- Bio-inspried Wireless
- Coding and Modulation
- Cooperative Comms
- Joint algorithm/hardware design
- Machine-to-Machine Communication
- MIMO Transceivers
- Multiuser Comms
- Optical Wireless Communications
- Quantum Communications
- Signal Processing
- Source Coding
- Wireless Networking
- Wireless Video Communications
- Millimeter Wave Communication

## You are here

# Plug for RobProb on the comms group CVS

Here's my first blog entry. Well done to Jos and Andy for putting together an excellent resource that will no doubt promote and assist collaborative research no end!

I'm going to use this opportunity to plug my RobProb code on the group's CVS. This well documented code provides data types for both normal- and logarithmic-domain probabilities and probability ratios. More specifically, there are data types for logarithmic domain L-values and Logarithmic Likelihood Ratios (LLRs), along with their normal-domain equivalents. These data types are capable of representing infinite values (which are associated with absolute confidences) without the need for clipping at some arbitrary large value. Furthermore, this code provides IT++ vectors of these data types to store frames of LLRs, for examples. Conversion functions both to and from the double and vec data types are provided to retain compatibility with existing code.

The RobProb code provides Jacobian and addition operators for use in the logarithmic domain, with approximate, lookup-table-aided and exact versions of the Jacobian operator available for use in the MaxLogMAP, LogMAP and MAP algorithims respectively, for example. The corresponding addition and multiplication operators are provided for use in the normal domain, although it should be noted that behind the scenes, all calculations are performed in the logarithmic domain to avoid the precision problems that are often associated with working in the normal domain. Furthermore, the number of Add-Compare-Select (ACS) operations performed when using these operators is automatically counted, allowing the assessment of your decoder's computational complexity.

Functions are provided for use with EXtrinsic Information Transfer (EXIT) charts, providing the ability to generate *a priori* information with a particular mutual information and to measure extrinsic mutual information. The mathematic method of measuring extrinsic mutual information can be employed in both the binary and non-binary case with arbitrary source entropy. Additionally, a historgram method for measuring the extrinsic mutual information in the binary case is provided. Unlike other EXIT chart code available within the group, the number of histrogram bins and their sizes are determined automatically in the RobProb code, making the use of the histogram method simple. For decoders that are not truely *A Posteriori* Probability (APP) decoders, such as if the MaxLogMAP or LogMAP algorithm is chosen over the MAP algorithm, the histogram method of measuring extrinsic mutual information can be more accurate than the mathematic method.

Take care all, Rob.