Cooperative Spectrum Sensing

Exploits geographically distributed sensors to combat multipath fading and shadowing effects. When one node experiences a deep fade, others with uncorrelated channel conditions maintain detection integrity. This diversity transforms the probability of missed detection from an exponential function of individual SNR to a much steeper function of the number of cooperating nodes, effectively eliminating the hidden node problem.

A centralized fusion center aggregates local sensing observations from all cooperating secondary users to form a global decision about spectrum occupancy. The fusion center implements a combining rule—ranging from simple logical operations to optimal likelihood ratio tests—to synthesize distributed information. This architecture shifts computational complexity from resource-constrained edge nodes to a more capable central processor.

Individual cognitive radios transmit binary local decisions (occupied/vacant) to the fusion center, which applies a K-out-of-N voting rule. Common implementations include:

• OR Rule: Declares occupied if any single node detects a signal—maximizes detection probability at the cost of higher false alarms
• AND Rule: Requires unanimous agreement—minimizes false alarms but risks missed detections
• Majority Rule: Balances the trade-off, requiring more than half of nodes to concur Bandwidth-efficient but discards valuable soft information.

Nodes transmit raw sensing statistics or quantized test metrics rather than binary decisions. Common approaches include:

• Equal Gain Combining (EGC): Fusion center sums all received energy measurements with equal weights
• Maximal Ratio Combining (MRC): Weights each node's contribution by its instantaneous SNR for optimal performance
• Likelihood Ratio Test (LRT): Statistically optimal fusion using complete probability distributions Soft fusion consistently outperforms hard decision methods, approaching centralized detection performance as reporting channel quality improves.

The control channel over which sensing data is transmitted to the fusion center introduces its own imperfections. Imperfect reporting channels suffering from fading, noise, or bandwidth limitations can degrade or even negate cooperation gains. Censoring schemes mitigate this by allowing only nodes with sufficiently reliable observations to report, conserving bandwidth while preserving detection performance. Quantized soft fusion balances the bandwidth-performance trade-off by transmitting multi-bit sensing metrics rather than full analog values.

Cooperative sensing networks are vulnerable to spectrum sensing data falsification (SSDF) attacks, where malicious nodes report fraudulent observations to manipulate the global decision. Byzantine defense mechanisms include:

• Reputation-based weighting: Nodes with historically consistent reports receive higher trust scores
• Outlier detection: Statistical tests identify and exclude anomalous reports before fusion
• Consensus algorithms: Distributed agreement protocols that tolerate a fraction of adversarial nodes These safeguards are critical for deployment in contested electromagnetic environments.

A central processing node that aggregates local sensing data from distributed cognitive radios to form a global decision. The fusion center applies a combining rule to mitigate individual sensor errors caused by fading or shadowing. Key responsibilities:

• Synchronizing reporting intervals
• Weighting sensor reliability
• Broadcasting the final spectrum occupancy verdict
• Acting as the single point of integration for network-wide situational awareness

A bandwidth-efficient strategy where individual nodes transmit only a binary local decision (0 or 1) to the fusion center. The center then applies a logical rule to generate the final verdict. Common combining rules:

• AND Rule: All nodes must agree; minimizes false alarms but increases missed detections.
• OR Rule: A single positive vote triggers detection; highly conservative against interference.
• K-out-of-N Rule: Requires K concurring votes, balancing the trade-off between sensitivity and specificity.

A high-fidelity strategy where nodes transmit raw or quantized sensing statistics (e.g., energy levels, likelihood ratios) to the fusion center. Preserving this granular data enables superior detection performance compared to hard fusion. Common techniques:

• Equal Gain Combining (EGC): Sums all energy measurements with uniform weight.
• Maximal Ratio Combining (MRC): Weights measurements by their instantaneous signal-to-noise ratio.
• Likelihood Ratio Test (LRT): The mathematically optimal fusion rule assuming known channel statistics.

A critical degradation in sensing reliability caused when a cognitive radio is shadowed by terrain or buildings relative to a transmitting primary user. The sensor fails to detect the active transmitter, leading to a missed detection and potential harmful interference. Mitigation strategies:

• Deploying geographically diverse cooperative sensors
• Utilizing relay nodes to forward sensing data around obstacles
• Implementing sensitivity margins in detection thresholds to account for deep fades

An adaptive threshold-setting algorithm that maintains a fixed probability of false alarm despite dynamic background noise. CFAR is essential for energy detectors operating in fluctuating environments. Common variants:

• Cell-Averaging CFAR (CA-CFAR): Estimates noise from adjacent range bins.
• Ordered-Statistic CFAR (OS-CFAR): Robust against multiple interfering targets by selecting a ranked noise estimate.
• Greatest-of CFAR (GO-CFAR): Designed to control false alarms at clutter edges.

A graphical plot illustrating the trade-off between probability of detection (Pd) and probability of false alarm (Pfa) as a detector's discrimination threshold varies. The ROC curve is the standard metric for benchmarking cooperative sensing performance. Key insights:

• The area under the curve (AUC) quantifies overall detector efficacy.
• Soft fusion curves consistently dominate hard fusion curves.
• The operating point is chosen based on the regulatory requirement for primary user protection.