A network coding approach to cooperative diversity

This paper proposes a network coding approach to cooperative diversity featuring the algebraic superposition of channel codes over a finite field. The scenario under consideration is one in which two "partners" - Node A and Node B - cooperate in transmitting information to a single destination; each partner transmits both locally generated information and relayed information that originated at the other partner. A key observation is that Node B already knows Node A's relayed information (because it originated at Node B) and can exploit that knowledge when decoding Node A's local information. This leads to an encoding scheme in which each partner transmits the algebraic superposition of its local and relayed information, and the superimposed codeword is interpreted differently at the two receivers - i.e., at the other partner and at the destination node - based on their different a priori knowledge. Decoding at the destination is then carried out by iterating between the codewords from the two partners. It is shown via simulation that the proposed scheme provides substantial coding gain over other cooperative diversity techniques, including those based on time multiplexing and signal (Euclidean space) superposition.