Analog Devices Interview Questions and Answers for India Embedded and DSP Roles
Last updated 2026-11-19
Analog Devices is known for data converters, signal-chain parts and the software that surrounds them. A software or firmware interview that touches this area tends to ask about what happens between the real-world signal and your code: how an ADC samples, how noise limits resolution, how a digital filter is built, and how you move the data safely over SPI and DMA. This page has twelve practice questions with worked answers in that area.
They are practice questions for the topics such roles are commonly reported to cover, not copies of a real interview. For coding problems in the same style, see Analog Devices coding round questions.
| Typical roles | Embedded software, firmware, DSP software, applications and test engineering |
|---|---|
| Core topics | ADC and sampling, filters, SPI and I2C, fixed-point maths, embedded C |
| Rounds (commonly reported) | Online or written test, technical interviews, HR |
Rounds as commonly reported
Prep guides usually describe a screening test, then technical interviews that start from your resume and move into electronics fundamentals, C and signal processing, followed by an HR round. The company does not publish a fixed process, and the weighting differs by team: a test-software role leans on programming, a DSP role on mathematics.
Be ready to do arithmetic on paper: LSB size, cut-off frequency, bits of resolution. Interviewers value a correct number with the units and the assumptions stated.
Topics to revise
- Sampling, aliasing, oversampling, and the main ADC architectures.
- SNR, SINAD and effective number of bits.
- Op-amp and RC filter basics; FIR and IIR digital filters.
- SPI modes, I2C basics, DMA with double buffering.
- Fixed-point arithmetic and saturation in C.
Hiring status
This site holds no sourced hiring statements for Analog Devices, so it quotes no drive dates, number of openings or packages. Treat any post that does without a source with caution.
Check the company's own careers site for current openings. For general interview preparation, see how to crack the technical interview.
Analog Devices Interview Questions and Answers for India Embedded and DSP Roles: questions with worked answers
1 What is the ideal SNR of an N-bit ADC, and what is ENOB?
easyFor a full-scale sine wave, the ideal signal-to-noise ratio is about 6.02 N + 1.76 dB, so a 12-bit converter gives about 74 dB. Real converters are worse because of thermal noise, distortion and clock jitter. The effective number of bits is derived from the measured SINAD: ENOB = (SINAD - 1.76) / 6.02. A measured 68 dB gives (68 - 1.76) / 6.02, about 11.0 bits, so a nominal 12-bit part behaves like an 11-bit one.
2 Compare SAR, sigma-delta and pipelined ADCs.
mediumA successive-approximation (SAR) ADC does a binary search with a DAC and comparator, one bit per clock, giving moderate speed and resolution with low power; it is common in microcontrollers. A sigma-delta ADC oversamples at a very high rate with one-bit quantisation, shapes the noise to high frequencies and removes it with a digital filter, giving high resolution at low bandwidth, as in precision measurement and audio. A pipelined ADC splits conversion into stages that work on different samples at once, giving high sample rates for communications and imaging at higher power.
3 What is oversampling and how many extra bits does it give?
mediumSampling faster than needed spreads the fixed quantisation noise over a wider band; a digital low-pass filter then removes the part outside the band of interest. For white noise, each factor of 4 in oversampling ratio gives about one extra bit of resolution (3 dB per doubling). So 16 times oversampling adds about two bits. The gains need noise at roughly one LSB to dither the signal, and they reduce the effective output rate.
4 An RC low-pass filter has R = 1.59 kohm and C = 100 nF. What is the cut-off frequency?
easyThe -3 dB frequency is fc = 1 / (2 pi R C). With R = 1590 ohm and C = 100 x 10^-9 F, RC = 159 microseconds, so fc is about 1 kHz. Beyond fc, a first-order filter rolls off at 20 dB per decade. As an anti-aliasing filter it must sit well below half the sample rate, and a single pole is rarely steep enough, so higher-order filters or oversampling are used.
5 Compare FIR and IIR digital filters.
mediumAn FIR filter output is a weighted sum of recent inputs only, so it is always stable, can have exactly linear phase if its coefficients are symmetric, but may need many taps for a sharp cut-off. An IIR filter feeds back past outputs, so it reaches a sharp response with a few coefficients, but it can be unstable, has non-linear phase and is sensitive to coefficient quantisation, so it is usually built from second-order sections. Choose FIR when phase matters, IIR when computation is tight.
6 Implement an FIR filter efficiently on a microcontroller.
mediumKeep the last N samples in a circular buffer so there is no data shuffling, multiply each by its coefficient and accumulate in a wider type than the data. Finally shift down and saturate. Cost is N multiply-accumulates per output sample, which DSP cores do one per cycle. Using Q15 data with a 32-bit accumulator avoids overflow of intermediate sums. The code below filters with a 4-tap average.
c#include <stdio.h> #include <stdint.h> #define N 4 int16_t fir(int16_t x, int16_t *hist, int *pos, const int16_t *h) { hist[*pos] = x; int32_t acc = 0; for (int k = 0; k < N; k++) { int idx = (*pos - k + N) % N; acc += (int32_t)hist[idx] * h[k]; } *pos = (*pos + 1) % N; return (int16_t)(acc >> 15); } int main(void) { const int16_t h[N] = {8192, 8192, 8192, 8192}; /* 0.25 each in Q15 */ int16_t hist[N] = {0}; int pos = 0; int16_t in[] = {1000, 1000, 1000, 1000, 1000}; for (int i = 0; i < 5; i++) printf("%d ", fir(in[i], hist, &pos, h)); puts(""); /* 250 500 750 1000 1000 */ return 0; }7 Explain the four SPI modes.
mediumSPI mode is set by clock polarity (CPOL), the idle level of the clock, and clock phase (CPHA), which says whether data is sampled on the first or second clock edge. Mode 0 is CPOL 0, CPHA 0: idle low, sample on the rising edge. Mode 1: idle low, sample on the falling edge. Mode 2: idle high, sample on the falling edge. Mode 3: idle high, sample on the rising edge. Master and slave must use the same mode, which is the first thing to check if reads return garbage.
8 Why use DMA with double (ping-pong) buffering for ADC data?
mediumThe ADC produces samples continuously, and the CPU must not lose any while it processes the last block. With two buffers, DMA fills one while the CPU processes the other; a half-transfer or complete-transfer interrupt tells software which buffer is ready, and the roles swap. This decouples the processing time from the sample interval, as long as processing a block takes less than the time to fill the other. If it does not, samples overrun and the firmware should flag it.
9 Why avoid floating point in an interrupt service routine on a small core?
mediumCores without a floating-point unit emulate it in software, which takes many cycles, and the ISR may also need to save more registers. Long ISRs raise latency for other interrupts. Prefer fixed-point integer arithmetic in the ISR, or just move samples to a buffer and process them in the main loop. If an FPU exists, check whether its registers are saved lazily on entry to the ISR.
10 How do you sign-extend a 12-bit two's complement ADC code stored in a 16-bit variable?
mediumIf bit 11 is set the value is negative. Flip bit 11 and subtract 2048: (x ^ 0x800) - 0x800 gives the signed value in one expression. 0xFFF becomes -1 and 0x800 becomes -2048. An alternative is to shift left by 4 into the top of a 16-bit signed variable and arithmetic shift right by 4, but the shift of a signed value is implementation-defined in C, so the XOR form is more portable.
11 HR: Why Analog Devices, and what interests you about signal processing?
easyAnchor your answer in something you did: a project that read a sensor, a filter you implemented in MATLAB or C, a lab where you saw aliasing happen. Say what the result taught you and what you want to learn next. Avoid generic statements about the company being a leader. Also read how to crack the HR interview.
12 HR: Tell me about a time you had to learn a topic quickly for a project.
mediumPick a real case, such as learning an unfamiliar sensor interface or a new toolchain. State the deadline, how you split the topic into learnable parts, which sources and people you used, what you built to test your understanding, and the outcome. Close with what you would do faster the next time.
Frequently asked questions
Do Analog Devices interviews include electronics for software candidates?
Commonly reported for embedded and firmware roles: sampling, ADCs and filters can come up beside C. The depth depends on the team.
Which DSP topics are most useful to revise?
Sampling and aliasing, FIR versus IIR filters, the DFT and FFT idea, quantisation and fixed-point effects. Be able to do one numerical example of each.
Are these real Analog Devices questions?
No. They are practice questions in the style of topics such roles commonly cover. The company does not publish its interview questions.
Where do I find current openings?
Use the company's own careers site. This site holds no sourced hiring facts for Analog Devices, so it quotes none.
Related guides
- Analog Devices Coding Round Questions with Solutions
- Texas Instruments Interview Questions and Answers for India Embedded Roles
- NXP Interview Questions and Answers for India Embedded Software Roles
- Micron Interview Questions and Answers for India Software and Firmware Roles
- C Language Interview Questions and Answers
- Bit manipulation interview questions with worked solutions
More in Company Interview Questions
- Apple Coding Round Questions with Solutions in Python and Java
- Apple Interview Questions and Answers for SDE Roles
- Arcesium Coding Round Questions With Solutions
- Arcesium Interview Questions and Answers
- Atlassian Coding Round Questions with Solutions in Python and Java
- Atlassian Interview Questions and Answers for SDE Roles
- Barclays coding round questions with solutions
- Barclays interview questions with answers (technology graduates)
- Birlasoft Coding Round Questions with Solutions
- Birlasoft Interview Questions for Freshers
- BNY coding round questions with solutions
- BNY interview questions with answers (technology roles)