Error Control Code Combining Techniques in Cluster-based Cooperative Wireless Networks

1. Error Control Code Combining Techniques in Cluster-based Cooperative Wireless Networks Su Yi Babak Azimi-Sadjad Shivkumar Kalyanaraman Vijaynarayan Subramanian Presenter: Jayasri Akella Rensselaer Polytechnic Institute

2. Outline • Introduction and Related work • Performance Analysis for Link Layer Cooperation • Simulations • Conclusions and Future Work Rensselaer Polytechnic Institute

3. Introduction • Objective: to improve the overall channel quality/ throughput for each transmitter/receiver pair • Method: link layer cooperation scheme for multi-hop wireless networks and sensor networks • Advantage: extract diversity gain out of the redundancy inherently present in all broadcast network transmission Rensselaer Polytechnic Institute

4. Previous Work: Physical Layer Cooperation • The information source reaches the first relay cluster. • The nodes in the relay cluster share their information for diversity gain. Then they relay the information to the next cluster. • The next cluster has a reliable channel with the destination node, hence there is no need of physical layer cooperation. A single node can relay the information to the final destination node. Rensselaer Polytechnic Institute

5. Link Layer Cooperation • Stage 1: Cluster head decides if cooperation is necessary • Stage 2: FEC and Code combining among cluster nodes • Stage 3: Use ARQ or transmit diversity if else fail Rensselaer Polytechnic Institute

6. Info pkt Encoded pkt Wireless channel Received pkt Received pkt Received pkt Received pkt Viterbi decoder Estimate of info pkt Code Combining Procedure Rensselaer Polytechnic Institute

7. Code Combining Technique • Combine L repeated packets encoded with a code of rate R • Thus obtain a lower rate R/L and more powerful • Viterbi (maximum-likelihood) decoding • The decoding function: • An alternate way is: where weight for the i th channel Rensselaer Polytechnic Institute

8. Code Combining with Convolutional Codes: An Example • A (3,1,2) code with an information sequence h =3 001 S S 3 3 1 1 0 0 1 0 0 1 1 1 0 0 S S S 1 1 1 1 1 1 0 0 0 1 1 1 S S S 2 2 2 0 0 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 000 000 000 000 000 S S S S S S 0 0 0 0 0 0 r = ( 000 , 000 , 001 , 000 , 000) Rensselaer Polytechnic Institute

9. Code Combining with Convolutional Codes: An Example • A (3,1,2) code with an information sequence h =3 • Cooperative nodes L=3 • Weight for each channel is w1, w2, w3 • If w1=w2=w3, the all zero path is chosen. • If w1=1, w2=2, w3=3, then the highlighted path is chosen. 001 001 001 S S 3 3 1 0 1 1 0 0 0 1 0 0 1 1 1 0 0 1 0 0 0 1 0 0 1 1 1 1 0 1 0 0 1 1 S S S 1 1 1 0 1 1 1 0 0 0 1 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 0 0 0 1 1 1 1 w1 w2 w3 1 S S S 2 2 2 1 1 1 0 1 0 1 1 1 1 1 1 1 1 1 1 1 0 0 1 0 1 1 1 1 1 1 1 1 1 1 1 0 0 1 0 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 S S S S S S 0 0 0 0 0 0 r = ( 000 001 111 , 000 100 101 , 001 010 001 , 000 000 000 , 000 000 000) Rensselaer Polytechnic Institute

10. Code Combining with Convolutional Codes in a Uniform Channel Condition • Bit-error prob of the non-combined code: • Bit-error prob of the L-repeated code: • p – BER for wireless channel • - coefficient of power term in B(X), the bit weight enumerating function (WEF) of the convolutional code • - minimum free distance Rensselaer Polytechnic Institute

11. Code Combining with Different Channel Conditions • Bit-error prob of the L-repeated code: • where • is the coefficient in the generating function of r.v. S: • S is a weighted sum of the received sequence Rensselaer Polytechnic Institute

12. Simulations • A simple network topology L = 4 50 or 100 m 250 m Sender Cluster head Rensselaer Polytechnic Institute

13. Link Layer Decoding Performance Decoded bit-error rate Pb vs. number of cooperative nodes L. PL is the amount of power deduction of the intra-cluster transmission upon the inter-cluster transmission. Decoded bit-error rate Pb vs. number of cooperative nodes L with different cluster radius. Smaller cluster radius has a better performance. Rensselaer Polytechnic Institute

14. Energy Consumption Aggregate energy consumption vs. number of cooperative nodes or packet repeats L. A decoded bit-error rate Pb=10-7 is fixed. SNR vs. number of cooperative nodes L. With a fixed objective Pb, the required SNR decreases with the increase of the cluster size L. Rensselaer Polytechnic Institute

15. Conclusion and Future Work • Cooperation architecture is effective in improving the link performance and reducing the energy consumption • Less power leads to less interference among nodes, thus can improve the capacity of the wireless networks. • Future work on designs which explicitly exploit physical layer, data link layer, and network layer cooperation among nodes • These designs include: • cooperation-intended cluster-based routing • medium access issues in the intra-cluster communications • network performance from all aspects • more information theoretic analysis of the coding technique and network capacity. Rensselaer Polytechnic Institute

16. Thank you! • For more information: yis@rpi.edu Rensselaer Polytechnic Institute