Maximilian Bauhofer,Silvio Mandelli,Marcus Henninger,Thorsten Wild,Stephan ten Brink
Abstract:In future wireless communication networks, existing active localization will gradually evolve into more sophisticated (passive) sensing functionalities. One main enabler for this process is the merging of information collected from the network's nodes, sensing the environment in a multi-static deployment. The current literature considers single sensing node systems and/or single target scenarios, mainly focusing on specific issues pertaining to hardware impairments or algorithmic challenges. In contrast, in this work we propose an ensemble of techniques for processing the information gathered from multiple sensing nodes, jointly observing an environment with multiple targets. A scattering model is used within a flexibly configurable framework to highlight the challenges and issues with algorithms used in this distributed sensing task. We validate our approach by supporting it with detailed link budget evaluations, considering practical millimeter-wave systems' capabilities. Our numerical evaluations are performed in an indoor scenario, sweeping a variety of parameter to analyze the KPIs sensitivity with respect to each of them. The proposed algorithms to fuse information by multiple nodes show significant gains in terms of targets' localization performance, with up to 35\% for the probability of detection, compared to the baseline with a mono-static setup.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to achieve multi - target localization in multi - static integrated sensing and communication deployments. Specifically, current research mainly focuses on single - sensing - node systems or single - target scenarios, paying attention to specific problems of hardware deficiencies or algorithmic challenges. However, for distributed sensing tasks with multiple targets and multiple nodes, there is a lack of a comprehensive perspective to process and fuse the information collected from multiple sensing nodes. This paper proposes an approach based on the scattering model and demonstrates the challenges and problems of existing algorithms in this distributed sensing task within a flexible and configurable framework. Through detailed link - budget evaluation and numerical simulation, it is verified that the proposed multi - node information - fusion algorithm significantly improves the target - location performance, especially in terms of the detection probability, which is 35% higher than that in the single - static setup.
### Main Contributions
1. **Complete (classical) sensing pipeline**: Similar to the existing cellular network structure.
2. **Set of multi - node peak - extraction and - fusion techniques**: Used to process and fuse the data collected by multiple nodes.
3. **Numerical evaluation**: Conducted in realistic multi - scattering - target scenarios, with a focus on the physical feasibility and consistency at millimeter - wave frequencies.
### Methods
- **System model**: Using OFDM radar technology, each sensing access point (SAP) receives signals through single - static acquisition.
- **Periodogram method**: Estimates the angle and distance information of the target through DFT.
- **Signal - to - noise ratio (SNR)**: Defines the symbol - level SNR and the sensing SNR at the communication center.
- **Network structure**: A structure of multiple SAPs and a central sensing management function (SeMF), similar to the 5G location management function (LMF).
- **Target extraction**: Performs peak extraction and CFAR threshold detection in SAP.
- **Fusion process**: Performs geometric clustering and multi - node fusion in SeMF to reduce false detections.
### Results
- **Multi - SAP fusion performance**: Multi - node fusion significantly improves the target - detection probability, especially in the low - noise - power range.
- **Influence of bandwidth and number of antennas**: More antennas and higher bandwidth can improve the detection probability and accuracy.
- **Influence of the number of targets**: As the number of targets increases, the detection probability decreases, but multi - static sensing can partially alleviate this problem through multi - node fusion.
- **Influence of room size**: A larger room size can reduce the probability of the target being occluded, and multi - node SAPs maintain a relatively high detection probability over a larger range.
### Conclusions
This paper proposes a framework for achieving multi - target localization in multi - static integrated sensing and communication deployments, which significantly improves the performance and robustness of the system through multi - node information - fusion. Future research will focus on optimizing subsystems, especially the extraction process, as well as more complex fusion techniques. At the same time, it is planned to expand to more realistic scenarios, including the fusion of clutter and bistatic acquisition.