A fully customizable hardware implementation for general purpose genetic algorithms

Peker, Murat

doi:10.1016/j.asoc.2017.09.044

Cited by 15 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since most of the existing hardware implementations are complete GA algorithm [9,11,12], they reported the quality of solutions and the search speed as performance metrics, so, there is no crossover performance evaluation individually and independently. Whereas detailed information of crossover module is presented only in Reference [5], therefore, we can only compare our simulation results with the reported results of that work.…”

Section: Experimental Results and Comparisonmentioning

confidence: 99%

See 1 more Smart Citation

FPGA Implementation of Crossover Module of Genetic Algorithm

Attarmoghaddam

Kanan

2019

Information

View full text Add to dashboard Cite

This paper proposes a hardware realization of the crossover module in the genetic algorithm for the travelling salesman problem (TSP). In order to enhance performance, we employ a combination of pipelining and parallelization with a genetic algorithm (GA) processor to improve processing speed, as compared to software implementation. Simulation results showed that the proposed architecture is six times faster than the similar existing architecture. The presented field-programmable gate array (FPGA) implementation of PMX crossover operator is more than 400 times faster than in software.

show abstract

Section: Experimental Results and Comparisonmentioning

confidence: 99%

“…In Reference [10], the adaptive GA is designed on FPGA, based on modular design that supports two-point crossover. Figure 1 shows two crossover methods provided in Reference [11], named as one-point and multi-point crossover.…”

Section: Existing Work In Hardware Implementation Of Crossover Modulementioning

confidence: 99%

FPGA Implementation of Crossover Module of Genetic Algorithm

Attarmoghaddam

Kanan

2019

Information

View full text Add to dashboard Cite

show abstract

“…Search heuristics have also been deployed in silica. Genetic algorithms (GA) have been deployed to field-programmable gate arrays (FPGA) for hardwarebased optimization [9], enabling rapid prototyping of fast, low-power search. Energy management is another concern for metaheuristics, specifically those involving smart power grids [5].…”

Section: Real-world Systemsmentioning

confidence: 99%

“…To focus on execution time, we intentionally did not introduce parallelism or distributed processing, however such a procedure can be used on a microcomputer (e.g., Python's multiprocessing package). Figure 1(a) compares the number of generations required for the algorithm to converge and Figure 1(b) compares the amount of time (seconds) required to reach convergence between a current-generation laptop 9 and the Raspberry Pis. As can be seen from these figures, there exists no difference in the number of generations required to converge to the optimal solution.…”

Section: Search@home Overviewmentioning

confidence: 99%

Search@Home: A Commercial Off-the-Shelf Environment for Investigating Optimization Problems

Fredericks¹,

Moore²

2020

Preprint

View full text Add to dashboard Cite

Search heuristics, particularly those that are evaluation-driven (e.g., evolutionary computation), are often performed in simulation, enabling exploration of large solution spaces. Yet simulation may not truly replicate real-world conditions. However, search heuristics have been proven to be successful when executed in real-world constrained environments that limit searching ability even with broad solution spaces. Moreover, searching in situ provides the added benefit of exposing the search heuristic to the exact conditions and uncertainties that the deployed application will face. Software engineering problems can benefit from in situ search via instantiation and analysis in real-world environments. This paper introduces Search@Home, an environment comprising heterogeneous commercial off-the-shelf devices enabling rapid prototyping of optimization strategies for real-world problems.

show abstract

“…In order to evaluate the performance of the proposed CSA system, two of the most common benchmark functions were implemented along with the spherical function as in [49], [50].…”

Section: B System Performance Evaluationmentioning

confidence: 99%

FPGA Implementation of Floating Point Based Cuckoo Search Algorithm

Issa

Ahmed

2019

IEEE Access

View full text Add to dashboard Cite

Cuckoo search algorithm (CSA) has been a candidate for numerous recent applications and showed great compatibility in solving optimization problems. It is a metaheuristic algorithm which is based on the odd breeding strategy of the Cuckoo bird spices. It is used to find an optimum or near optimum solution for a certain problem. In this research, we propose an FPGA hardware implementation for the CSA based on single precision IEEE floating point (FP). The FP format provides a wider range and higher precision when compared to fixed point format. To the best of our knowledge, this is the first study to consider implementing FP format-based CSA on FPGA. The proposed design is implemented using pipelined and parallel techniques to get a high throughput and speed. The design is controlled and coordinated using finite state machines (FSMs) modules and is configured on Cyclone IV E FPGA chip from Intel. Three common benchmark functions are used to evaluate the performance of the proposed design. The design has a maximum operating frequency of 99 MHz. It was found out the maximum power consumption for the most complex function is 610.28 mW, mainly due to the use of FP format. In addition, the proposed design is implementation and evaluated for multidimensional operation. Accordingly, the proposed design is suitable for path planning for unmanned aerial vehicles (UAVs), sensor deployments for wireless sensor networks (WSNs) in addition to medical diagnostic and DSP applications. INDEX TERMS Cuckoo search algorithm (CSA), field programmable gate array (FPGA), finite state machine (FSM), IEEE floating point format.

show abstract

A fully customizable hardware implementation for general purpose genetic algorithms

Cited by 15 publications

References 12 publications

FPGA Implementation of Crossover Module of Genetic Algorithm

FPGA Implementation of Crossover Module of Genetic Algorithm

Search@Home: A Commercial Off-the-Shelf Environment for Investigating Optimization Problems

FPGA Implementation of Floating Point Based Cuckoo Search Algorithm

Contact Info

Product

Resources

About