Receiver Algorithm Trainable Parameter Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communication systems face challenges in optimizing trainable parameters of receivers to minimize loss functions effectively, especially when error correction codes are used, due to the complexity of the optimization problem and the lack of access to transmitted coded bits.
Innovation Solution
The system employs a receiver algorithm with trainable parameters that generates refined estimates of coded data using error correction codes, calculates loss functions, and iteratively updates these parameters using gradient descent or reinforcement learning until convergence conditions are met, ensuring accurate decoding and parameter optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If trainable parameters are used in receiver algorithms to improve communication system performance, then accuracy of data bit estimation is improved, but device complexity increases
Solution Approach 1:
The patent applies parameter changes by introducing trainable parameters into the receiver algorithm that can be optimized through iterative training. The receiver algorithm's parameters are adjusted based on loss function calculations comparing estimated coded data with refined estimates, enabling the system to adapt parameters for improved performance without redesigning the entire receiver architecture.
2Reliability
If iterative parameter updates are performed to minimize loss function, then block error rate is reduced, but loss of time increases
Solution Approach 1:
The patent implements periodic action through iterative parameter updates where the receiver algorithm repeatedly processes the same received data to generate updated estimates. Each iteration involves decoding, re-encoding, loss calculation, and parameter updates, creating a periodic refinement cycle that continues until convergence or maximum iterations are reached, thereby reducing block error rates through systematic improvement.
Solution Approach 2:
The patent applies preliminary action by performing parameter training offline before actual communication operations. The trainable parameters are optimized using training data and iterative updates in advance, so that during actual communication, the receiver can directly use the pre-optimized parameters without performing iterative updates in real-time, thus reducing time loss during operational phases.
3Reliability
If error correction codes are used to improve data transmission reliability, then block error rate is reduced, but device complexity increases
Solution Approach 1:
The patent applies universality by designing a receiver algorithm that performs multiple functions: it decodes error correction codes, generates estimates of coded data, and simultaneously trains its parameters through iterative optimization. This multi-functional approach consolidates what could be separate complex subsystems into a unified algorithm that handles both error correction and adaptive optimization, managing complexity through functional integration.
4Measurement precision
If refined estimates are generated through encoding of estimated data bits, then measurement precision is improved, but productivity decreases
Solution Approach 1:
The patent applies copying by creating a refined estimate of coded data through re-encoding the estimated data bits. Instead of directly using the estimated coded data from the receiver, the system decodes it to data bits, re-encodes these bits back to coded data format, and uses this refined copy for training purposes. This copying process enables accurate loss function calculation while maintaining the original received data for actual communication operations.
Data Source
AI summary
An apparatus, method and computer program is described comprising receiving data at a receiver of a transmission system; using a receiver algorithm to convert data received at the receiver into an estimate of the first coded data, the receiver algorithm having one or more trainable parameters; generating an estimate of first data bits by decoding the estimate of the first coded data, said decoding making use of an error correction code of said encoding of the first data bits; generating a refined estimate of the first coded data by encoding the estimate of the first data bits; generating a loss function based on a function of the refined estimate of the first coded data and the estimate of the first coded data; updating the trainable parameters of the receiver algorithm in order to minimise the loss function; and controlling a repetition of updating the trainable parameters by generating, for each repetition, for the same received data, a further refined estimate of the first coded data, a further loss function and further updated trainable parameters.