Spectral density refers to the distribution of power or energy over frequency in a signal. The unit of spectral density depends on whether it refers to power spectral density (PSD) or energy spectral density (ESD):
- The unit of power spectral density (PSD) is usually watts per hertz (w/hz) or dbm/hz (decibels relative to a milliwatt per hertz). The PSD quantifies the power distribution of a signal across its frequency spectrum, indicating how power is distributed across different frequency components.
- The unit of energy spectral density (ESD) is joules per hertz (j/hz) or dbm/hz. ESD measures the energy distribution of a signal across its frequency spectrum, providing insight into the energy content across different frequency components.
Spectral density is measured using various techniques depending on the type of signal and the desired spectral analysis. For continuous signals, the spectral density can be calculated using Fourier transform techniques such as the Fourier transform or estimating the power spectral density using methods like the Welch method. For discrete signals, spectral density can be estimated using techniques such as periodogram or multiple spectral estimation.
Power spectral density in decibels (dB) is commonly used to express the power distribution of a signal over frequency in a logarithmic scale relative to a reference level. The unit of power spectral density in DB is usually DBM/Hz (decibels relative to one milliwatt per hertz). This unit indicates the power level of a signal relative to a reference power of one milliwatt across each hertz of bandwidth. The use of DB allows convenient representation and comparison of signal power levels over different frequency ranges or bandwidths.