Stochastic Bit Loading Algorithm for Multi-Carrier Modulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bit loading algorithms for multi-carrier modulation systems, such as HomePlug AV, face challenges in efficiently exploiting error-correction capabilities of turbo codes while maintaining a low bit error rate, often resulting in reduced throughput due to high computational complexity and inability to adapt to uniform power allocation constraints.
Innovation Solution
The Jump Probability Computation (JPC) algorithm introduces a stochastic approach that dynamically switches between digital transmission constellations based on signal-to-noise ratio, allowing for increased throughput by probabilistically determining the use of higher bit constellations while maintaining a pre-established bit-error rate, thereby efficiently exploiting error-correction capabilities with reduced complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If deterministic bit loading algorithms are used to maintain low bit error rate, then reliability is improved, but throughput is reduced due to inability to exploit error-correction capabilities
Solution Approach 1:
The patent applies dynamics by transitioning from deterministic bit loading to a stochastic algorithm that probabilistically selects between different digital transmission constellations. The system dynamically adjusts the probability of using higher-order constellations based on channel conditions and error-correction capabilities, allowing throughput to increase while maintaining reliability through controlled randomness rather than fixed deterministic rules
Solution Approach 2:
The patent changes the fundamental parameter of bit loading from a deterministic value to a probability distribution. Instead of assigning a fixed number of bits per carrier, the system uses a stochastic algorithm that determines the probability of using different constellation sizes (e.g., QPSK, 16-QAM, 64-QAM), thereby changing the state space from discrete fixed values to continuous probability values that can be optimized for both throughput and error rate
2Productivity
If higher order digital transmission constellations are used to increase throughput, then productivity is improved, but bit error rate increases beyond acceptable levels
Solution Approach 1:
The patent implements feedback by using the error-correction capabilities of turbo codes to inform the stochastic bit loading decisions. The system monitors the performance and adjusts the probability distribution of constellation selection based on the actual error rates and channel conditions, creating a closed-loop system where higher-order constellations can be used more aggressively while maintaining reliability through feedback-driven probability adjustment
Solution Approach 2:
The patent applies partial or excessive action by allowing the system to occasionally use higher-order constellations that would normally exceed the bit error rate threshold, relying on the stochastic nature and error-correction codes to handle the excess errors. This enables the system to push beyond traditional deterministic limits and achieve higher average throughput while maintaining acceptable overall error rates
3Productivity
If stochastic algorithm is used to increase throughput, then productivity is improved, but device complexity increases due to probabilistic constellation selection
Solution Approach 1:
The patent applies segmentation by dividing the stochastic bit loading process into discrete, manageable steps: identifying the first digital transmission constellation with the greatest signal-to-noise ratio, identifying a second constellation with immediately greater bits per symbol, and determining probabilities for each. This segmented approach breaks down the complex stochastic optimization into a series of simpler, more manageable operations that reduce overall computational complexity
Data Source
AI summary
A method of transmitting symbols of a digital transmission constellation from a set thereof, ordered from a smallest to a greatest number of bits per symbol, may include identifying a first constellation from the set that is configured to communicate with a threshold error rate and has a greatest signal-to-noise ratio smaller than a signal-to-noise ratio of a received signal. The method may also include identifying a second constellation from the set that corresponds to a constellation with a number of bits per symbol immediately greater than the first constellation. The method may further include determining first and second probabilities of use of the first and second constellations that would generate an expected number of erroneous bits corresponding to the threshold error rate. The method may further include transmitting a symbol with a constellation selected randomly between the first and second constellations according to the first and second probabilities, respectively.