An ordered crossover based approach to designing labeling diversity mappers

Solwa, Shaheen

doi:10.1002/ett.4431

Cited by 4 publications

(12 citation statements)

References 24 publications

(133 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the case of the GA found in [4], LD mapper designs have illustrated significant diversity gains from approximately 3 dB upto 17 dB. The GA found in [8] was able to improve upon the LD values achieved by the GA in [4], but small performance gains of approximately 0.5dB up to 4 dB were achieved. Leveraging on the performance of the aforementioned evolutionary algorithms, another subset of AI, machine learning, can be applied to the LD problem.…”

Section: Introductionmentioning

confidence: 90%

“…The application of meta-heuristic algorithms to the LD problem had produced significant advantages such as reduced complexity [8] and improved error performance [8], [9]. Hence, in this paper the authors' apply another subset of artificial intelligence called machine learning (ML) which is applied to the LD problem to predict the amount of LD achieved.…”

Section: A Motivations and Contributionsmentioning

confidence: 99%

“…The higher the value of φ, the more LD has been achieved. Hence, the end goal of the LD mapper design problem is the maximise the minimum product Euclidean distance between symbols on a constellation [7,8].…”

Section: A Theoretical Error Performance and Mapper Design Of Ustld S...mentioning

confidence: 99%

“…More recently, the authors of [4] and [8] have proposed meta-heuristic algorithms in the form of a genetic algorithm (GA). GAs imitate the process of evolution and survival of the fittest in nature [4].…”

Section: Introductionmentioning

confidence: 99%

“…This is performed over a number of iterations until a good, stable solution is found. A GA can produce a solution of either a local optimal or global optimal, which depends on the 'mating' technique used [8,10]. In the case of the GA found in [4], LD mapper designs have illustrated significant diversity gains from approximately 3 dB upto 17 dB.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Machine-Learning-Based Labelling Diversity Model for Predictive Analysis: Using 16QAM as a Case Study

et al. 2022

Self Cite

View full text Add to dashboard Cite

The recent advancement and enhancement that optimized uncoded space-time labelling diversity (USTLD) have provided significant diversity gains. By adopting the use of evolutionary algorithms, labelling diversity (LD) mapper designs produced are near-optimal in quality. The only disadvantage to the use of evolutionary algorithms is that the produced solution is not always optimal. To ease the calculation of how much a mapper design had achieved LD, this paper proposes a machine learning-based analysis to predict the amount of LD achieved by a mapper. In this paper, only the 16QAM constellation is studied as a simple case. Six machine learning-based algorithms were proposed in this paper, namely multi-linear regression (MLR), support vector regression (SVR), decision trees (DT), random forest (RF), K-nearest neighbours (KNN) and a simple artificial neural network (ANN). From the results obtained from the experiments, it can be seen that the MLR algorithm is the least time complex while the ANN is the most time complex. It is also important to note that the DT and KNN algorithms take a comparatively short amount of time to execute. When compared in terms of machine learning metrics, it was shown that the ANN algorithm performed the best with the least amount of error while the MLR algorithm performed the worst with the highest amount of error. Thus, it could be seen that the results from this paper provide a positive outlook on applying machine learning algorithms to the LD problem.INDEX TERMS Labelling diversity, machine learning, mean square error, neural networks, predictions.

show abstract

Section: Introductionmentioning

confidence: 90%

Section: A Motivations and Contributionsmentioning

confidence: 99%

Section: A Theoretical Error Performance and Mapper Design Of Ustld S...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Machine-Learning-Based Labelling Diversity Model for Predictive Analysis: Using 16QAM as a Case Study

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

A modified global neighborhood algorithm for designing labeling diversity mappers

Solwa

Ayanda

2022

Trans Emerging Tel Tech

Self Cite

View full text Add to dashboard Cite

Genetic algorithms (GAs) have been recently applied to uncoded space‐time labeling diversity systems (USTLD) to improve upon diversity in wireless links. GAs were used to optimize the secondary mappers that are used to encode the information to be transmitted. Existing literature only achieved a local optimal design, at the fraction of the computational costs. Hence, this article focuses on improving upon the GAs for USTLD systems by merging it with a global‐local search technique, called the global neighborhood algorithm (GNA). The GNA algorithm was tested on M‐QAM, M‐PSK, M‐APSK, and asymmetric M‐APSK constellations of modulation order M =16,32$$ =16,32 $$, and 64$$ 64 $$. In the case of M‐QAM constellations, the GNA was able to match, but not improve upon existing mapper designs. In the case of the 64PSK constellation, the GNA was able to produce mapper designs that improved upon the error performance of existing mapper designs by approximately 1 dB. In the case of the 32APSK constellation, the GNA was able to improve upon the error performance of existing mapper designs by approximately 0.5 dB. Moreover, the GNA produced mapper designs for the 11+5APSK and asymmetric 16APSK constellations that improved upon the error performance of existing mapper designs by approximately 0.5 and 4 dB respectively. Finally, the computational complexity of the proposed GNA algorithm was analyzed and the algorithm was shown to be Ofalse(normalM2+Pm,1+Pm,2false)$$ O\left({\mathrm{M}}^2+{P}_{m,1}+{P}_{m,2}\right) $$ which is slightly more complex than the initial GA approach with complexity Ofalse(normalM2false)$$ O\left({\mathrm{M}}^2\right) $$, but significantly less complex than the enhanced GA with complexity Ofalse(normalM!false)$$ O\left(\mathrm{M}!\right) $$. The increased complexity also came with significant gains such as improved LD mapper designs.

show abstract

A multi-optimization-based genetic algorithm mapper design for three transmit uncoded space-time diversity systems

Solwa

Bamisaye

Ayanda

2022

Int. J. Model. Simul. Sci. Comput.

View full text Add to dashboard Cite

Three transmit antennas Uncoded Space-Time Labeling Diversity (USTLD) systems were recently proposed as a direct extension to the two transmit antennas USTLD system. By employing more antennas at the transmitter, antenna — as well as — labeling diversity (LD) can be further achieved. Existing mapper designs for three transmit antennas USTLD systems are based on symmetrical heuristics, and hence no asymmetrical mapper designs can be produced. This paper proposes a multi-function optimization genetic algorithm for the three transmit antennas USTLD system that produces mapper designs irrespective of size or symmetry. The genetic algorithm was tested on 16-, 32QAM, 16-, 32APSK and 16APSK constellations that do not exhibit diagonal symmetry. The proposed genetic algorithm had produced mapper designs that match but did not improve upon existing mapper designs for 16QAM and 32QAM, respectively. The 16APSK and 32APSK mapper designs produced achieved diversity gains of [Formula: see text] and [Formula: see text] when compared to existing third mapper designs, as well as [Formula: see text] and [Formula: see text] gain over the two transmit antennas USTLD scheme at the BER of [Formula: see text]. Furthermore, when tested on 16APSK constellations that do not exhibit diagonal symmetry, these mappers achieved diversity gains between [Formula: see text], [Formula: see text] and [Formula: see text] over the two-transmit USTLD scheme at the BER of [Formula: see text], respectively. The computational complexity of the proposed genetic algorithm was analyzed based on a previous study, which was shown to be [Formula: see text].

show abstract

An ordered crossover based approach to designing labeling diversity mappers

Cited by 4 publications

References 24 publications

A Machine-Learning-Based Labelling Diversity Model for Predictive Analysis: Using 16QAM as a Case Study

A Machine-Learning-Based Labelling Diversity Model for Predictive Analysis: Using 16QAM as a Case Study

A modified global neighborhood algorithm for designing labeling diversity mappers

A multi-optimization-based genetic algorithm mapper design for three transmit uncoded space-time diversity systems

Contact Info

Product

Resources

About