A PCA-PSO-tuned inverse ELM framework for multi-objective antenna design for Wi-Fi 6E IoT applications Chaya Devi Pothala, Madhu Ramarakula Engineering Research Express, 2026 A fast and reliable inverse modelling framework based on Extreme Learning Machine (ELM) is presented for multi-objective antenna design targeting Wi-Fi 6E IoT applications. Conventional antenna design depends on repeated full-wave simulations to achieve desired electromagnetic (EM) performance resulting in high computational cost and design cycles. Inverse antenna design predicts geometry from target EM features, but the complex mapping between many EM inputs and few geometric variables often causes overfitting. To address these challenges, a non-iterative inverse design approach is developed using an ELM framework integrated with Principal Component Analysis (PCA) and Particle Swarm Optimization (PSO). PCA reduces EM feature dimensions to improve generalization, while PSO optimizes hidden-layer weights and biases, removing dependence on random initialization and enhancing accuracy. Four inverse ELM configurations along with a baseline ANN model were analysed. Among them, the PCA_PSO_ELM achieved a 5-fold cross-validated Mean Absolute Percentage Error (MAPE) of 7.72%, demonstrating good prediction reliability. The framework is validated through synthesis of a compact dual-band Wi-Fi 6E antenna. Using the inverse model, the geometry corresponding to target EM specifications was directly estimated to obtain a 40 × 10 mm 2 FR-4 monopole antenna operating at 2.4 GHz and 6.1 GHz. The fabricated prototype achieved peak gains of 2.7 dB and 2.4 dB, with bandwidths of 300 MHz and 650 MHz, respectively. These results confirm the efficiency, speed, and practical suitability of the proposed inverse ELM approach for intelligent multi-objective antenna design in emerging IoT devices.
ML-Augmented Optimization of LoRa Antennas for Drone Telemetry Pothala Chaya Devi, Ramarakula Madhu IEEE Access, 2025 A compact printed monopole antenna for drone telemetry communication, operating at 433 MHz, with a gain of 2.2 dBi, is designed. The antenna is fabricated on an FR-4 substrate with dimensions of 0.116 λ₀ × 0.073 λ₀ and is optimized for long-range communication. A Machine Learning augmented Optimization (MLaO) method is proposed to reduce the antenna design time compared to traditional electromagnetic simulation techniques. Typically, antenna design involves complex computer simulations like CST and computationally intensive parameter sweeps. However, in this work a surrogate Artificial Neural Network (ANN) model trained on 1080 antenna designs replaces the heavy CST simulations. This ANN model is then coupled with a Simulated Annealing (SA) optimizer to generate antennas with the desired characteristics, reducing the total design time to 57% compared to traditional techniques. Three antenna designs were simulated using MLaO for different long-range (LoRa) frequency bands, with Ata1 (433 MHz) achieved a return loss (S₁₁) of -23.3 dB, Ata2 (865 MHz) had an S₁₁ of -30.6 dB, and Ata3 (dual band at 433 MHz and 865 MHz) with S₁₁ of -15.6 dB and -35.4 dB respectively. The fabricated antenna (433 MHz) was mounted on a drone and tested with a 3DR-433 telemetry transceiver, recording an average Received Signal Strength Indicator (RSSI) of -57.8 dBm up to 470 m. These results demonstrate the proposed antenna’s efficiency, compactness, and the effectiveness of the MLaO approach for fast and accurate antenna design.
Implementation and Evaluation of Approximate Multiplier for Improved Efficiency P. Chaya Devi, Y. Bhanu Prakash, T. Divya, B. Dakshayani, B. Sai Kumar 2024 IEEE International Conference on Smart Power Control and Renewable Energy Icspcre 2024, 2024 Multipliers stand as essential components that play a pivotal role in computational tasks. Their importance is found not just in achieving accurate and high-speed calculations but also in optimizing power efficiency. Approximate multipliers are useful in applications where a modest level of imprecision is tolerated but does not affect the overall outcome. In this paper, two approximate 4:2 compressors are proposed and utilized in an 8 × 8 Dadda multiplier. By strategically introducing controlled approximations, by sacrificing precision, multipliers with proposed approximate compressors offer reduction in power consumption, area utilization, and delay, by compromising on accuracy. The error analysis is further verified on the scales of Error Rate (ER), Error distance (ED) and multiple other metrics. Xilinx Vivado software is used to simulate and synthesize the proposed designs.
Smart agriculture using iot Bammidi Deepa, Chukka Anusha, P. Chaya Devi Advances in Intelligent Systems and Computing, 2021
Notice of Removal: Design of low interference high directive planar antenna with schelkunoff polynomial method P.Chaya Devi, K. Anusha, M. Kalpana, L. Ganesh International Conference on Electrical Electronics Signals Communication and Optimization Eesco 2015, 2015 The words broadcasting, unicasting and multicasting are the well familiar terms in the field of communications and are very important in describing various parameters in designing the system. In the field of communications, the applications which involve broadcasting needs an antenna of high directivity and should use the power effectively (low side lobes). Where in applications that needs unicast-reception mainly needs high directivity (Zone of reception) and low interference (low side lobes) antennas. There are some special applications like tracking radars, surveillance antennas demand patterns with special characteristics (Beam widths). Due to the non complexity in implementation and ability to produce symmetrical-high directive beams, generally planar array antennas are used in above mentioned applications. Planar antennas, with compromise in increased number of elements and size, produces high directive beams. Other way of generating desired beam pattern is by synthesizing the antenna radiation pattern. Schelkunoff polynomial synthesis method is one being used in linear array design for suppressing radiation in undesired directions, there by increases the directivity. In order to produce the desired beam with high directivity this paper proposes a planar array design method and also extends the schelkunoff polynomial method (confined as linear array synthesismethod ) to planar array design to produce cost effective high directive antennas. The radiation pattern characteristics (RPC) Directivity, 3dB beam width, Null-Null beam width and side lobe levels are used to analyse the performance of proposed design and algorithm.
