Dynamic Spectrum Access (DSA)

The foundational principle of DSA. Secondary users (SUs) are permitted to transmit only when a spectrum hole is detected—a frequency band allocated to a primary user (PU) that is temporarily unused in a specific geographic location and time.

• Non-interference basis: SUs must vacate the channel immediately upon detecting a returning PU.
• Temporal granularity: Access windows can range from milliseconds in tactical networks to hours in TV white spaces.
• Example: A Wi-Fi-like device operating in the 3.5 GHz CBRS band opportunistically using a Navy radar channel when the coastal area is inactive.

DSA radios must continuously monitor the electromagnetic environment to build and maintain an accurate picture of spectrum occupancy. This involves spectrum sensing—the real-time detection of PU signals—often augmented by external data sources.

• In-band sensing: Monitoring the currently used channel for a returning PU.
• Out-of-band sensing: Scanning alternative channels to maintain a ranked list of backup frequencies.
• Geolocation database access: Querying a regulatory database (e.g., for TV white spaces) to cross-reference sensed data with licensed transmitters.
• Key metric: Detection probability must exceed 90% for primary users at very low signal-to-noise ratios (e.g., -114 dBm for DTV signals).

The absolute operational constraint of any DSA system. Secondary transmissions must not cause harmful interference to licensed primary users. This is enforced through strict regulatory thresholds.

• Interference Temperature Limit: A metric defined by the FCC quantifying the maximum tolerable RF power from all secondary sources at a PU receiver.
• Protection contours: Geographic zones around primary transmitters where secondary access is prohibited or power-limited.
• Hidden node mitigation: Cooperative sensing or database lookup is required because a single SU may be shadowed from a PU transmitter and falsely detect a hole.

Unlike static radios, a DSA-enabled cognitive radio dynamically adjusts its transmission parameters to fit the characteristics of the available spectrum hole and avoid interference at the new frequency.

• Transmit Power Control (TPC): Adjusting power to the minimum necessary level, reducing the interference footprint.
• Adaptive Modulation and Coding (AMC): Switching modulation schemes (e.g., QPSK to 64-QAM) based on the signal-to-noise ratio of the new channel.
• Bandwidth adaptation: Aggregating multiple narrowband holes or using a single wideband hole, depending on availability.
• Example: An SU vacating a 10 MHz channel and reconfiguring to transmit on two non-contiguous 5 MHz holes using carrier aggregation.

Cognitive radios do not have unrestricted freedom. A Policy Engine enforces a set of machine-interpretable rules that constrain the radio's autonomous decisions to ensure regulatory compliance and operational objectives.

• Regulatory policies: Enforce FCC/ITU rules for specific bands (e.g., max power limits, exclusion zones).
• Operational policies: Define user priorities, quality of service requirements, and security constraints.
• Policy reasoning: The cognitive engine must prove that a proposed action is compliant before execution.
• Example: A DSA policy might state: "In band 3.55-3.65 GHz, max EIRP is 23 dBm/10 MHz, and you must vacate within 60 seconds of receiving a SAS suspension order."

Advanced DSA systems incorporate machine learning to move from reactive to proactive spectrum access. By modeling historical usage patterns, the radio can predict future spectrum availability.

• Spectrum Occupancy Prediction: Using time-series models (e.g., LSTMs) to forecast idle periods, reducing sensing overhead and latency.
• Reinforcement Learning (RL): An agent learns an optimal channel selection policy through trial-and-error, maximizing throughput while minimizing collisions.
• Multi-Armed Bandit (MAB): A simplified RL model that balances exploration (sampling new channels) with exploitation (using the best-known channel).
• Benefit: Reduces the frequency of reactive handoffs, improving link stability for applications like video streaming.

The intelligent core of a cognitive radio that uses AI models to observe the RF environment, learn from it, and autonomously decide on optimal transmission parameters. It implements the observe-orient-decide-act (OODA) loop, processing inputs from spectrum sensing modules and directing the software-defined radio to execute actions. Modern cognitive engines leverage reinforcement learning and deep neural networks to adapt policies in real-time without human intervention.

The fundamental cognitive radio function of monitoring the electromagnetic environment to detect the presence or absence of primary user signals. Key techniques include:

• Energy detection: Simple threshold-based sensing
• Matched filter detection: Requires prior knowledge of primary signal
• Cyclostationary feature detection: Exploits periodicity in modulated signals
• Cooperative sensing: Multiple radios share observations to overcome the hidden node problem

A frequency band that is allocated to a primary user but is temporarily unused in a specific geographic location. Also called white space or spectrum opportunity, these gaps represent the exploitable resource for DSA systems. A hole is defined by three dimensions:

• Frequency: The specific band available
• Time: The duration of availability
• Space: The geographic area where the hole exists Secondary users must vacate immediately upon primary user return.

The process by which a secondary user seamlessly vacates its current frequency channel upon detecting a returning primary user and transitions ongoing communication to another available spectrum hole. Key requirements:

• Proactive handoff: Predicts primary user arrival using spectrum prediction models
• Reactive handoff: Triggers immediately upon primary user detection
• Target channel selection: Must find a new hole with sufficient bandwidth and expected dwell time Latency during handoff directly impacts quality of service for real-time applications.

A regulatory-approved, location-aware database that a cognitive radio queries to determine which TV white space frequencies are available for unlicensed use at its current geographic coordinates. The database contains:

• Protected contours of licensed TV broadcasters
• Registered wireless microphone venues
• Exclusion zones for radio astronomy sites Radios must report their location periodically and update their channel list accordingly. This approach provides deterministic protection without requiring real-time sensing.