The distributed Brillouin optical time domain reflectometry (BOTDR) sensing system is based on the linear relationship between the frequency shift of spontaneous Brillouin scattering (SpBS) and temperature/strain on the optical fiber. It utilizes a convenient structure of single-ended incident light detection to measure temperature and strain along the fiber over long distances, reducing measurement complexity and application costs. Meanwhile, it allows for continued fiber detection even when the optical fiber is broken on an engineering construction site, avoiding the limitations associated with employing a double-ended loop Brillouin optical time domain analysis (BOTDA) sensing system after the fiber loop breakage. This makes the BOTDR system more practical in engineering applications. However, due to limitations imposed by stimulated Brillouin scattering (SBS), if only SpBS occurs in the system,the power can not exceed SBS’s threshold power. Consequently, the low incoming fiber optical power results in weak signal energy and a low signal-to-noise ratio (SNR) within the system, affecting overall detection accuracy. To improve the SNR without increasing complexity or detection costs, we propose a method that enhances the compactness and affordability of BOTDR systems by local stimulated scattering excitation.

To validate the method’s feasibility, we conduct three experiments. Firstly, a 20 ns pump pulse is utilized at room temperature with a pulse frequency of 40 kHz to measure the 1009 m optical fiber under test. By comparing the root mean square error (RMSE) of the frequency shift distribution at different positions and various current levels, it is confirmed that increasing the operating current of EDFA can enhance the SNR of the BOTDR system in the absence of SBS occurrence. Subsequently, by setting the pulse cycle frequency to 10 kHz, we measure a 2019 m optical fiber under test. The RMSE of the optical fiber’s frequency shift distribution is calculated in segments. Comparison between eight segments with different current levels verifies that appropriate SBS can improve the SNR of the BOTDR system. Finally, by employing an optimized system configuration, temperature measurement experiments are conducted. At room temperature (23 ℃), SOA modulation enables a pulse width modulation to 50 ns and sets a frequency of 20 kHz. In these conditions, the maximum detection distance reaches up to 5000 m with a tested optical fiber length of 2 km obtained. A section consisting of two separate portions (nearby distances: approximately 300 m and 800 m) within this range is heated to reach temperatures as high as 50 ℃ using a water bath technique, which achieves temperature measurement accuracy of 0.33 ℃.

The accuracy of system frequency shift detection is a crucial metric for evaluating the BOTDR system, primarily determined by the system’s frequency resolution. The resolution is influenced by factors such as the SNR, short-time Fourier transform (STFT) frequency step, central frequency of Brillouin gain spectra, and full width at half-maximum. Traditional BOTDR systems rely on spontaneous Brillouin scattering for temperature and stress measurements along the fiber. However, this method is characterized by low scattered light signal strength and limited detection accuracy. In contrast, stimulated Brillouin scattering offers higher scattered light signal strength but results in significant energy loss of the pump pulse. By employing locally stimulated scattering to enhance the detection accuracy of compact low-cost BOTDR systems, we achieve temperature measurement accuracy of 0.33 ℃ in short-distance measurements (Table 3). This success validates the feasibility of our approach while requiring a relatively simple system structure, lower detection costs, and improved engineering practicality (Fig. 3).

We propose a method to enhance the detection accuracy of BOTDR systems by utilizing local SBS. When SBS occurs in the optical fiber, temperature measurement accuracy of 0.33 ℃ is achieved on the sensing fiber ranging from 20 to 900 m. This indicates that although SBS may reduce the sensing distance of BOTDR systems, it can improve the measurement accuracy. Based on a simple single-ended incident light structure, our approach employs local SBS to improve the detection accuracy of BOTDR systems over short distances. Compared with traditional BOTDR systems, our method features higher SNR and more accurate detection without increasing complexity or application costs. Meanwhile, it enables continued detection even under fiber breakage. These advantages further enhance the engineering utility of compact and low-cost BOTDR systems.