Question 1

What is the Nyquist frequency in DSP?

Accepted Answer

The Nyquist frequency is exactly half of the sampling rate. It represents the absolute maximum frequency that a digital system can represent without distortion. For example, CD audio sampled at 44,100 Hz has a Nyquist frequency of 22,050 Hz, which covers the full human hearing range of roughly 20 Hz to 20,000 Hz. Any frequency above the Nyquist frequency will fold back into the spectrum as aliasing artifacts.

Question 2

What is the Nyquist-Shannon sampling theorem?

Accepted Answer

The Nyquist-Shannon sampling theorem states that to perfectly reconstruct a continuous signal from its samples, the sampling rate must be at least twice the highest frequency present in the signal. This minimum rate (2 x fmax) is called the Nyquist rate. Sampling below this rate causes aliasing, where high-frequency components masquerade as lower frequencies in the digital representation.

Question 3

Why is oversampling used in audio and DSP?

Accepted Answer

Oversampling means sampling at a rate higher than the Nyquist rate minimum. It provides three key benefits: it relaxes anti-aliasing filter requirements (allowing a gentler rolloff), it spreads quantization noise across a wider bandwidth (improving effective SNR), and it provides a guard band between the signal bandwidth and the Nyquist frequency. Common oversampling ratios are 2x, 4x, and 8x. Professional audio often uses 96 kHz or 192 kHz sampling for a 20 kHz signal bandwidth.

Question 4

How does bit depth affect dynamic range in digital audio?

Accepted Answer

Each additional bit of depth adds approximately 6.02 dB of dynamic range. A 16-bit system provides about 98 dB of dynamic range (6.02 x 16 + 1.76), which is sufficient for most consumer audio. A 24-bit system provides roughly 146 dB, exceeding the range of human hearing (about 120 dB). The 1.76 dB constant accounts for the quantization noise floor of a uniform quantizer.

Question 5

What is frequency resolution and how does FFT size affect it?

Accepted Answer

Frequency resolution equals the sampling rate divided by the FFT size (fs / N). A larger FFT provides finer frequency resolution but requires more samples and processing time. For example, at 44,100 Hz sampling with a 1,024-point FFT, the resolution is 43.07 Hz. Doubling the FFT to 2,048 points halves the resolution to 21.53 Hz. This creates a fundamental tradeoff between frequency precision and time resolution in spectral analysis.

Question 6

What causes aliasing and how do you prevent it?

Accepted Answer

Aliasing occurs when a signal contains frequency components above the Nyquist frequency (half the sampling rate). These components fold back into the spectrum as false lower frequencies that cannot be distinguished from genuine signals. Prevention requires two steps: first, apply an analog anti-aliasing low-pass filter before the ADC to attenuate frequencies above the Nyquist frequency; second, ensure the sampling rate is at least twice the highest frequency of interest. Oversampling provides additional protection.

Application	Sampling Rate	Nyquist Freq	Bit Depth	Dynamic Range
Telephone (narrowband)	8,000 Hz	4,000 Hz	8 bits	49.92 dB
AM Radio	22,050 Hz	11,025 Hz	16 bits	98.08 dB
CD Audio	44,100 Hz	22,050 Hz	16 bits	98.08 dB
DVD Audio	48,000 Hz	24,000 Hz	24 bits	146.24 dB
Professional Studio	96,000 Hz	48,000 Hz	24 bits	146.24 dB
Hi-Res Audio	192,000 Hz	96,000 Hz	24 bits	146.24 dB
Ultrasound Imaging	40,000,000 Hz	20,000,000 Hz	12 bits	74 dB
Radar (Pulse)	100,000,000 Hz	50,000,000 Hz	14 bits	86.04 dB

Calculadora DSP

Insira os valores

Resultado

Formula

Exemplo Resolvido

Fundamentos do Processamento Digital de Sinais

Common Sampling Rates and Applications

Perguntas Frequentes

O que e a frequencia de Nyquist?

O que e o teorema de Nyquist-Shannon?

Como a profundidade de bits afeta a faixa dinamica?

O que e resolucao de frequencia?

Por que usar sobreamostragem?

O que e aliasing e como evita-lo?

Como adiciono esta Calculadora DSP ao meu site?

Calculadoras Relacionadas

Accurate and Reliable