Skip to main content

Digitally Assisted Analog Circuits Unraveled

It may be the first time you've heard of the term, with function implied by semantics. But what are digitally assisted analog circuits? How do they perform or give "assistance" to the analog part exactly? What are their advantages and disadvantages? Does this field have any potential in emerging technologies (in my opinion they do, which is why I've decided to write all about them in this article).

Before anything else, a proper background to digitally assisted analog circuits is in order (what a mouthful! I'll be writing this as DAAC or DAA for short throughout the article).  

It all started when scaled CMOS technology began progressing rapidly, challenges in analog design becoming more difficult to surmount. Remember that with decreasing area, precision decreases and devices become more nonlinear. In addition, estimates show advances in the digital sector equating to 3 times longer in analog for catching up. Obviously, there was a burgeoning demand for a solution to this lag. One such solution that was cooked up is DAAC design.


What are all these D.A.A.C.s, Anyhow?


DAACs supposedly bring the best of both analog and digital by integrating them. The digital fabric, which could be a digital signal processor, implements corrections to the unregulated output from the analog block. The most famous applications are in DACs and ADCs [2], with impressive power efficiency and performance improvements suggested by most recent studies. Digital calibration (to be discussed later) solved analog deficiencies in deep submicron (that's <0.1 um) CMOS technology with increased tolerance in mismatch at almost no cost in precision.

Other applications involve amplifiers, where digital assistance relaxes requirements on linearity and gain precision with the critical requirement at input referred noise. Aside from lower power consumption in conventional amplifiers, many alternative amplifier topologies can be used.  Another application is a 28nm Mobile SoC (based from a study conducted by Broadcom Corp. in Irvine, CA) [3]. You can observe the contrast in performance from the table (Fig. 10)

Note: "This Work" denotes testing on the 28nm Mobile SoC - a DSP and application switching regulator (ASR).

Some DAAC applications don't just stop with a combination of analog and digital. They also add a bit of machine learning. Now we have a machine-learning digitally assisted analog circuit. But why would we want such a feature for a DAAC? Well, for a delta-sigma DAAC - gradient descent purports improving online calibration of parameters (especially to effects of local minima).


How do these digitally-assisted-analog-circuits "assist"?


There are in general 2 types of assistance: digital compensation and digital calibration. Digital compensation refers to a digital mapping of raw output bits to a corrected, final set of output bits. On the other hand, digital calibration involves using digital calibration settings to adjust analog circuit parameters, directly correcting output values from the analog block.
Fig. 1 provides a simple illustration of how both types work.

Fig. 1 Shown are block diagrams comparing (a) Digital Compensation (b) Digital Calibration


There are many ways to apply digital assistance in circuit design. One direct approach is by adjustment of comparator thresholds (which are at the heart of most ADC architectures). Popular methods for calibration include varying the load capacitance, the body voltages of the transistors at the input of the comparator, and using a redundant input pair of those said transistors. 

Comparator adjustment shifts limitations in accuracy from offset to electrical noise, which significantly reduces the power consumed by the comparator, as well as its input capacitance. In addition, some alternative architectures that potentially further power efficiency become possible (as with amplifiers). In DACs, digital assistance reduces required area for matching, which is useful at high resolutions.

From theory, digital signals carry a lot more noise than analog signals and a structure that combines both analog and digital provides a path for the noise from the digital block to get coupled to analog. This is an important consideration because it will be difficult to provide separate isolated paths for the analog and digital ground if they are close to each other on the layout.

Given the arguments above, it is easy to presume that DAACs have the potential of offering a bright future to analog. Of course, the ultimate decision will lay upon how the industry and consumers will respond, as well as on whether any competing solutions can outshine this approach. 



References

[1] https://people.eecs.berkeley.edu/~boser/presentations/2004-0518%20Digitally%20Assisted%20ADCs.pdf

[2] Fang-shi Lai, Yung-Fu Lin, et.al.(2010). Digitally-assisted analog designs for submicron CMOS technology. Proceedings of 2010 International Symposium on VLSI Design, Automation and Test, IEEE Conference Publications, pp. 49-52

[3] Xicheng Jiang; Narayan Prasad Ramachandran, et.al. (2014). Digitally-assisted analog and analog-assisted digital design techniques for a 28 nm mobile system-on-chip. ESSCIRC 2014 - 40th European Solid State Circuits Conference, pp. 475-478


Comments

  1. Hello Justin,

    First of all, I would like to thank you for following Power Electronics News.

    I just visited your blog and your articles are great. Do you mind if we feature you on our website Sixty Second Tech and then exchange links at the same time?

    If this is something that interests you, please email me at andrei@sixtysecondtech.com

    Best regards,
    Andrei

    ReplyDelete

Post a Comment

Popular posts from this blog

Calculator Techniques for the Casio FX-991ES and FX-991EX Unraveled

In solving engineering problems, one may not have the luxury of time. Most situations demand immediate results. The price of falling behind schedule is costly and demeaning to one's reputation. Therefore, every bit of precaution must be taken to expedite calculations. The following introduces methods to tackle these problems speedily using a Casio calculator FX-991ES and FX-991EX.


►For algebraic problems where you need to find the exact value of a dependent or independent variable, just use the CALC or [ES] Mode 5 functions or [EX] MENU A functions.


►For definite differentiation and integration problems, simply use the d/dx and integral operators in the COMP mode.


►For models that follow the differential equation: dP/dx=kt and models that follow a geometric function(i.e. A*B^x).

[ES]
-Simply go to Mode 3 (STAT) (5)      e^x
-For geometric functions Mode 3 (STAT) 6 A*B^x
-(Why? Because the solution to the D.E. dP/dx=kt is an exponential function e^x.
When we know the boundary con…

Common Difficulties and Mishaps in 6.004 Computation Structures (by MITx)

Updated: 
May 6, 2018
VLSI Project: The Beta Layout [help needed]Current Tasks: ►Complete 32-bit ALU layout [unpipelined] in a 3-metal-layer C5 process. ►Extend Excel VBA macro to generate code for sequential instructions (machine language to actual electrical signals).
Current Obstacles/Unresolved Decisions:
►Use of complementary CMOS or pass transistor logic (do both? time expensive, will depend on sched.
►Adder selection: Brent-Kung; Kogge Stone; Ladner Fischer (brent takes up most space but seems to be fastest, consider fan-out) [do all? time expensive, will depend on sched.)
►layout requirements and DRC errors

Please leave a comment on the post below for advise. Any help is highly appreciated.




Yay or Nay? A Closer Look at AnDapt’s PMIC On-Demand Technology

Innovations on making product features customizable are recently gaining popularity. Take Andapt for example, a fabless start-up that unveiled its Multi-Rail Power Platform technology for On-Demand PMIC applications a few months back. (read all about it here: Will PMIC On-Demand Replace Catalog Power Devices?) Their online platform, WebAmp, enables the consumer to configure the PMIC based on desired specifications. Fortunately, I got a hands-on experience during the trial period (without the physical board (AmP8DB1) or adaptor (AmpLink)). In my opinion, their GUI is friendly but it lacks a verification method for tuning (i.e. the entered combination of specs). How would we know if it will perform as expected or if there are contradicting indications that yield queer behavior? Also, there is not just one IP available, but many that cater to a differing number of channels and voltage requirements (each with their own price tag).
Every new emerging technology has the potential to oversh…