"A standard FDM scheme for telephone signals is to multiplex 12 voice channels, each occupying 4kHz bandwidth, into a group signal with 48kHz. The next basic building block is the 60-channel supergroup, which is formed by frequency division multiplexing five group signals. The next level of the hierarchy is the mastergroup, which combines 10 supergroup inputs." My question is why do we need this grouping of voice channels at different stages and with different carriers? Why don't we just modulate our individual voice channels with the supergroup carrier frequencies?