Project

Back to overview

Energy Proportional Decoding of Polar Codes for 5G with Variable Complexity Algorithms

English title Energy Proportional Decoding of Polar Codes for 5G with Variable Complexity Algorithms
Applicant Burg Andreas Peter
Number 175813
Funding scheme Project funding (Div. I-III)
Research institution Laboratoire de circuits pour télécommunications EPFL - STI - IEL - TCL
Institution of higher education EPF Lausanne - EPFL
Main discipline Information Technology
Start/End 01.02.2018 - 31.01.2019
Approved amount 124'164.00
Show all

All Disciplines (2)

Discipline
Information Technology
Electrical Engineering

Keywords (4)

Polar Codes; Integrated Circuits; VLSI design; Channel Coding

Lay Summary (German)

Lead
Systeme zur drahtlosen Kommunikation verwenden digitale Kodierung zur Vermeidung und zur Korrektur von Fehlern bei der Datenübertragung. Die zweite, dritte und vierte Generation von Mobilfunksystemen (2G, 3G, und 4G) verlässt sich hierbei vor allem auf sogenannte Turbo-codes, die in den vergangenen zwei Jahrzehnten immer weiter verbessert wurden, die aber auch eine Reihe von Schwachstellen aufweisen. Um diese zu beheben, setzt die kommende, fünfte Mobilfunkgeneration eine völlig neue Art von Codes ein. Diese sogenannten Polar Codes versprechen einen besseren Datendurchsatz, mehr Flexibilität und kürzere Latenzen, sind aber auch anspruchsvoller in der Dekodierung. Während die Dekodierung von Turbo-codes auf mehr als 20 Jahren Forschung aufbaut, herrscht in der effizienten und stromsparenden Dekodierung von Polar codes noch Nachholbedarf.
Lay summary
Ziel dieses Projektes ist es, schnelle, effiziente und vor allem stromsparende Dekoder für Polar codes zu entwickeln. Dabei konzentrieren wir uns vor allem auf Anwendungen im Bereich der fünften Mobilfunkgeneration. Ein Schwerpunkt unserer Forschung liegt dabei darin, den Aufwand zur Dekodierung (und damit den Energieverbrauch) den Gegebenheiten des Kanals anzupassen und somit sicher zu stellen, dass nur soviel Energie verwendet wird, wie absolut notwendig ist. Dies ermöglicht zum Beispiel Kompromisse zwischen Datenrate und Energieverbrauch und führt zu einer Verlängerung der Batterielaufzeit. Hierzu kommen Algorithmen und Ideen zum Einsatz die bereits in vorangehenden Projekten im Ansatz entwickelt wurden. Ziel dieses Projektes ist es nun die Algorithmen weiter zu verbessern und entsprechende Architekturen und Integrierte Schaltungen zu entwerfen.
Direct link to Lay Summary Last update: 09.05.2018

Responsible applicant and co-applicants

Employees

Publications

Publication
On the computational complexity of blind detection binary linear codes
Balatsoukas-StimmingAlexios, Filos-RatsikasAris (2019), On the computational complexity of blind detection binary linear codes, in Proc. of ISIT, Paris, FranceIEEE, New York, USA.

Scientific events

Active participation

Title Type of contribution Title of article or contribution Date Place Persons involved
Special Session on Polar codes at WCNC 2018 Talk given at a conference Implementation Aspects of Polar Codes 15.04.2018 Barcelona, Spain Burg Andreas Peter; Balatsoukas Stimming Alexios Konstantinos;
ITA 2018 Talk given at a conference Blind Detection of Polar Codes 16.02.2018 San Diego, United States of America Burg Andreas Peter;


Associated projects

Number Title Start Funding scheme
169443 Gbps Decoders for Polar Codes in 5G Networks 01.10.2016 Project funding (Div. I-III)
149447 Efficient Application Specific Integrated Circuits for Decoding Polar Codes 01.05.2014 Project funding (Div. I-III)

Abstract

Forward-error-correction (channel) coding is an essential ingredient of all modern communication systems. Decoders of such codes are often one of the most computationally intensive and power hungry task in modern wireless receivers. One of the latest innovations in channel coding are polar codes, a new type of codes that can provably achieve the capacity of practically relevant channels. The widely-recognized disruptive impact of these codes is illustrated by the recent, rapid adoption of this young class of codes as part of the new 5G communication standard under development by 3GPP. Besides the communication-performance benefits of polar codes, an important ingredient to their success has been the recent progress in the design of advanced decoding algorithms and corresponding decoders with manageable complexity and power consumption. Nevertheless, compared to the achievements of 20+ research years in efficient decoder design for the competing LDPC and turbo codes, polar decoders are still in their infancy, lagging behind in terms of area and energy efficiency. An important and fundamental reason for the limited (area and energy) efficiency of today’s polar decoders is deeply rooted in the nature of close-to-optimal decoding strategies which are key to achieve the best error-correction performance. These advanced algorithms explore multiple hypotheses to maximize the chance for success (i.e., finding the correct code word). Unfortunately, in order to always meet expectations on sampling rate, throughput, and latency, all of these hypotheses must typically be considered in parallel. In practice, this leads to severe overhead, notably in terms of area and power consumption, which is unnecessary under the most frequent average operating conditions. Most of the resources are wasted on calculations for results that will, in the end, be discarded. This project builds upon an initial 2-year project that was followed by a subsequent (ongoing until 31 Sept. 2017) project. In the first project, we laid the foundations for the implementation of efficient high-speed polar decoders by pioneering key algorithmic optimizations and architectural templates for SC and list-based decoding. In the second short project, we improved the flexibility of that technology, and conducted a careful study to compare polar codes to LDPC and turbo codes. We also identified some of the key weaknesses of current polar decoding strategies in terms of energy efficiency. In this third project phase, we are proposing a 12-month research effort to attack these weaknesses (i.e., the fundamental lack of efficiency and waste of resources) in today’s polar decoders designed to always match system requirements in throughput and latency. To this end, we first study the dynamics of having a component with variable execution time in the context of a complete communication system with stringent deadlines and how to overcome these deadlines. Second, we work on improving a recent promising decoding algorithm for polar codes that received little attention thus far because of its inherent variable execution time. Third, we fabricate a chip that implements this new decoding algorithm to demonstrate that we can live with uncertainty in the computational load to save energy, while causing almost no impact to communication system performance with respect to critical latency constraints, error rate, and throughput.
-