The simulation was performed using FDTD method in order to investigate the optical behavior of the proposed DEMUX and the output spectrum was plotted in terms of wavelength. To use this method, accurate meshing and time calculations were required. The following equation was applied [55]:
\(\varDelta t\le \frac{1}{c\sqrt{(\frac{1}{{\varDelta x}^{2}}-\frac{1}{{\varDelta z}^{2}})}}\) [2]
where Δx and Δz denote mesh sizes of the structure, Δt represents time steps of the structure and c is the light speed in open space.
The 8-channel DEMUX is a key process in data transmission. In this study, the main goal was to design an optical DEMUX with 8 output channels at very low bandwidth in the third telecommunication window due to low losses. The lattice constant (Ʌ) had to be changed in order to place the output channels in the range of 1.55 µm in the proposed structure. The output spectrum of the structure tended toward lower wavelengths by reducing the lattice constant, so that the output spectrum of the structure was in the wavelength range of C-band at the lattice constant of 0.522 µm. Then, the appropriate output spectrum was obtained for the eight output channels in terms of quality factor, transmission efficiency and crosstalk by managing the refractive index and radius of dielectric rods of the structure and performing the trial and error of defect rods in the output path.
Figure 3 illustrates the output spectrum of the proposed DEMUX, which included 8 channels with the central wavelengths of λ1 = 1.5271 µm, λ2 = 1.5337 µm, λ3 = 1.5395 µm, λ4 = 1.5446 µm, λ5 = 1.55 µm, λ6 = 1.5542 µm, λ7 = 1.5587 µm and λ8 = 1.5632 µm, all of which were in the C-band telecommunication window.
As shown in Fig. 3, the average overall efficiency and channel spacing were 98% and 5 nm, respectively.
Quality factor (QF) is another important parameter in DEMUX. The quality factor of DEMUX for each channel was calculated as follows:
\(\text{Q}=\frac{{{\lambda }}_{0}}{\varDelta {\lambda }}\) [3]
where QF is the quality factor, λ0 is the central wavelength and ∆λ is the channel bandwidth. According to the above equation, lower bandwidth was associated with a higher quality factor. Therefore, in the proposed structure, the minimum and maximum quality factors were 2,739 and 4,277, respectively. Table 2 presents details of the output spectrum of each channel such as central wavelength, bandwidth, quality factor and transmission efficiency.
Table 2
The output characteristics of the eight PhC DEMUX
| Ch1 | Ch2 | Ch3 | Ch4 | Ch5 | Ch6 | Ch7 | Ch8 |
Central wavelength (µm) | 1.5271 | 1.5337 | 1.5395 | 1.5446 | 1.55 | 1.5542 | 1.5587 | 1.5632 |
Band width (nm) | 0.36 | 0.56 | 0.36 | 0.44 | 0.52 | 0.48 | 0.42 | 0.38 |
Transmission | 96% | 98% | 99% | 98% | 98.5% | 95.5% | 98% | 99.5% |
Quality Factor | 4242 | 2739 | 4277 | 3511 | 2981 | 3238 | 3712 | 4114 |
Figure 4 (a-h) indicates the electric field distribution of the proposed DEMUX in channels 1-8, respectively. All the input signals reached channel 1 at λ1 = 1.5271 µm. Therefore, the signal power of channel 1 reached the maximum value at the desired wavelength. Similarly, outputs of channels 2, 3, 4, 5, 6, 7 and 8 received the highest input signals at λ2 = 1.5337 µm, λ3 = 1.5395 µm, λ4 = 1.5446 µm, λ5 = 1.55 µm, λ6 = 1.5542 µm, λ7 = 1.5587 µm and λ8 = 1.5632 µm.
The degree of interference between adjacent channels (crosstalk) is another important indicator for evaluating the performance of the PhC-based DEMUX. The effect of crosstalk on optical communications is serious because the signal sent to a channel can create an adverse effect on another channel. In designing DEMUXs, there is less effort to achieve crosstalk. According to the output spectrum of the structure (dB) shown in Fig. 5, Table 3 presents the crosstalk. The inter-channel crosstalk ranged from -60.5 to -25.5 dB, which was very lower than previous similar works.
Table 3
Inter-channel crosstalk of the 8 outputs of the proposed DEMUX
Channel | Ch1 | Ch2 | Ch3 | Ch4 | Ch5 | Ch6 | Ch7 | Ch8 |
Ch1 | --- | -33.3 dB | -48 dB | -45.2 dB | -37 dB | -52 dB | -59 dB | -56 dB |
Ch2 | -26.6 dB | --- | -30 dB | -39.5 dB | -30 dB | -46 dB | -51.5 dB | -51 dB |
Ch3 | -39 dB | -37.5 dB | --- | -33 dB | -34.6 dB | -60.5 dB | -44.5 dB | -47 dB |
Ch4 | -46 dB | -36 dB | -28.5 dB | --- | -32.5 dB | -44 dB | -45 dB | -42 dB |
Ch5 | -48.5 dB | -40 dB | -34 dB | -28.5 dB | --- | -31.5 dB | -40.5 dB | -42.5 dB |
Ch6 | -44.6 dB | -40 dB | -42.5 dB | -40.4 dB | -25.5 dB | --- | -33 dB | -46.6 dB |
Ch7 | -47 dB | -49.8 dB | -41.7 dB | -49 dB | -30.4 dB | -30.5 dB | --- | -41.5 dB |
Ch8 | -50 dB | -54 dB | -51 dB | -39.7 dB | -32.4 dB | -39.2 dB | -30 dB | --- |
Table 4 compares the obtained results with those of previous works in order to value the present paper. The proposed structure had higher quality factor and transmission efficiency as well as lower crosstalk than previous works.
Table 4
Comparing results of this paper with similar previous works
Reference | Number of channels | Channel spacing (nm) | Spectral width(nm) | Quality Factor | Transmission (%) | Crosstalk (dB) |
[4] | 4 | --- | 0.217 | 7358.5 | 99.25 | -9.79 up to -46.68 |
[15] | 8 | 4 | 1.2 | 1577.7 | 94.5 | -8 up to -48.3 |
[26] | 5 | --- | 0.22 | 6236 | 95.98 | -18.04 up to -50.2 |
[49] | 8 | --- | 0.65 | 2955 | 56 | -8 up to -40 |
[50] | 8 | 2.1 | 0.67 | 2300 | 94-99 | -11.2 up to -40 |
[51] | 8 | 4.2 | 1.8 | 825 | 80 | --- |
[52] | 8 | --- | 0.59 | 4320 | 93 | -11 up to -46 |
[53] | 8 | 3 | 1.2 | 1319 | 98 | -16 up to -40 |
[54] | 8 | 2.25 | 1.5 | 1000 | 96 | -35 up to -160 |
Proposed DEMUX | 8 | 5 | 0.44 | 3602 | 98 | -25.5 up to -60.5 |