Secondary User (SU)

The defining characteristic of an SU is its legal and technical obligation to avoid harmful interference to the Primary User (PU) . An SU does not have guaranteed access to spectrum; it opportunistically exploits spectrum holes—frequency bands that are unoccupied at a specific time and geographic location. This requires continuous spectrum sensing to detect the return of a PU, triggering immediate vacation of the channel.

To maintain seamless communication, an SU must possess spectrum mobility—the ability to vacate a channel and re-establish a link on a different frequency without service disruption. This process, known as spectrum handoff, is triggered by PU detection or channel quality degradation. Effective handoff protocols are critical for maintaining Quality of Service (QoS) in dynamic environments.

SUs operate within a strict regulatory hierarchy. In frameworks like the Citizens Broadband Radio Service (CBRS) , SUs are the lowest tier (General Authorized Access), subordinate to Incumbent Access and Priority Access Licensees. They must query a central Spectrum Access System (SAS) for authorization and frequency assignments, ensuring they never transmit on channels reserved for higher-tier users.

Unlike static legacy radios, an SU is often a Cognitive Radio (CR) capable of understanding its RF environment. This cognition goes beyond simple energy detection to include Automatic Modulation Classification and predictive modeling of spectrum occupancy. By learning temporal usage patterns, an SU can proactively switch channels before a PU even appears, minimizing latency.

To facilitate underlay spectrum access, SUs must dynamically adjust their transmit power. By constraining their signal to remain below a strict interference temperature limit at the PU receiver, an SU can theoretically transmit concurrently with a PU. This requires precise channel estimation and closed-loop power control to avoid crossing the threshold from noise to harmful interference.

In multi-agent environments, SUs can adopt cooperative or selfish strategies. Cooperative SUs share spectrum sensing data to improve PU detection accuracy and coordinate channel access to avoid collisions with other SUs. Selfish SUs optimize only their own throughput, potentially causing congestion. Multi-Agent Reinforcement Learning (MARL) is often used to train optimal sharing behaviors.