spectral efficiency

Deepanshu sharma Hi,
Can you attach complete paper. From the snap shot it is showing coding efficient 1 bit / symbol, spectral efficiency .5 bits/sec/Hz.

The spectral efficiency of a digital communication system is measured in bit/s/Hz. It is the net bitrate (useful information rate excluding error-correcting codes) or maximum throughput divided by the bandwidth in hertz of a communication channel.
It can also be measured in bit/symbol, showing the net bit rate divided by the symbol rate.

Hi Deepanshu Sharma,
Downloaded the article from the link.

From the article:
New speed for the signal

We now effectively deal with two different speeds.
First is the bit rate ftx, measured in bits per second, also referred to as “transmission rate.”
Second, we define the symbol rate S that quantifies the number of symbols transmitted per second,
With the coding efficiency in bits/symbol: e = log2 (number of symbols in alphabet),

the symbol rate calculates as: the bit rate ftx measured in bits per second / the coding efficiency in bits/symbol

The minimum optical bandwidth required by the signal (in Hz) is determined by dividing the symbol rate by 2.
Therefore, the bandwidth required by complex modulated signals doesn’t depend on the data rate but only on the symbol rate.
That also means the more bits encoded into one symbol at a given data rate, the greater the reduction in occupied optical bandwidth.

So, there is not any affect of DD method on the spectral efficiency.

Regards.
Sethi Rk

Hi Rk Sethi,

I understood your point about the differences in 2 bit rates. However, I have few following questions according to your explanation about the definition of Spectral Efficiency itself. 1.What are the main units for spectral efficiency? 2.Do we consider ‘coding symbol efficiency’ as a ‘source coding efficiency’ or we don’t? 3.Which of these efficiencies can we consider as the Internal Spectral Efficiency (or ISE), which we consider for optical systems using superchannels, for example?