Feature: Communications system design
see Figure 6, to evaluate its scattering parameters (S-parameters), comparing the measured data with the simulation results. Figures 7 and 8 show the measured and simulated S11 (return
loss) and S21 (insertion loss) over the frequency range of 1-8GHz. T e simulated S11 (solid red line) shows a deep resonance with a minimum of approximately -32dB in the passband (3.07-6.13GHz), indicating excellent impedance matching. T e measured S11 (dashed red line) follows the simulated
Figure 8: Comparison of simulated and measured results Another group of researchers presented a compact microstrip
bandpass fi lter employing a novel folded open‐loop resonator topology. T ey combined advanced design optimisation techniques with a rigorous parametric analysis to achieve a fi lter with low insertion loss, high selectivity and a wide stopband. Table 7 compares these designs with ours. As seen, despite using
a relatively higher loss substrate like FR4, our proposed design demonstrates competitive input matching, low insertion loss and a compact size (17 x 13 x 1 mm) compared to other designs. Our studies provided a clear rationale for the chosen dimensions,
balancing the competing demands of passband fl atness, insertion loss and fabrication tolerance. To further refi ne the design, a multi-objective optimisation
approach was implemented: • Electromagnetic fi eld distribution analysis: Full-wave simulations were used to analyze the fi eld concentration around discontinuities and slots, guiding the refi nement of geometrical features.
• Sensitivity analysis: We assessed the impact of dimensional variations on key performance metrics (S11 and S21) to ensure resilience against manufacturing tolerances.
• Trade-off analysis: T e optimisation targeted minimisation of insertion loss while maintaining a symmetric passband response. Adjustments were made iteratively, particularly to the fl oating block dimensions and slot geometries, to balance coupling strength and impedance matching.
Fabrication, testing and future developments T e fabricated microstrip bandpass fi lter was implemented on an FR4 substrate and manufactured with high precision using an advanced etching process; see Figure 5. T e substrate selection was guided by its durability and cost-
eff ectiveness in RF applications, with a dielectric constant of 4.4 and a loss tangent of 0.02. To ensure minimal deviations from the simulated design, careful attention was given to cutting accuracy, conductor patterning and alignment during fabrication. T e fi nal structure was tested using a Vector Network Analyzer (VNA),
28 December 2025/January 2026
www.electronicsworld.co.uk
curve closely but exhibits slight discrepancies in certain frequency regions. Specifi cally, the return loss in the measured data is marginally higher than the simulated values at some points, suggesting minor impedance mismatches likely caused by fabrication tolerances or slight variations in the dielectric constant of the FR4 substrate. T e transmission coeffi cient S21 (solid green line) in the
simulated response confi rms a low insertion loss of approximately -0.48dB within the passband, demonstrating eff ective signal transmission. T e measured S21 (dashed green line) follows a similar trend but with slightly higher insertion loss across the band. T is discrepancy can be attributed to multiple factors, including conductor and dielectric losses, fabrication imperfections and potential parasitic eff ects introduced by the SMA connectors and soldering process. Outside the passband, both the simulated and measured results
show strong attenuation, confi rming that the fi lter eff ectively suppresses unwanted frequencies beyond 6.5GHz. Despite the overall strong agreement between measured and
simulated results, some deviations were observed, particularly in the depth of return loss and the insertion loss across the band. T ese variations can be linked to several contributing factors: 1. Fabrication tolerances: Small variations in the etching process and substrate thickness can introduce slight shiſt s in the resonator’s electrical length, aff ecting the resonance frequency.
2. Connector and soldering eff ects: T e presence of SMA connectors and solder joints introduces additional parasitic reactances, which are not fully accounted for in the simulation.
3. FR4 material variability: T e actual dielectric properties of FR4 oſt en exhibit slight variations from their nominal values, infl uencing the fi lter’s frequency response.
4. Measurement artefacts: T e calibration of the VNA, along with possible cable losses, may slightly alter the measured S-parameters, particularly at higher frequencies. T e experimental results confi rm the successful realisation of
the proposed microstrip bandpass fi lter, with strong agreement between the simulated and measured responses. Whilst minor deviations exist, they remain within acceptable limits for practical RF applications. Further optimisation could involve refi ning the microstrip
etching process to enhance dimensional precision, improving impedance matching techniques, or exploring alternative substrates with lower loss characteristics to further improve performance. Additionally, compensating for fabrication uncertainties in the simulation model through tolerance-aware design methods could help mitigate these minor discrepancies in future iterations.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48
