Full-Duplex Cognitive Radio: A New Design Paradigm for Enhancing Spectrum Usage

I. INTRODUCTION

• Half-duplex Listen-before-Talk (LBT) sacrifices transmission time for sensing and cannot detect primary-user state changes during secondary-user transmission.These blind intervals can cause collisions when primary users arrive and spectrum waste when they leave.Source: §I · p. 2
• Full-duplex (FD) radios can send and receive on the same time and frequency resources, offering a way to sense continuously while transmitting.Self-interference once made FD impractical, but progress in cancellation has renewed its promise for wireless networks.Source: §I · p. 2
• The proposed Listen-and-Talk (LAT) protocol equips secondary users with FD radios to sense the target band and transmit or remain silent based on sensing results.FD-CR requires new signal processing, resource-allocation, and protocol designs, including transmit-power optimization for throughput.Source: §I · pp. 2–3
• In FD-CR, residual self-interference degrades sensing SINR, whereas half-duplex sensing avoids that interference but occupies only a small fraction of each slot.FD transmission can continue while primary users are absent, but transmit power must be constrained to maintain acceptable sensing performance.Source: §I · p. 3

A. Full-Duplex Communication System

• A full-duplex node transmits and receives signals in the same time interval and frequency resources using separate transmit and receive antennas.Practical systems retain residual self-interference from leakage between the transmit and receive RF units; its variance is proportional to transmit power.Source: A. Full-Duplex Communication System · pp. 3–4

B. Conventional Cognitive Radio Communication System

• In a conventional cognitive-radio network, each secondary user's slot begins with spectrum sensing, followed by scheduled data transmission if the channel is judged idle.Sensing may use energy detection, matched filtering, cyclostationary features, or cooperative detection.Source: B. Conventional Cognitive Radio… · p. 4

A. The LAT Protocol

• In LAT, SU1 continuously senses with Ant1 and uses Ant2 to transmit data when a spectrum hole is detected, deciding at each slot's end whether to transmit in the next slot.The model contains one primary user and one secondary transmitter-receiver pair.Source: A. The LAT Protocol · p. 5
• The four spectrum-use cases distinguish normal primary-only or secondary-only use from spectrum waste when neither transmits and collision when both transmit.Spectrum-waste and collision ratios are identified as key sensing-performance metrics.Source: A. The LAT Protocol · p. 5
• When SU1's transmit antenna is idle, its sensing antenna receives only the potential primary-user signal and noise; transmission introduces residual self-interference that may sharply reduce sensing SINR.The residual self-interference is irreducible in this setup and strongly affects sensing performance.Source: A. The LAT Protocol · pp. 5–6

B. Key Design Parameters

• The design constrains SU interference to the primary network by collision probability, then derives sensing strategy, thresholds, and slot length from that constraint.Source: B. Key Design Parameters · p. 6

C. Performance Analysis

• Saturated SU throughput depends on channel rate R and the fraction of spectrum holes not wasted, (1-Pw).R denotes channel transmission rate and Pw the ratio of wasted spectrum hole.Source: 1) Power-throughput Tradeoff · p. 6

1) Power-throughput Tradeoff:

• Higher transmit power raises channel rate but RSI degrades sensing, increasing wasted spectrum holes and reducing the usable-hole fraction under the same collision constraint.Source: 1) Power-throughput Tradeoff · pp. 6–7
• Lower RSI shifts the optimal transmit power upward and raises corresponding throughput; sufficiently large RSI makes sensing unreliable and removes the local optimum.At high power, throughput can approach the C = 0.5R line associated with 0.5 spectrum waste.Source: 1) Power-throughput Tradeoff · pp. 7–8
• In low power, SU throughput rises then falls to a local optimum; in high power, it rises monotonically, so maximum power may not maximize throughput.The figure compares analytical curves with simulation circles.Source: Figure 3 · p. 7, 1) Power-throughput Tradeoff · p. 8

• The LBT–LAT comparison uses the same model for fairness.Source: 2) Comparison between the LAT and… · p. 8
• LAT performs better than LBT in most cases because it uses spectrum holes more efficiently and reacts promptly to primary-user state changes.Source: 2) Comparison between the LAT and… · p. 9
• LAT may not outperform LBT when secondary transmit power is sufficiently large and residual self-interference becomes unbearable.Source: 2) Comparison between the LAT and… · p. 9

IV. FULL-DUPLEX COGNITIVE RADIO APPLICATIONS

• The section introduces key applications of FD cognitive radio systems.Source: §IV · p. 9

A. Distributed Scenarios

• Without central coordination, each transmitting FD-CR user must contend for channel access and resolve possible collisions.Source: A. Distributed Scenarios · p. 9

1) Distributed Spectrum Access Scheme:

• Continuous sensing lets non-cooperative FD-SUs detect PU and SU state changes during transmission and back off immediately after collisions.The proposed distributed dynamic spectrum access scheme may reduce spectrum waste and collision length.Source: 1) Distributed Spectrum Access Scheme · p. 10

2) Full-Duplex MIMO System:

• A two-node FD MIMO system partitions antennas among sensing, transmission, and reception, enabling bidirectional communication while sensing but creating complicated antenna interference.When all three antenna groups are used, the system supports bidirectional communication while sensing.Source: 2) Full-Duplex MIMO System · p. 10

B. Centralized Scenarios

• The centralized scenario places an access point in a coverage area to establish connections with mobile users.Source: B. Centralized Scenarios · p. 10

1) Full-Duplex Cognitive Access Point System:

• A cognitive OFDM access point uses one antenna to sense PU spectrum and another to serve SUs, so downlink RSI affects sensing results.Assigning SUs to subcarriers and adjusting transmit power are essential to system performance.Source: 1) Full-Duplex Cognitive Access Point… · p. 10

2) Full-Duplex Cognitive Relay System:

• In the FD cognitive relay, source transmission, relay sensing, and forwarding of previously received signals occur in the same time-frequency resource.The relay uses three antennas: one for sensing, one for reception, and one for transmission; its spectral efficiency can improve over a traditional HD relay.Source: 2) Full-Duplex Cognitive Relay System · pp. 10–11
• The relay's sensing and data-receive antennas each receive a combination of PU, source, and RSI signals because it operates in the same frequency band.Source: 2) Full-Duplex Cognitive Relay System · p. 11

V. RESEARCH PROBLEMS

• FD-CR research problems involve managing space, time, frequency, power, and node resources while accounting for performance effects from RSI.Source: §V · p. 11

1) Spectrum Sensing:

• High transmit power can make sensing degradation unbearable; cooperative sensing combines several SUs' results, but transmitting-SU interference makes local sensing differ from the conventional HD case.Source: 1) Spectrum Sensing · p. 12
• FD-CR needs wideband sensing to exploit opportunities across broad frequency ranges because RSI changes the sparsity underlying conventional wideband sensing.Source: 1) Spectrum Sensing · p. 12

2) Multiple Antenna Techniques:

• Multiple antennas at FD-CRs can be used for beamforming and antenna selection to improve secondary-network performance.Source: 2) Multiple Antenna Techniques · p. 12
• Transmit beamforming can direct information toward desired users while reducing interference to the node's own sensing signals, with system sum rate jointly optimized.A multi-antenna FD cognitive AP may support multiple downlink transmissions while maintaining reliable sensing using antenna structure to minimize interference.Source: 2) Multiple Antenna Techniques · p. 12
• Antenna selection assigns groups to sensing, transmission, and reception; relays can adapt these choices to instantaneous channel conditions for reliable sensing and maximum SINR.The selection problem becomes more complex as the number of antennas increases.Source: 2) Multiple Antenna Techniques · p. 13

B. Dynamic Spectrum Access and Management

• Dynamic spectrum access lets a cognitive wireless node adapt transmission and reception as the radio environment changes.Source: B. Dynamic Spectrum Access and… · p. 13

1) Distributed Dynamic Spectrum Access:

• FD-CR SUs can detect both primary-user presence and collisions with other SUs during transmission, reducing collision duration, though RSI may degrade collision-detection accuracy.Distributed access still requires SUs to independently gather, exchange, and process environmental information when a central controller is unavailable.Source: 1) Distributed Dynamic Spectrum Access · p. 13

2) Centralized Dynamic Spectrum Access:

• In centralized access, an AP or relay coordinates spectrum decisions; subcarrier allocation must account for tolerable RSI to preserve reliable sensing.The AP matches subcarriers with downlink SUs, while relay systems allocate subcarriers for multiple secondary source-destination pairs.Source: 2) Centralized Dynamic Spectrum Access · pp. 13–14

3) Power Control:

• Power control in FD-CR must balance system performance against RSI-related sensing degradation, with designs depending on whether power constraints are total or individual.The power-control objective and solution differ across FD-CR system types.Source: 3) Power Control · p. 14
• In FD cognitive MIMO, sensing, receiving, and transmitting antennas are separated into sets, so RSI-aware optimal power pouring can differ from conventional water-filling.The described MIMO system uses individual power constraints.Source: 3) Power Control · p. 14
• For FD relays, greater transmit power raises destination SNR but can reduce sensing accuracy and blur source signals; APs instead split total power across subcarriers while jointly considering sensing and transmission.Relay power control has an individual constraint, whereas the AP scenario uses a total-power constraint.Source: 3) Power Control · p. 14

C. Multiple Networks Coexistence

• Heterogeneous networks sharing a band can suffer severe performance degradation from uncoordinated mutual interference, and FD techniques may make their coexistence more flexible.The section identifies reducing RSI impact and enabling efficient spectrum-access management as remaining research problems.Source: C. Multiple Networks Coexistence · pp. 14–15

VI. CONCLUSIONS

• Analytical and simulated results indicate that the proposed LAT protocol can efficiently improve spectrum utilization.The article also discusses FD-CR applications in centralized and distributed scenarios and outlines associated signal-processing and spectrum-access problems.Source: §VI · p. 15
• Future work may study CRNs with multiple PUs and SUs and apply economic theories to dynamic spectrum access and spectrum trading.Source: §VI · p. 15