Should primaries be considered victims or police?

This paper investigates spectrum-jail-based enforcement from the perspective of a primary. We first treat the primary as a victim, to understand the expense of protecting primaries in different situations. In all cases, the overhead can be limited by reducing the rate of wrongful convictions. We then allow the primary to participate in its own protection - ideally, we want the primary to report interference but otherwise operate as though the secondary were not present. However, the primary can report interference when there is none; it can also transmit “gibberish” to claim more than its required transmit time. Because the primary will be affected in some way by secondary presence, it unfortunately has incentive to use these behaviors to kick the secondary out of the band. Crying wolf, an obviously undesirable behavior, is too easily done if reporting is free. So, it must be countered by making reporting interference sufficiently expensive. “Gibberish,” on the other hand, requires more serious thought because it is not easily deterred and may in fact be useful in encouraging compatible subcontracting. Finally, we show that both primary and secondary do better with better coexistence strategies, so the jail system gives natural incentives for better receiver designs, registration in a database, or other approaches for making coexistence simpler.