The two source nodes, 1 and 2, have messages A and B that need to be transmitted to the two destination/sink nodes, 5 and 6, which need both A and B to be delivered. Each edge in this network can only carry a single value. If nodes could only route/retransmit, then the central link would only be able to carry message A or message B, but not both of them at the same time. Assuming we sent message A through the center, then node 5 would receive message A twice and would never receive message B. The same problem would occur at node 6 if we sent message B through the center node, with node 6 never being able to receive message A. We can immediately see here that routing would be insufficient since no routing scheme can simultaneously transmit A and B to both sink nodes. This is where the operation on data at relay nodes come into play; a simple linear code is demonstrated here, where A and B are encoded using their sum. As such, in this example node 5 would receive both A and A+B, from which it can decode B by subtracting these two values. Node 6 would utilize the same procedure to decode A after receiving both A and A+B. From this simple example we can see that various other encoding techniques could be
…show more content…
Linear Network Coding (LNC) computational complexity makes it unsuitable for practical use in devices that operate on battery power, such as mobile phones and wireless sensors. The triangular pattern based packet coding scheme is performed in two stages. First, redundant “0” bits are selectively added at the head and tail of each packet to make sure that all packets are of uniform bit length. The packets are then XOR-coded bit by bit where the “0” bits are added in such a way that they generate a triangular pattern, known as triangularization, as shown in fig.
Linear Network Coding (LNC) computational complexity makes it unsuitable for practical use in devices that operate on battery power, such as mobile phones and wireless sensors. The triangular pattern based packet coding scheme is performed in two stages. First, redundant “0” bits are selectively added at the head and tail of each packet to make sure that all packets are of uniform bit length. The packets are then XOR-coded bit by bit where the “0” bits are added in such a way that they generate a triangular pattern, known as triangularization, as shown in fig.