GHz vs. sampling rate.
Fig. 2. (Color online) Performance comparison of state-of-the-art DACs: (a) FoM1, (b) FoM2, (c) FoM3 versus sampling rate.
Fig. 3. (Color online) Block diagram of a high-speed high-resolution current steering DAC with a typical switching current cell.
Fig. 4. A typical switching current cell proposed in Ref. [21].
Fig. 5. Quad-switching current cell with switch cascodes reported in Ref. [38].
Fig. 6. A simplified quad-switch cell proposed in Ref. [34].
Fig. 7. RZ current cell with resampling switches proposed in Ref. [11].
Fig. 8. (Color online) Magnitude of the frequency response for NRZ, RZ, and MRZ waveforms reported in Ref. [53].
Fig. 9. A typical fast latch proposed in Ref. [21]
Fig. 10. High-speed latch presented in Ref. [57].
Fig. 11. DAC output stack, with the switch driver proposed in Ref. [37].
Fig. 12. Block diagram of dummy trigger proposed in Ref. [56].
Fig. 13. Master-slave latch presented in Ref. [58].
Fig. 14. Double-edge current switch driver with enhanced reset circuit reported in Ref. [59].
Fig. 15. Two-parameter calibration loop configuration and CAL_DACs proposed in Ref. [28].
Fig. 16. Conceptual block diagram of a dual-rate hybrid DAC architecture proposed in Ref. [49].
Fig. 17. OIC technique with compensation resistor proposed in Ref. [35].
Parameter | Ref. [4]
| Ref. [11]
| Ref. [17]
| Ref. [30]
| Ref. [35]
| Ref. [36]
| Ref. [37]
| Ref. [38]
| Ref. [39]
| Ref. [40]
|
---|
Process (nm) | 28 | 130 | 16 | 40 | 28 | 16 | 40 | 65 | 28 | 65 | Resolution (bit) | 13 | 14/12 | 16 | 12 | 14 | 14 | 14 | 16 | 9 | 16 | Sampling rate (GS/s) | 9 | 7.2/12 | 6 | 1.6 | 10 | 6.8 | 8.9 | 10 | 11 | 9/12 | SFDR@Nyquist frequency (dBc) | N/A | 67/55 | 67 | 70 | 65 | 62 | 50 | 69 @3GS/s | 51 | 56/52 | IM3@DC-Nyquist frequency (dBc) | <–45 | N/A | <–82 | <–70 | <–70 | <–71 | <–65 | <–73 @3GS/s | <–51 | <–67/<–67 | NSD (dBm/Hz) | N/A | –161/–159 | –162 @2.6GHz | –150 @800MHz | –158 @5GHz | –160 | N/A | N/A | N/A | –130 @6GHz | Power (mW) | 360 | N/A | 350 | 40 | 162 | 330 | 1200 | 800 | 110 | 758/1065 | Area (mm2)
| 1.16 | N/A | 0.52 | 0.016 | 0.07 | 0.855 | N/A | N/A | 0.04 | 0.97 | FoM1 (GHz/mW) | N/A | N/A | 6.5×105 | 4.4×105 | 4.5×105 | 2.7×105 | 5.6×103 | 2.5×105 | 3.8×104 | 6.3×104/
1.9×104 | FoM2 (GHz/mW) | N/A | N/A | 18.7 | 102.4 | 10.1 | 12.4 | 1.4 | 16.4 | 27.9 | 8.6/6.2 | FoM3(GHz/(mW·mm2))
| N/A | N/A | 2.4×106 | 2.6×107 | 2.4×106 | 2.4×105 | N/A | N/A | 3.6×105 | 5.8×105/
4.2×105 |
|
Table 1. Performance summary and comparison with state-of-the-art high-speed high-resolution DACs.
| FoM1 | FoM2 | FoM3 |
---|
Definition | | | | Reference | [30]
| [41]
| [42]
| Explanation | SFDRBest/SFDRworst: Best/Worst measured SFDR in whole Nyquist bandwidth;
fclk: Sampling rate;
Ptotal/Pload: Power consumption of the whole DAC/load;
N: Resolution;
fs@6(N–1): Output signal frequency where the SFDR has dropped with 6 dB (= 1 bit) in comparison with the expected result (≈ 6N) (Note: If the measured SFDR cannot reach 6(N–1), 0.1 GHz is selected here for calculation);
Area: The core area of the DAC.
|
|
Table 2. Detailed definitions of DAC FoMs.
Parameter | INL at 40 °C (14-bit level LSB) | DNL at 40 °C (14-bit level LSB) | Temperature drift 1σ (LSB) |
---|
No CAL | 6.02 | 3.33 | 1.6 | Ref. [24]
| 0.20 | 0.16 | 1.6 | Ref. [27]
| 0.23 | 0.16 | 0.8 | Ref. [28]
| N/A | N/A | 0.6 |
|
Table 3. INL, DNL and temperature drift summary of proposed foreground calibration techniques.