TL;DR
This paper tackles the challenge of secure wireless communication in untrusted relay networks that also harvest energy. The authors propose and analyze four different transmission and reception schemes, deriving closed-form expressions for secrecy and connection outage probabilities, and introducing the Effective Secrecy Throughput (EST) metric to balance security and reliability. They find that while some schemes offer better security, others provide superior reliability, and an optimal antenna configuration exists for maximizing EST.
What Problem Does This Paper Address?
Wireless communication systems are inherently vulnerable to eavesdropping. Traditional security measures often rely on complex cryptographic methods. Physical Layer Security (PLS) offers a complementary approach by leveraging the inherent randomness of wireless channels to secure communications. However, in cooperative relay networks, where intermediate nodes (relays) help transmit data, the trustworthiness of these relays becomes a critical concern. If a relay is untrusted, it could potentially eavesdrop on the communication. Furthermore, many modern wireless devices are energy-constrained, necessitating energy harvesting capabilities. This paper addresses the combined challenges of ensuring secure data transmission in amplify-and-forward (AF) untrusted relay networks that also harvest energy, while simultaneously considering the trade-off between security and reliable data delivery.
Methods
The authors investigate four distinct transmission and reception schemes for a system comprising a source (S), a destination (D), and an untrusted energy-harvesting relay (R). Both S and D are equipped with multiple antennas, while R has a single antenna. The schemes differ in how S transmits information and how D receives it:
- Transmission Schemes:
- Maximal Ratio Transmission (MRT): Utilizes all transmit antennas at S to maximize the received signal strength at R.
- Transmit Antenna Selection (TAS): Selects a single transmit antenna at S to reduce complexity and potentially improve performance.
- Reception Schemes:
- Maximal Ratio Combining (MRC): Combines signals from multiple antennas at D to maximize the received signal-to-noise ratio (SNR).
- Receive Antenna Selection (RAS): Selects a single receive antenna at D.
The paper derives closed-form expressions for:
- Secrecy Outage Probability (SOP): The probability that the achievable secrecy rate falls below a required threshold, indicating a security failure.
- Connection Outage Probability (COP): The probability that the destination cannot correctly decode the information, indicating a reliability failure.
- Effective Secrecy Throughput (EST): A novel metric that combines the secrecy rate and the probability of successful transmission, providing a comprehensive measure of system performance.
The authors also analyze the asymptotic performance of EST in the high transmit SNR regime and investigate the impact of various parameters like power allocation, power splitting ratio, and antenna numbers.
Datasets
Not discussed in the paper. The evaluation is based on analytical derivations and Monte Carlo simulations.
Key Insights and Ideas
- EST as a Holistic Metric: The introduction of the Effective Secrecy Throughput (EST) is a key contribution, providing a single metric to balance the often-conflicting goals of security (low SOP) and reliability (low COP).
- Trade-off Between Security and Reliability: As transmit SNR increases, security (SOP) improves due to jamming, but reliability (COP) can decrease. EST captures this trade-off.
- MRT vs. TAS: While MRT generally offers better reliability, TAS can be superior in terms of security, especially at high SNRs. This is attributed to MRT potentially leaking more information to the untrusted relay.
- MRC vs. RAS: MRC consistently outperforms RAS in terms of reliability, as expected from its signal combining capabilities.
- Optimal Antenna Configuration: For both MRT and TAS schemes, there exists an optimal number of antennas at the source (Ns) that maximizes the EST. Beyond this optimum, increasing antennas can degrade security.
- Impact of Power Allocation (β): There's an optimal power allocation factor (β) between the source and destination that maximizes EST. Too little power for the source hinders reliable communication, while too much for the destination compromises security.
- Energy Harvesting Allocation (ω): Increasing the proportion of harvested energy used for forwarding information (ω) improves EST, as it also reduces the energy available for eavesdropping by the relay.
Key Results
- Scheme Performance:
- EST: MRT-MRC generally outperforms other schemes across various parameters. However, in the high SNR regime, TAS-MRC becomes superior.
- Security (SOP): TAS schemes (TAS-RAS, TAS-MRC) offer better security than MRT schemes (MRT-RAS, MRT-MRC).
- Reliability (COP): MRC schemes (TAS-MRC, MRT-MRC) offer better reliability than RAS schemes (TAS-RAS, MRT-RAS).
- Asymptotic Analysis: In the high SNR regime, the EST of all schemes converges to a constant value, indicating that further increases in transmit power are not beneficial.
- Optimal Parameters: The paper identifies optimal values for the power allocation factor (β), power splitting ratio (p), and number of antennas (Ns) that maximize the EST for the proposed schemes. For instance, optimal Ns values were found to be 4 or 17.
Limitations and Future Work
- Untrusted Relay Assumption: The paper assumes the relay is untrusted. Future work could explore scenarios with partially trusted relays or multiple untrusted relays.
- Channel State Information (CSI): The analysis assumes perfect CSI is available. Investigating the impact of imperfect CSI would be a valuable extension.
- Single Relay: The current model focuses on a single untrusted relay. Extending the analysis to multiple untrusted relays, or considering different relaying protocols (e.g., decode-and-forward), could be explored.
- Specific Fading Models: While Rayleigh fading is considered, analyzing performance under other fading distributions (e.g., Nakagami-m) could provide broader insights.
- Interference: The current model primarily considers additive white Gaussian noise (AWGN). Investigating the impact of co-channel interference or other forms of interference would be relevant for more complex scenarios.
Summary updated