This work is a part of my Bachelor Thesis.
This Repository presents an efficient design of Finite Impulse Response (FIR) filter using G-best guided Cuckoo Search (GCS) algorithm. To reduce the parameter dependency in traditional Cuckoo Search Algorithm (CSA) as well as better searching of optimal coefficient in the filter designing problems, some modification in the approach of CSA is done in the proposed GCS, which results in a noteworthy faster convergence with an optimal solution. Here, both low-pass and band-pass filters have been designed, for both Type1 and Type 2 FIR, using GCS, CSA, and Artificial Bee Colony (ABC) algorithm.
From the graphical and statistical observations (here) of GCS, CSA, and ABC, it is clear that GCS surpass its competitors in terms of convergence rate, execution time, and filter response.
Furthermore, the stop-band attenuation (As ) and pass-band ripple (Rp) obtained for the filters developed using the proposed GCS is much better than that designed using CSA, ABC or, Parks and McClellan (PM) method.
An average study suggests that GCS shows 9.23% increase in As and 26.05% decrease in Rp for lower order LPF(for both Type 1 and Type 2, order less than 40); 4.27% increase in As and 18.85% decrease in Rp for higher order LPF(for both Type 1 and Type 2, order greater than 40 and less than 100); 32.47% increase in As and 28.18% decrease in Rp for lower order BPF(for both Type 1 and Type 2, order less than 40); 6.92% increase in As and 10.44% decrease in Rp for higher order BPF(for both Type 1 and Type 2,order greater than 40 and less than 100), compared to PM.
- For GCS BPF-Type1 Filter, go to this folder
- For GCS BPF-Type2 Filter, go to this folder
- For Cuckoo Search Algorithm , Go to here for Type 1 BPF and here for Type 2 BPF
- Artificial Bee Colony codes are available as well. Go to here for Type 1 BPF and here for Type 2 BPF
1. This repo contains only the code for the BPF.
2. Ensure that you have installed Matlab and have CPU with sufficient RAM (>=4GB).