Finite state Markov chain analysis of Hybrid Automatic Repeat request (HARQ) schemes
Professors: Francesca Bassi, Pierre Duhamel
Students: Shu Hao Yang , Alaa Khreis (me)
February – March 2015
First Meeting (6 february 2015)
- Presentation of the context (wireless transmission, ARQ principle) and motivation by cooperative communications by Pr. Bassi.
- Quick panoramic of HARQ (Type I, Type II, chase combining) by Pr. Bassi.
- Quick panoramic view of finite state Markov Chains applied to HARQ problems (example: point-to-point Type I HARQ transmission) by Pr. Bassi.
- Scheduling of the next meeting and reading materials.
Reading Transactions Papers (7 – 14 february 2015)
- Markov Analysis of Selective Repeat Type II Hybrid ARQ Using Block Codes (Leonardo Badia, Marco Levorato, Michele Zorzi)
- Applications of Error-Control Coding (Daniel J. Costello, Joachim Hagenauer, Hideki Imai, Stephen B. Wicker)
- Hybrid ARQ: Theory, State of the Art and Future Directions (Christopher Lott, Olgica Milenkovic, Emina Soljanin)
- Steady State Analysis of Coded Cooperative Networks with HARQ Protocol (Stefano Tomasin, Michele Zorzi)
- A Novel Markov-Chain-Based Method for Throughput Analysis in Truncated Cooperative HARQ Systems (Hualiang Chen, Yueming Cai, Weiwei Yang, Yingbo Hu)
Second Meeting (18 february 2015)
- Discussion about the papers and exchange of information.
- Decision to simulate and analyse HARQ type II chase combining/incremental redundancy and soft decoding/hard decoding.
Conception of the simulation scenario (20 february 2015)
- Simulation 1: A toy example of ARQ schèmes with Hamming Code error checking to compare hard decoding and chase combining.
- Simulation 2: ARQ scheme with Reed-Solomon channel encoding. Comparison of chase combining and incremental redundancy.
- Definition of the performance evaluation metrics.
Coding the simulations (21-26 february 2015)
Third meeting (27 february 2015)
- Evaluation of the simulation output.
- Suggestion to implement ARQ with CRC-24 to to compare hard decoding and chase combining.
Complete analysis of ARQ schemes (28 february – 12 march 2015)
- Implementation of ARQ with CRC-24 and comparison of hard decoding and chase combining.
- Markov state analysis of ARQ schemes.
Fourth meeting (12 march 2015)
- Presentation of a report about ARQ schemes. This report defines the various ARQ schemes, performance evaluation metrics, simulation senarios, Markov state analysis, simulation code, simulation output, evaluation and comparison of the results.
- Evaluation and refinements of the report content.
Simulation of HARQ type I schemes (13 – 19 march 2015)
- Simulation of HARQ with a CRC-24 error correction code and channel coding on top using convolutional codes with Viterbi Decoding. We also study hard decoding and chase combining in this case.
- The evaluation of the output showed an error in the simulation code that was fixed later.
Fifth meeting ( 20 march 2015)
- Hints to fix the simulation code.
- Modeling HARQ type I using Markov Chains.
- Suggestions to start studying cooperative HARQ.
Analysis of HARQ type I schemes and emphasis on cooperative HARQ (21-29 march 2015)
- Completing with success the simulation of HARQ type I and analysis of the output.
- Reading many of the latest publications about cooperative HARQ and documentation with basic analysis of the various approaches in treating cooperative HARQ.
- Preparation of the final project report and presentation.
Final project meeting (30 march 2015)
- Defense of the project.
- Constructive comments about the presentation and discussion of the perspectives of using cooperative HARQ.
By doing this project, I gained valuable knowledge about HARQ and I feel more curious about investigating many ideas concerning cooperative HARQ and finding a suitable model for these complex systems.
* I will share the reports or at least parts of the reports later. The content of the reports helps understanding ARQ and HARQ type I in depth, in addition to presenting cooperative HARQ and research perspectives in this area.