• new1
  • new2
  • new4
  • new5

Church Units

Basically INTERSYMBOL INTERFERENCE deals with symbols (inform of 1 and 0) smearing or interfering with each other in a transmitted signal which becomes distorted when it gets to the receiving end due to a process known as FADING.
The signal gets distorted within the CHANNEL in a communication setup before it finally gets to the receiving end. The design of the transmitters and receivers depends on the assumption that the channel transfer function must be known. But, in most of the digital communications applications, the channel transfer function is not usually known.
The only solution to preventing this intersymbol interference is through the use of EQUALIZERS, now what the equalizer does it tracks the channel characteristics by the use of some algorithms which I am studying, examples of such algorithm includes LMS (Least Mean Square) ALGORITHM, RLS (Recursive Least Square) ALGORITHM and CMA (Constant Modulus) ALGORITHM. All this are found within the communication tool box of MATLAB.
Now the above ALGORITHMS all have their weaknesses, the first 2 uses a training signal which is usually known in advance by the receiver this training signal uses extra bandwidth which is an extra disadvantage in that bandwidth is used. This brings me to using an ALGORITHM which is the last one that doesn’t use any training signal instead it uses the characteristics of the incoming signal to detect the channel hence saving bandwidth.
The main aim of this project is to get an ALGORITHM which will read the fast varying channel hence converging fast and at the same time having a LARGE SNR (SIGNAL to NOISE RATIO).
NOW, the LMS ALGORITHM which uses a training signal is the simplest so far but it converges SLOWLY.
The next one which is RLS also uses a training signal has a fast convergence speed but it is unstable in nature.