Cognitive Resource Analyzer for Cellular Network Ecosystems

Modern cellular infrastructure has grown in complexity and become an ecosystem that includes the host technologies, such as 4G Long Term Evolution (LTE) and 5G New Radio (NR), along with subsystems such as narrowband Internet of Things (NB-IoT), category M1 (Cat-M1/LTE-M), observed time difference of arrival (OTDOA) and support for 4G/5G coexistence using dynamic spectrum sharing (DSS). Such an increased complexity has driven the need for network monitoring to enable load tracking, congestion control, spectral efficiency analysis, intrusion detection, and cognitive radio communications. However, no approach can passively monitor the entire cellular ecosystem because some reservations are configured with high-layer messages that are only accessible to in-network, active, and sometimes user-specific equipment. In this work, we design the Resource Analyzer for Variable Ecosystem Networks (RAVEN) to monitor the individual technologies within the cellular ecosystem. RAVEN reduces the synchronization search space for multiple subsystems and enables a fine-grained subsystem-level tracking of spectrum usage. RAVEN also provides a method to detect anomalies and find subsystems that are not yet supported. RAVEN is validated through simulation and live experiments and is compared to other cellular network monitors.