Modern Channel Coding for Wireless Communications
Lecturer: Dr.-Ing. Marc Adrat
Type: Part of a master program
Course language: English
Lecture notes and exercise problems are all in English and are sold in the first lecture as well as by Irina Ronkartz.
from Thursday, April 4, 2019
08:30 - 10:00
Lecture room 4G
from Thursday, April 11, 2019
10:15 - 11:00
Lecture room 4G
Dates will be announced.
The lecture on "Modern Channel Coding for Wireless Communications" addresses the latest Forward Error Correction (FEC) techniques for present and next generations of wireless communications.
After a brief review of classical approaches to FEC, further insights are presented on the information theoretical limits. For this purpose and various relevant scenarios, the minimum required signal-to-noise ratio is determined for allowing a theoretically error-free transmission.
In this lecture, several practical applications for advanced FEC techniques are discussed which provide error-correcting capabilities close to these information theoretical limits.
As the first advanced channel coding technique, Turbo Codes are introduced since they are used e.g. in 3G, 4G and 4.5G applications like UMTS, LTE Advanced Pro, and IEEE 802.16 WiMAX. Turbo Codes achieve their performance objectives by iteratively exchanging the so-called extrinsic information between two (or more) constituent decoders.
In addition, Extrinsic Information Transfer (EXIT) Charts are discussed which allow to visualize the stepwise improvements in extrinsic information by the iterations. These charts help on one hand to understand the convergence behavior of Turbo Codes and on the other hand to design powerful Turbo Coding schemes.
Other related applications of the Turbo Principle are also introduced using Bit-Interleaved Coded Modulation with Iterative Decoding (BICM-ID) as an example.
Following Turbo Codes, another advanced channel coding technique presented is the Low Density Parity Check (LDPC) Code. Such codes are used e.g. in DVB-S2 and IEEE 802.11n WiFi. LDPC Codes are a special class of Linear Block Codes which can also be decoded by iterative algorithms similar to the Turbo Decoding approach.
At the end of this lecture, the concept of Polar Codes is introduced. Polar Codes transform the channel in such a way that on the information side there is a clear separation between reliable and unreliable channel usage. Polar Codes are a promising candidate for the 3GPP New Radio (5G).
The results of the evaluation are summarized below.