China
Africa
Explore chapters and articles related to this topic
Storage and transmission of spatial sound signals
Published in Bosun Xie, Spatial Sound, 2023
In digital audio, a signal is discretized in time and amplitude and approximately represented by binary codes. Converting an analog signal into a digital signal includes three steps, e.g., sampling, quantization, and coding. Sampling is conducted to represent a continuous time signal by its values at discrete times. Sampling frequency is determined by Shannon–Nyquist sampling theorem. Quantization is performed to represent the amplitude of a discrete time signal with a finite number of discrete values. Mapping the continuous amplitude of a signal to finite discrete values leads to quantization noise. The signal-to-quantization noise ratio increases as the quantization bit increases. Quantization noise shapingimproves thesignal-to-quantization noise ratio.
Discrete signals
Published in J. Dunlop, D. G. Smith, Telecommunications Engineering, 2017
Near instantaneously companded audio multiplex (NICAM) is an alternative compression technique which is used for digital transmission of high-fidelity television stereophonic sound signals. It is essentially a bit reduction technique which is designed to maintain a constant signal-to-quantization noise ratio over a wide dynamic range. In fact A law compression may also be considered as a bit reduction technique. This is illustrated in Fig. 3.19 which depicts an example of non-linear quantization with 5 to 4 bit compression. This is, in essence, a scaled down version of Fig. 3.18, in which the compression is from 12 to 8 bits.
Noise and Digital Signal Processing
Published in Gillian M. Davis, Noise Reduction in Speech Applications, 2018
Stephan Weiss, Robert W. Stewart, Gillian M. Davi
where σx2 and σe2 are the powers of x[n] and e[n], respectively. Specifically for the assessment of quantization noise, SNR is often labeled as the signal-to-quantization-noise ratio (SQNR).
Deep Learning Techniques for OFDM Systems
Published in IETE Journal of Research, 2021
M. Meenalakshmi, Saurabh Chaturvedi, Vivek K. Dwivedi
The PAPR is defined as the ratio of peak power of an OFDM signal to its average power. OFDM signal suffers from high PAPR because many independently modulated subcarriers are added together, which leads to high peak values in the time domain [18]. The high peak value drives the power amplifier into saturation and decreases its efficiency. This high peak also affects the signal-to-quantization noise ratio of DAC and ADC. Therefore, the PAPR value has to be reduced to accomplish better performance in the OFDM system. There are two possibilities for PAPR reduction: The first possibility is to reduce the average power of the signal, but it decreases the signal-to-noise ratio (SNR) and degrades BER performance. The second possibility is to reduce the peak power of the OFDM signal to diminish the PAPR value. The PAPR of a discrete-time signal x(n) can be expressed as the following equation: where E[.] is the expectation operation, and |x(n)| is the magnitude of x(n).
An improved MASH sigma-delta modulator with all-anolog interstage and input-feedforward architecture
Published in Journal of the Chinese Institute of Engineers, 2020
Li-Jun Ge, Ze Jun Qiao, Jun Wang, Ming Zhou Li, De-Bao Zhu, Bo Dai, Qing Qing Zhao
In most of the literature, two-stage MASH architectures have become the focus due to low complexity. For the signals with wide band, it is necessary for the two-stage MASH architecture to reduce the oversampling rate, which causes low signal-to-quantization noise ratio (SQNR) (Márkus and Temes 2004; Hatami et al. 2014). Though high resolution quantizer based MASH architectures can provide high SQNRs, the large number of quantization bits makes the data-weighted averaging (DWA) logic very complicated (Sonika, Neema, and Patel 2016). Another problem of MASH architectures is that the output swing of the integrator without useful signal as an input becomes large. To solve this problem, a cascade of integrators with feedforward (CIFF) architecture is applied to MASH, which reduces the output swing of the integrator, enhancing the performance of the modulator (Song and Park 2010; Honarparvar et al. 2017).