Polar Code Construction Using Precomputed Index Arrays in 5G
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing polar code construction methods for 5G new radio standards are computationally inefficient, leading to high CPU cycle requirements that are not suitable for the shorter slot time durations in 5G networks, as they involve complex calculations and high computational complexity.
Innovation Solution
Pre-calculating two one-dimensional index arrays, an input bits index array sorted by channel quality and a sub-block interleaving array, to determine frozen bit positions and assign information, CRC, and PC bits efficiently during polar encoding and decoding, allowing for faster polar code construction and decoding processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional polar code construction methods are used, then code performance meets requirements, but computational complexity is high and CPU cycles are excessive
Solution Approach 1:
The patent pre-calculates and stores the bit-channel quality ordering (array Q) and frozen bit positions (array Af) during code construction phase. These pre-computed arrays are then directly used during encoding without re-calculation, transforming the complex real-time computation into a simple array lookup operation that runs forty times faster
Solution Approach 2:
The patent creates simplified copy representations of the complex channel quality calculations in the form of lookup arrays Q and Af. Instead of performing complex computations during encoding, the system copies the essential information into pre-computed arrays that can be quickly referenced, reducing computational complexity while preserving code performance
2Manufacturing precision
If complex polar code construction calculations are performed in real-time, then accurate code construction is achieved, but the processing time exceeds 5G slot duration requirements
Solution Approach 1:
The patent moves the computationally intensive calculations of bit-channel quality ordering and frozen bit positions to a preliminary offline phase. The results are stored in arrays Q and Af that are then directly applied during real-time encoding, ensuring construction accuracy while reducing latency to meet 5G slot duration requirements
Solution Approach 2:
The patent extracts the time-consuming computational elements (channel quality calculations and frozen bit determination) from the real-time encoding process and places them in a separate preliminary computation phase. This separation allows the main encoding operation to proceed quickly using pre-computed data, meeting strict timing requirements
Data Source
AI summary
According to some embodiments, a method for polar encoding includes obtaining an input bits index array Q, wherein each element Q[i] is an index of a polar coded transmission channel and corresponds to an input U[i] of a polar encoder. The elements of Q are ordered according to their associated channel quality. The method further includes obtaining an integer number X of bits for polar encoding and wireless transmission, wherein X is not greater than N. Upon determining the polar code block length N is not greater than the number of rate matched bits M available for transmission, the method includes assigning each of the bits as inputs to the polar encoder ordered according to the input bits index array Q.