Traditional cognitive radio networks operate under a strict hierarchy: primary users hold the exclusive license to a frequency band, while secondary users opportunistically scavenge idle spectrum. In underlay systems, secondary users must strictly limit their transmit power to avoid disrupting the primary user. In overlay systems, they must constantly sense the channel and vacate it the moment the primary user returns. This creates a zero-sum game where any secondary transmission inherently threatens primary performance.
As Rui Zhang 0006 demonstrates in Cognitive Wireless Powered Network: Spectrum Sharing Models and Throughput Maximization, we can break this zero-sum constraint by shifting from interference avoidance to cooperative relaying. Instead of treating secondary users as a source of noise, the primary network treats them as an infrastructure extension. Long Gao et al. frame this paradigm in Throughput and Delay Scaling in Supportive Two-Tier Networks as "active support": the secondary network actively improves the primary network's throughput and delay scaling, earning dedicated spectrum access as a direct reward.
How Relaying Creates Spectrum
To understand why this works, consider a primary user with a weak or fading channel to its base station. Because the channel is poor, the primary user must transmit at a low data rate, keeping the spectrum occupied for a long duration.
If a nearby secondary user acts as an amplify-and-forward or decode-and-forward relay, it can exploit its superior spatial position to boost the primary user's signal quality. Gan Zheng et al. show in Cooperative Cognitive Networks: Optimal, Distributed and Low-Complexity Algorithms that this cooperative relaying can reduce primary outage probabilities to near zero.
Crucially, accelerating the primary user's transmission alters the temporal dynamics of the network. As Fotis Foukalas et al. explain in Packet Relaying Control in Sensing-based Spectrum Sharing Systems, forwarding primary packets clears the primary network's queue much faster. This rapid queue clearance artificially creates more "idle" spectrum states.
Similarly, Ahmed El Shafie et al. demonstrate in Protocol Design and Stability Analysis of Cooperative Cognitive Radio Users that by relaying failed primary packets, secondary users reduce the total duration of primary activity. This expands the overall stability region of the network and unlocks clean transmission windows for the secondary users.
The Currency of Cooperation
The exchange of relaying services for spectrum access can be structured in several ways, depending on how resources are partitioned:
- Time and Subcarrier Allocation: In a time-division framework, a primary user leases a fraction of its transmission frame to a secondary user. Tao Yang et al. in Cooperative Spectrum Sharing Between D2D Users and Edge-Users: A Matching Theory Perspective utilize a three-phase frame structure: broadcast, relaying, and private transmission. The secondary user uses the second phase to forward primary data and is rewarded with exclusive spectrum use in the third phase. Alternatively, Dawei Wang et al. show in Reciprocally Benefited Spectrum Access Scheme With Joint Power and Subcarrier Allocation in a Software-Defined Network how this can be done in the frequency domain by allocating dedicated subcarriers to the secondary user during the relaying process.
- Concurrent and Parallel Transmission: Instead of separating transmissions in time, networks can use spatial or power-domain multiplexing. Ping Xie et al. propose a protocol in Cooperative spectrum leasing using parallel communication of secondary users where one secondary node relays primary traffic while other secondary nodes simultaneously transmit private data in parallel. In the power domain, Lu Lv et al. demonstrate in Cognitive Non-Orthogonal Multiple Access with Cooperative Relaying: A New Wireless Frontier for 5G Spectrum Sharing that a strong secondary user can use Non-Orthogonal Multiple Access (NOMA) to multiplex its own signal with the primary user's signal, providing a reliability benefit to the primary user while gaining opportunistic access.
- Information and Energy Symbiosis: For energy-constrained Internet of Things (IoT) devices, the currency can include power. Daqian Zhao et al. show in A Novel Information and Energy Cooperation Transmission Scheme in Cognitive Spectrum Sharing-Based D2D Communication Systems that secondary devices can harvest radio-frequency energy directly from the primary signal they are helping to forward, sustaining their own communication entirely through the cooperative act.
- Physical Layer Security: Cooperation is not limited to data forwarding. When a primary network faces eavesdropping threats, secondary users can act as friendly jammers. Ding Xu in Resource Allocation for Secure Communications in Cooperative Cognitive Wireless Powered Communication Networks and Alireza Shamsoshoara et al. in A reputation-based stackelberg game model to enhance secrecy rate in spectrum leasing to selfish IoT devices demonstrate that providing protective jamming services to enhance the primary network's secrecy rate serves as a highly effective payment mechanism for spectrum access.
Algorithmic Coordination: Games, Matching, and Learning
Because primary and secondary users are often independent or selfish entities, establishing cooperation requires robust algorithmic frameworks to handle pricing, pairing, and resource allocation.
Game Theory
The interaction is frequently modeled as a leader-follower dynamic. Michael Bloem et al. in A Stackelberg game for power control and channel allocation in cognitive radio networks introduce a virtual pricing mechanism where the primary user regulates access by adjusting a cost parameter based on its own signal quality. This allows for non-binary spectrum sharing where secondary users can remain on a channel as long as they mitigate their interference to meet the primary user's quality-of-service requirements.
Matching Theory
When multiple primary users and secondary users coexist, they must be paired optimally. Nima Namvar leverages a distributed matching game framework in Spectrum sharing in cooperative cognitive radio networks: A matching game framework to ensure stable, mutually beneficial pairings between license holders and relays based on their localized channel conditions and utility functions.
Reinforcement Learning
In highly dynamic environments like Unmanned Aerial Vehicle (UAV) networks, centralized coordination is impractical. Alireza Shamsoshoara et al. demonstrate in Distributed Cooperative Spectrum Sharing in UAV Networks Using Multi-Agent Reinforcement Learning that multi-agent reinforcement learning allows individual UAVs to autonomously balance the cost of helping a primary user against their own mission objectives, achieving optimal spectrum sharing without a central controller.
Key Foundational Work
Coalition Games with Cooperative Transmission: A Cure for the Curse of Boundary Nodes in Selfish Packet-Forwarding Wireless Networks (2007) — contributed 1 concept elements to this direction.
Delay-Throughput Tradeoff for Supportive Two-Tier Networks (2008) — contributed 1 concept elements to this direction.
A stackelberg game for power control and channel allocation in cognitive radio networks (2007) — contributed 1 concept elements to this direction.