if your calculating with hardware implemenation you can use bellow equation to find the

Total execution time = max no of bits per packets * clock time period

Priyansi, in general terms, encryption time is the time required to take the plaintext and convert it to ciphertext. And decryption time is the other way around, restoring the plaintext, from the received ciphertext.

In a stream cipher, you should see this manifested as a time lag, in transmission, and then again in the receiver. Extra delay, caused only by the encryption and decryption processes, not to be confused with time required just for the typical buffering.

Execution time varies, depending on the security protocol and on the security algorithms used.

The execution time depends on many factors in your hardware and software. It also depends on the code you used for measuring the time. I would suggest you contact the authors and ask them to kindly share their code. Going back to your second question, the execution time can also be affected by buffering. As a general note, it is better to measure the execution time of an algorithm by CPU clock cycles per byte rather than seconds.

I don't know what do the mentioned authors meant, but there is type of estimation - to measure the actual time, which the encryption/decryption process took. It is handy for comparison, when several algorithms are implemented on the same hardware.

The difference in time might be caused by the difference in hardware and used software development tools. Python provides useful tools to make development process easy, but the price is vast loss of productivity. It might happen, that the authors used C++, for instance, and/or dived deep into algorithm optimization to meet used hardware means. From my experience, 30-40 times performance difference in cryptography related algorithms implementation might be caused purely by usage of Python/Java/JavaScript/PHP languages instead of Assembler/C++ for their implementations.

The encryption time depends on the implementation platform. If you implement the code using a standard platform (C++/Python), it will not give you the exact encryption time since the time required for many system calls will also be added to the total encryption time. The code has to be implemented in emulator (e.g., ATEMU, used to calculate CPU cycles) or lightweight simulator (e.g., TOSSIM) to get more accurate time for an encryption operation.

Hope, you include the time taken for sequence generation too.

Hi. The calculations in details in the link:

https://iopscience.iop.org/article/10.1088/1757-899X/745/1/012039/pdf