NR-5G, Polar vs LDPC Coding Techniques
From the first (1G) to the fifth (5G) generation of mobile communications, it has been challenging to provide quick and secure communication because data transfer happens in a channel environment with noise from amplification, distortion, and other impairments.
In order to achieve quick communication with low error probability, channel coding is essential, and selecting the right channel coding scheme is a difficult and important task.
In 5G there are two codes are getting used:-
The very first question comes to our mind is why polar is selected for the control channel and LDPC for the shared channel
Shannon showed that error-free communication over a noisy channel is possible if the information transmission rate is below or equal to a specific bound i.e the?Channel Capacity bound.?Since then efforts were put into finding the different channel coding techniques to get the rate closer to the channel capacity.
The three main 5G usage scenarios are?enhanced mobile broadband (eMBB), ultra-reliable and low latency communications (URLLC), and massive machine-type communications, according to the 3GPP technical report 38.913 [1]. In comparison to a 4G system, these scenarios call for increased throughput, latency, and reliability. In addition, 5G channel codes should allow hybrid automated repeat request (HARQ) for user data and variable code rate and length for both control information and user data, comparable to a 4G system.
Based on the aforementioned requirements, polar coding has been used for control information whereas LDPC coding has been used for user data in the eMBB scenario as well as the release-15 scope of the URLLC scenario, which focuses on low latency in 5G cellular communications.
领英推荐
5G deployment scenarios, especially the eMBB case, require the support of high throughput which can be up to 20 Gbps, and the encoding and decoding process of 5G channel codes especially for data need to be designed to handle this. 5G LDPC codes adopt the structure of quasi-cyclic (QC) LDPC codes which naturally enables parallelism in encoding and decoding, and high-throughput encoder and decoder can be realized by such parallelism. 5G channel codes for data should also support HARQ, and 5G LDPC codes are designed to efficiently support incremental redundancy (IR) HARQ. As will be described later, such IR-HARQ design of 5G LDPC codes effectively reduces the size of the encoding and decoding graphs when the operating code rate is high, which also helps the realization of high throughput. Rate compatibility to select an arbitrary amount of transmitted bits from mother code output and a variable code length are other important functionalities of 5G channel codes, and the 5G LDPC design realizing such functionalities will be described later in this paper. Fig. 2(a) describes an encoder/transmitter processing chain of 5G LDPC codes.
Control information is typically transmitted with a smaller amount of information bits and with a shorter codeword length. Control information is also typically transmitted with a lower code rate while good performance in a lower BLER regime is required. 5G polar codes are designed to satisfy these requirements. Adoption of polar codes for control information is reasonable, especially considering low BLER application given the fact that polar codes are analytically shown not to suffer from error floor. While the capacity achievability of polar codes with simple SC decoding is attractive, its finite length performance with SC decoding is shown to be noticeably worse than other channel coding candidates. A well-known solution for such a concern especially for polar codes is list decoding. Simply speaking, successive cancellation list (SCL) decoding tries to overcome the fundamental drawback of the premature decision of SC decoding by keeping multiple decision candidates throughout the SC decoding process, i.e., by not making a firm decision during the SC decoding stage. To further improve BLER performance of SCL decoding, CRC-aided (CA) SCL is considered in which CRC is used not only for error detection but also for error correction. Another concern with SC decoding is its latency issue, and there are several practical solutions to reducing the latency of SC decoding. In addition to such an implementation-specific effort, 5G polar codes adopt distributed CRC to support early decoding termination. Fig. 2(b) describes an encoder/transmitter processing chain of 5G polar codes.
References:
Thanks a lot for reading, if you really like my article you can follow me on?Medium,?Twitter,?Instagram,?ResearchGate