Aggregate Interference Margin

The Aggregate Interference Margin is a calculated safety buffer representing the maximum total interference power allowed from all non-incumbent transmitters at the protected contour of a primary receiver. It is the difference between the incumbent's interference-to-noise ratio (I/N) protection criterion and the actual predicted aggregate interference level. The AIM ensures that the cumulative effect of multiple low-power secondary users does not degrade the primary user's link budget below its operational threshold.

Calculating the AIM involves summing the contributions of all secondary transmitters using a statistical propagation model.

• Deterministic Component: Accounts for known, registered transmitters with fixed locations and powers.
• Stochastic Component: Models the aggregate effect of a large population of mobile or uncoordinated devices using log-normal shadowing and Monte Carlo simulations.
• Formula: AIM = I_limit - 10*log10(∑ 10^(I_i/10)), where I_limit is the incumbent's interference threshold and I_i is the received interference power from the i-th secondary user.

In a Spectrum Access System (SAS), the AIM is the central constraint for the Coexistence Manager (CxM). The SAS must ensure that the sum of interference from all authorized Citizen Broadband Radio Service (CBRS) devices (both PAL and GAA tiers) does not exceed the margin at any protected Dynamic Protection Area (DPA). If the predicted aggregate interference approaches the margin, the SAS must deny access or reduce the power of lower-tier devices.

The total AIM is often partitioned into sub-margins for different user classes or operators to ensure fairness.

• Tiered Allocation: A larger portion of the margin is reserved for Priority Access License (PAL) holders, guaranteeing their interference protection.
• Per-Operator Budget: In multi-operator scenarios, the margin is divided into operator-specific budgets to prevent one network from consuming the entire interference allowance.
• Geographic Partitioning: The margin can be spatially divided to manage interference in high-density urban canyons versus rural macro-cell deployments.

A Dynamic Protection Area (DPA) is a geospatial zone activated by a SAS when a federal incumbent, such as a naval radar, is operating. Upon activation, the SAS recalculates the AIM for all CBRS devices within and near the DPA. If the aggregate interference margin is exceeded, the SAS must issue a suspension order to specific secondary users, forcing them to cease transmissions or move to a different channel within a mandated timeframe, typically 300 seconds.

To account for real-world imperfections, the AIM includes several conservative safety factors:

• Propagation Model Uncertainty: Margins for errors in terrain-based path loss predictions (e.g., Longley-Rice model).
• Device Geolocation Error: Buffers for GPS inaccuracies in reporting the position of mobile secondary users.
• Temporal Fading Margin: An allowance for slow-fading effects that are not captured by median path loss models, ensuring protection even during constructive interference peaks.

A geospatial polygon defining the protected contour around a federal incumbent receiver, such as a naval radar. When a SAS activates a DPA, it must compute the aggregate interference from all CBRS devices within and near that zone. If the predicted Aggregate Interference Margin is exceeded, the SAS issues suspension orders to secondary users. DPAs are not static; they move with shipborne radar systems, requiring continuous recomputation of the margin.

Deep learning systems that categorize the type and source of interference contributing to the aggregate margin. These models distinguish between:

• Co-channel interference from other secondary users
• Adjacent-channel leakage due to non-ideal filtering
• Non-linear intermodulation products from colocated transmitters Accurate classification enables the SAS to apply targeted mitigation rather than blanket power reduction, preserving network capacity.

The mathematical foundation for computing the Aggregate Interference Margin. Models such as Irregular Terrain Model (ITM) and Hata-Davidson predict how RF energy attenuates over distance and obstacles. The SAS uses these models to estimate the path loss from each secondary transmitter to the incumbent receiver. Errors in propagation modeling directly translate to errors in the interference margin, making high-resolution terrain data and real-time weather inputs critical for accurate protection.

A coexistence paradigm where secondary users transmit simultaneously with the incumbent by spreading their signal below the noise floor. The Aggregate Interference Margin in underlay systems is managed through ultra-wideband (UWB) or direct-sequence spread spectrum (DSSS) techniques. Each additional underlay device raises the noise floor incrementally, and the margin defines the maximum permissible number of concurrent underlay transmitters before the incumbent's signal-to-noise ratio degrades below threshold.

A privacy-preserving machine learning approach where multiple SAS instances or cognitive radios collaboratively train an interference prediction model without sharing raw spectrum sensing data. Each node computes local gradient updates on its observed interference patterns and shares only the model parameters. This enables a more accurate, geographically distributed estimate of the Aggregate Interference Margin without exposing sensitive information about incumbent locations or operational patterns.