GATE Electronics (EC): Signals and Systems (LTI systems and convolution, Fourier series and transform, DTFT, Laplace and z-transform, sampling, DFT and digital filters), previous-year questions solved, part 2
Part 2 of 4 of the Signals and Systems hub: 4 more official GATE EC questions on Fourier series and DTFS, each solved step by step and checked against the official answer key. Part 1 covers signals and system properties, impulse response and convolution, Fourier transform, DTFT and frequency response, Laplace and z-transform and sampling, DFT and digital filters.
Questions from official GATE EC papers (IIT Kharagpur 2022, IIT Kanpur 2023, IISc Bengaluru 2024, IIT Roorkee 2025, IIT Guwahati 2026); solutions by Skill-Lync. · Updated 17 Nov 2026 · 4 of 31 questions on this page
DTFS of period-5 signal, a0=1, a1=3j, a2=2j, a3=-2j, a4=-3j: sum of x[n] sin(4 pi n/5)
- (A)
- (B)
- (C)
- (D)
Answer: (A) Official key: A
- Write the sine with exponentials: , which involves harmonic .
- The DTFS analysis equation is . So and .
- and .
- The sum is , option (A).
Trap: option (B), , comes from using where belongs, which flips the sign. Option (C), , is the answer with the factor forgotten. The code built from the coefficients (and again with 5*ifft(a)), summed it against and got . The swapped-index path gives and the missing- path gives , which are options (B) and (C).
The DTFS is what an N-point FFT computes for a periodic signal, so extracting a single harmonic with a weighted sum like this is how a lock-in or single-bin (Goertzel-style) detector measures one tone, for instance a DTMF key or a mains harmonic in a power meter. The factor N is the normalisation that FFT libraries leave to the user.
Periodic f(t) with period T0 and y(t)=f(alpha t), alpha gt 1: Fourier coefficients and fundamental period
Let be a periodic signal with fundamental period . Consider the signal , where .
The Fourier series expansions of and are given by
Which of the following statements is/are TRUE?
- (A) for all
- (B) is periodic with a fundamental period
- (C) for all
- (D) is periodic with a fundamental period
Answer: (A) and (D) Official key: A;D
- Substitute in the series of : .
- Compare with . The exponents are identical, and the exponentials are orthogonal, so the coefficients match term by term: for all . Time scaling moves the harmonics but does not change their amplitudes. (A) is true and (C) is false.
- The harmonics of sit at , so the fundamental frequency is and the fundamental period is . Directly: . (D) is true.
- is a period of only when is an integer, and even then it is a multiple of the fundamental period, not the fundamental period. (B) is false.
- Answer: (A) and (D).
Trap: (C), by confusing the expansion with the Fourier transform scaling property, where brings a factor. For Fourier series coefficients there is no such factor, because they are averages over one period. The code took with and and : it computed the coefficients by numeric integration over one period (largest was 1e-16, while reached 0.67) and found the smallest with by scanning: 0.333 for (that is ) and 0.400 for (that is ), never .
Playing a periodic waveform faster, as in a function generator or a DDS (direct digital synthesis) chip stepping through a lookup table, scales the period by and moves every harmonic up by the factor with the same amplitude. This is why a faster-played tone keeps its tone colour (harmonic amplitudes) but changes pitch.
Real periodic x(t) with x(t - T/2) = -x(t): which relation holds for the Fourier series coefficients a_2m?
The Fourier series representation of and satisfies the following: .
For any integer , which of the following options is correct?
- (A)
- (B)
- (C)
- (D)
Answer: (A) Official key: A
- Write the series for the shifted signal: . So the coefficients of the shifted signal are .
- The condition says these equal . Fourier coefficients are unique, so , that is .
- For odd the bracket is and is free. For even the bracket is , so . With this is for every integer , including the mean, : option (A).
- Options (B) and (D) fail at once, since is , not . Option (C) would force as well, which is not true: a square wave obeys the condition and has .
Trap: mixing this up with odd symmetry, . That only removes the cosine parts and says nothing about even harmonics: the sawtooth on one period is odd but its second coefficient has magnitude , not . The code built signals that satisfy (random odd-harmonic sums and a square wave), took the Fourier coefficients with an FFT of one period, and found for on all of them, while (B), (C) and (D) failed on every one.
Half-wave symmetry is why symmetric push-pull stages and full-bridge inverters produce output with only odd harmonics, and why their distortion analysis starts at the third. Spotting the symmetry saves you from computing a whole family of coefficients that are zero.
x1 = cos(2 pi n t), x2 = 2 sin(4 pi n t): orthogonality, orthonormality and harmonic statements?
Which of the following statements about and is/are valid?
- (A) and are orthogonal to each other over .
- (B) and are orthonormal to each other over .
- (C) is a harmonic of .
- (D) and are non-orthogonal to each other over .
Answer: (A) and (C) and (D) Official key: A;C;D
- Orthogonal over an interval means there. Orthonormal also needs each signal to have unit energy on that interval.
- Over (one period of ): , since each sine completes whole cycles. So (A) is true.
- Energies over the same interval: and . Neither is (for any they are and ), so (B) is false.
- has frequency and has , an integer multiple, so is the second harmonic of and (C) is true.
- Over half of that interval, , substitute : . Orthogonality is lost, so (D) is true.
- Answer: (A), (C) and (D).
Trap: assuming that harmonics are orthogonal on every interval. They are orthogonal over a whole period (or a multiple of it), not over half a period. The code computed the inner product numerically for : over to it is below , over to it equals ( for , never zero). It also found the energies and , so (B) fails.
Orthogonality of sinusoids over a whole period is what lets a spectrum analyser or an OFDM receiver separate tones: each carrier or harmonic can be picked out by correlating over an integer number of cycles. Windows that cut a cycle short leak energy between bins.