Heterogeneous Electromagnetic Spectrum Data Fusion Method Based on RSSI Detection Error Minimization

In the field of radio spectrum detection, distributed electromagnetic spectrum detection systems can collect and process data more efficiently and flexibly than one-time mobile tests or fixed-point long-term tests. However, in a distributed system, multiple detection nodes need to be arranged in the monitoring area, each with its own data acquisition module. Due to individual differences between different devices, the measured electromagnetic data can also have errors. When comprehensively analyzing the data collected by each node, it will cause certain errors and affect the final calculation results. This paper proposes a method to normalize Received Signal Strength Indicator(RSSI) features. First, multiple HackRF modules are used to collect a set of energy data for the same signal source in the same environment, and the RSSI curve of the signal source in this environment is fitted by calculation. Next, specify a standard RSSI curve equation, adjust all curves to standard curves through formula transformation, increase the similarity between them, and derive a correction formula for each acquisition device. Then use the same equipment in the same environment for the second round of data collection, use the data of the second round as verification data, and calculate the RSSI curve of the second round. After obtaining the curve equation, the correction formula is used to calibrate, and the error before and after the correction of the verification data is compared. Experiments show that this method successfully increases the similarity of RSSI curves corresponding to different devices, and reduces the measurement differences of different HackRF devices on the same signal source in the same environment.