Frequency Selective Fading
Frequency-selective fading occurs when multiple copies of a transmitted signal arrive at the receiver with different propagation delays, typically due to multipath propagation with varying path lengths. In channel modeling, a tap or cluster represents a group of multipath components (MPCs) arriving at approximately the same time. While there may be theoretically an infinite number of taps, practical mobile communication systems consider a finite number.
The frequency response of the channel varies because the different delayed paths cause constructive and destructive interference at different frequencies. Modeling this effect can be done using a linear, time-variant (LTV), and causal system description of the mobile radio channel.
Flat Fading vs Frequency Selective Fading
Flat fading occurs when the signal bandwidth is smaller than the channel's coherence bandwidth, meaning all frequency components experience similar fading. Narrowband signals are typically affected by flat fading, and diversity reception or error-correction coding is used to mitigate its effects.
Frequency-selective fading occurs when the signal bandwidth exceeds the channel's coherence bandwidth (or equivalently, when the symbol duration is smaller than the channel's delay spread). In this case, different frequency components of the signal experience different levels of fading, leading to inter-symbol interference (ISI) unless equalization techniques are applied.
Summary
In the case of frequency-selective fading, multipath propagation occurs, and each multipath component is affected by different fading characteristics. The signal reaches the receiver through multiple paths in such a way that one symbol interferes with another. The root cause of this interference is that multiple delayed copies of the same signal arrive at the receiver with delays that exceed the symbol duration. As a result, the current symbol interferes with the subsequent symbol, leading to inter-symbol interference (ISI).
In the case of flat fading, this phenomenon does not occur because the signal bandwidth is much smaller than the channel’s coherence bandwidth. In other words, the channel response remains approximately constant (flat) over the entire signal bandwidth. You can think of it as the channel remaining relatively unchanged over a certain duration, while the symbol duration is much shorter than the time over which the channel varies significantly, preventing ISI.