Antenna geometry

The SIW cavity of size 3.6λ _g × 1.8λ _g is used to enclose six L-slot antenna pairs. The antenna elements were further separated from each other by the via boundary to form sub-cavities of size 1.0λ _g × 0.6λ _g as shown in Fig. 1.

Fig. 1. The proposed antenna design.

The dual L-slots are clubbed together in inverted form to produce inductive and capacitive loading. The direction of the wave along the y-axis is affected by the parallel combination of an inductor (due to slot) and capacitor (due to the gap between two slots). The corners of the L-slots are curved to improve the gain by almost 0.3 dB. The overall structure is divided into seven different cavities. Cavity-1 is used to feed the L-slots available in the remaining six sub cavities. The main Cavity-1 is having a metallic vias in-front of the sub-cavities, which was introduced to help individually feed the slots in cavities. The positions of these vias are optimized to improve the front to back and side lobe radiation ratio and hence achieve the desired gain. The design is fed by a lumped port, the position of which is adjusted after a detailed parametric study. The L-slots are created on the top copper layer of the substrate, while DGS is created at the bottom surface. The position and the inner /outer radius of the DGS are also optimized to achieve a low side lobe level. The design also integrates two metallic vias just below the DGS structure. These vias help to modulate the current till the last L-shaped slots. Various parameters of the design are given in Table 1.

Table 1. Design parameters

Working principle

The resonant frequency of the sub-cavity containing the L-slots can be calculated by using equation-1[Reference Cassivi, Perregrini, Wu and Conciauro8].

(1)$$F_{R( {TEm0q} ) } = \displaystyle{{c_o} \over {2.\sqrt {\varepsilon _r} }}\sqrt {{\left({\displaystyle{m \over {W_{eff}}}} \right)}^2 + {\left({\displaystyle{q \over {L_{eff}}}} \right)}^2} $$

where,

(2)$$L_{eff } = L-\displaystyle{{D^2} \over {0.95\cdot p}}$$

and

(3)$$W_{eff } = W-\displaystyle{{D^2} \over {0.95\cdot p}}$$

‘D” is the diameter of the via, whereas “p” is the pitch of the via. The above equations are only valid for $b < \lambda _o.\sqrt {\varepsilon _r} /2$ and b < 4D. The above formula got verified for sub-cavity, by putting the value of “D” = 1 mm, “p” = 1.5 mm, “L” = 13.5 mm, “W” = 8 mm, “ɛ _r” = 2.33 and the height of the substrate “h” = 3.18 mm. The cavity resonates at 15.5 GHz in its fundamental TE₁₀₁ mode. The resonant frequency in the measured result is 14.29 GHz whereas in the simulation it is 14.59 GHz. This proves the closest approximation of the resonant frequency in theoretical, simulated and measured results. The resonant frequency shifted to the left due to the inductive loading of the slots created inside the cavity. This also helps to miniaturize the antenna for the given frequency of operation.

The array factor of the proposed 1 × 6 antenna array can be evaluated by using equation (4) and (5) [Reference Balanis17].

(4)$$( {AF} ) _N = \displaystyle{1 \over N}\left[{\displaystyle{{{\rm sin}\left({\displaystyle{N \over 2}\psi } \right)} \over {{\rm sin}\left({\displaystyle{1 \over 2}\psi } \right)}}} \right]$$

where,

(5)$$\psi = {k.d.cos\theta+\beta}$$

Here, β is the successive phase difference between the antenna elements, “d” = 8 mm/0.55λ _o is the distance between the two antenna elements and “N  = 6”, k is the free space wave number.

The advantage of using end metallic vias and DGS structure can be seen in Fig. 2. It is clear from Fig. 2(b) that the outer metallic vias helps to restrict the maximum current to flow on the outer L-slots. Figure 2(c) shows that the DGS structure allows some amount of power to flow to the outer L-slot, which in turn improve the overall gain of the antenna. The comparison of the designs shown in Fig. 2 is explained on the basis of S ₁₁ (dB) parameter through Fig. 3.

Fig. 2. The current distribution in structure at 14.59 GHz (a) without DGS and outer slots vias (b) without outer vias (d) without DGS (c) proposed design.

Fig. 3. Comparison S ₁₁ (dB) graph for various designs.

The S ₁₁ (dB) comparison graph shows one important aspect of the design. The design without DGS shows one more prominent frequency of operation at 14.08 GHz, which was quite suppressed in the proposed antenna design with the introduction of DGS. The reason not to include the lower frequency is that the radiation pattern at 14.08 GHz for the design without DGS shows a dip in the center.

Simulated and measured results

The device is fabricated through a PCB prototyping machine on an RT/Duroid Roggers substrate of thickness 3.18 mm.

Due to the use of the thicker substrate, the process of creating the printed though hole across it was typical, hence the vias are filled with copper wire and soldered at both ends as shown in Fig. 4. During this process the copper layer on the top and bottom layers were tined. Tinning is a common phenomenon in PCB design and does not affect antenna parameters to a larger extent. The proposed design was tested for S ₁₁ (dB) parameter on the Antrisu Site Master, S820E VNA for 30 GHz. The test bench of the antenna under test can be seen in Fig. 5(a). Figure 5(b) shows the Anechoic Chamber setup for testing antenna radiation pattern and gain.

Fig. 4. The fabricated antenna prototype (a) Top layer (b) Bottom Layer (c) Top layer after tinning process (d) Bottom Layer after tinning.

Fig. 5. Antenna under test at (a) Antrisu Site Master, S820E VNA (b) Anechoic Chamber for testing antenna radiation pattern and gain.

The S ₁₁ (dB) of the proposed antenna is shown at Fig. 6. The measured results shifted left by 0.3 GHz. The shift can be explained by the loading done by tin material at the top and bottom layer of the antenna. The lower frequency band in the measured S ₁₁ (dB) is suppressed by a considerable amount, so the band can be neglected. The measured result shows that the antenna achieves a 10 db bandwidth of 44 MHz.

Fig. 6. Tested and Measured S ₁₁ (dB) parameter with Gain (dBi).

Figure 6 also shows that at the operating frequency of 14.59 GHz the antenna shows a gain of 8.63 dBi. The measured gain at 14.29 GHz is calculated as 8 dBi when the distance between the test antenna and referenced antenna is kept 1 m and the gain of the referenced double ridge horn antenna at 14.29 GHz is 14.35 dBi. Figures 7(a) and 7(b) shows the measured and simulated normalized electric field for two planes (a)angle phi = 0° (xz plane) and (b)angle phi = 90° (yz plane) for the proposed antenna.

Fig. 7. Electric Field (a) measured and simulated for angle phi = 0^o (b) measured and simulated for angle phi = 90^o.

The two dimensional and three-dimensional electric field in Figs 7 and 8 respectively, shows that the radiation pattern is projected only in a single direction. A high directive beam with low side and back lobe radiation leads to achieving a high gain. The FBR of the proposed antenna structure is 24.9 dB.

Fig. 8. Three-dimensional electric field.

The comparison of the proposed design is done in Table 2, with seven different references cited in this communication. From Table 2 it is clear that the proposed antenna design is smaller in size and achieves good gain.

Table 2. Comparison Table

Since, there are limited numbers of designs of the L-slot antenna, hence the major comparison was done with other types of the slot antenna enclosed in a SIW cavity and further divided into sub cavity. In [Reference Wen, Wang and Ding11], the author further replicated his design to form a phased array antenna, hence the comparison was done only with a single element slot antenna.