INFORMATION THEORY AND CODING LDPC CODES

PREPAIRED BY PUNIT DADHICH ECE - B

CONTENTS
Introduction Representation of LDPC codes Regular and irregular LDPC codes Encoding of LDPC codes Decoding of LDPC codes Advantages Disadvantages Applications

Low-Density Parity-Check Codes

LDPC codes were originally invented by Robert Gallager in the early 1960s but were largely ignored until they were rediscovered in the mid-1990s by MacKay. Low-Density Parity-Check (LDPC) codes are a class of linear block codes . The name comes from the characteristic of their paritycheck matrix which contains only a few 1s in comparison to the amount 0s. Can perform within 0.0045 dB of Shannon limit.

Representations for LDPC codes

There are two different possibilities to represent LDPC codes. Like all linear block codes they can be described via matrices. The second possibility is a graphical representation.

Matrix representation :
Lets look at an example for a low-density parity-check matrix first.

The matrix defined below is a parity check matrix with dimension k x n for a (8, 4) code.

CONTI
We can now define two numbers describing these matrix wr and wc. wr is the number of 1s in each row and wc for the columns. For a matrix to be called low-density the two conditions wc << n and wr << m must be satisfied.

Tanner introduced an effective graphical representation for LDPC codes. These graphs provides a complete representation of the LDPC codes and they also help to describe the decoding algorithm. Tanner graphs are bipartite graphs. That means that the nodes of the graph are separated into two distinctive sets and edges are only connecting nodes of two different types.

Graphical Representation :

CONTI There are two classes of nodes:

Variable-nodes: Correspond to bits of the codeword or equivalently, to columns of the parity check matrix. There are n v-nodes Check-nodes: Correspond to rows of the parity check matrix. There are m = n-k c-nodes

Example :

It consists of m check nodes and n variable nodes . Check node fi is connected to variable node cj if the element hij of H is a 1.

Regular and irregular LDPC codes

A LDPC code is called regular if wc is constant for every column regular and wr = wc (n/k) is also constant for every row. In the previous example matrix is regular with wc = 2 and wr =4 Its also possible to see the regularity of this code while looking at the graphical representation. There is the same number of incoming edges for every v-node and also for all the c-nodes. If H is low density but the numbers of 1s in each row or column arent constant the code is called a irregular LDPC code.

Encoding LDPC Codes

A codeword c is generated by the matrix multiplication c = mG, where m is the k-bit long message and G is a k by n generator matrix

A common method for finding G from H is to first make the code systematic by adding rows and exchanging columns to get the H matrix in the form H = [PT I] Then G = [I P] The parity check matrix H is a n-k by n matrix of ones and zeros, such that if c is a valid codeword then, cHT = 0

Decoding of LDPC codes

The most common decoding algorithm are the belief propagation algorithm , the message passing algorithm and the sum-product algorithm.

Message passing algorithm :

Sent data word = 10010101

Received code word (C) = 11010101

DECODING (2)
STEP 1 : All v-nodes cj send a message to their cnodes fi. That means for example , c0 sends a massage containing 1 to f1 and f3 , c1 sends massage containing 1 to f0 and f1 and so on.

STEP 2 : Every fi calculate a response to every

connected cj. If we look at the example , every fi is connected to 4 cj node. So a fi looks at the massage received from three cj nodes and calculate the bit that the fourth cj should have in order to fulfill parity check equation.

DECODING (3)

DECODING (4)
STEP 3 : The cj nodes receive the massage from the fi and use this additional information to decide if their originally received bit is OK. In our example , each cj has three sources of information concerning its bits . The original bit received and two suggestions from the fi nodes. Now the cj can send another massage with their decision for the correct value to the fi nodes.

DECODING (5)

ADVANTAGES
Their main advantage is that they provide a performance which is very close to the capacity for a lot of different channels and also a lower error rate. Means they have near capacity error correction performance. Decoding algorithms are simple due to less number of ones.

. Shannon's theory tells us that "long" and "random" codes achieve capacity. The problem people were having was how to decode these "long" and "random" codes. LDPC codes provide the solution.

DISADVANTAGES
The main disadvantages is that encoders are somehow more complex The code length has to be rather long to yield good results.

APPLICATIONS
In satellite transmission of digital TV In 10GBase-T Ethernet Also used in Wi-Fi 802.11 In CMMB (China Multimedia Mobile Broadcasting)