Table 2. Sybil attack detections in literature.
| Citation | Detection method | Performance evaluation metric | Strengths |
|---|---|---|---|
| (Saravanakumar et al., 2022) | *Encryption Method | *Packet Loss Ratio (PLR) | *Achieve higher data delivery with a minimum delay |
| *Computation Overhead | |||
| *Artificial deep neural networks | *Throughput | ||
| *End-to-End Delay | |||
| (Singh & Saini, 2021b) | *Received Signal Strength (RSS) | *Energy Consumption | *Utilized for the usage of energy consumption, effectiveness of detecting Sybil attacks inside clusters. |
| *PCTBC: Power Control Tree Based Cluster Approach | |||
| (Angappan et al., 2021) | *Received Signal Strength (RSS) | *True Detection Rate (TDR) | *Highly efficient in detection ratio, energy utilisation, memory usage, computation, and Communication requirement |
| *Communication Overhead | |||
| *Localization Based | *Energy Consumption | ||
| *Computation Overhead | |||
| *Cluster based | |||
| *Memory Overhead | |||
| (Raghav, Thirugnansambandam & Anguraj, 2020) | Bee algorithm | *Efficiency | *Better data efficiency with security |
| (Dong, Zhang & Zhou, 2020) | Distance Vector Hop (DV Hop) Algorithm | *Localization Error | *Improve the security of the node localization in WSN. |
| *Reduces the average localization error by 3% than the traditional DV Hop. | |||
| (Wang & Feng, 2020) | Received Signal Strength (RSS) | *True Detection Rate (TDR) | *Resolve time difference |
| (Jamshidi et al., 2019a) | *Learning Automaton (LA) | *True Detection Rate (TDR) | *Detects 100% of Sybil nodes |
| *Communication Overhead | |||
| *False Detection Rate (FDR) | |||
| *Computation Overhead | |||
| *Client puzzles theory | *5% false detection rate | ||
| (Jamshidi et al., 2019c) | Received Signal Strength (RSS) | *True Detection Rate (TDR) | *Detect 99.8% of Sybil nodes |
| *Communication Overhead | |||
| *0.008% false detection rate | |||
| *False Detection Rate (FDR) | |||
| *True detection rate | |||
| (Jamshidi et al., 2019b) | Information based | *True Detection Rate (TDR) | *Detect 99% of Sybil nodes |
| *5% false detection rate | |||
| *False Detection Rate (FDR) | |||
| (Li & Cheffena, 2019) | *Multi Kernel Based Expectation Maximization (MKEM) | *Detection Accuracy | *High accuracy on detecting Sybil |
| *Gap statistical analysis method | |||
| *Can guarantee the detection accuracy even if the number of Sybil attackers increases. | |||
| *Kernel parameter optimization | |||
| (Vaniprabha & Poongodi, 2019) | *Elliptic Curve Cryptography (ECC) | *False Positive Rate (FPR) | *Enhances data accuracy of the collected data with minimized delay. |
| *97% packet delivery ratio | |||
| *0.8 ms packet dropping | |||
| *2.4 ms for key generation | |||
| *Packet delivery Ratio (PDR) | *0.96 ms for secret key exchange. | ||
| *Transmission model | |||
| (Shehni et al., 2018) | Watchdog | *True Detection Rate (TDR) | *Low extra Communication overhead |
| *Communication Overhead | *Detection measures of performance *True | ||
| Detection Rate (TDR) | |||
| *False Detection Rate (FDR) | *False Detection Rate (FDR) | ||
| *Network Performance | |||
| (Yuan et al., 2018) | *Received Signal Strength (RSS) | *True Detection Rate (TDR) | *Requires minimal overhead |
| *Works well based on the received signal strength | |||
| *Effective in detecting and defending against sybil attacks | |||
| *Communication Overhead | |||
| *Localization Based | *High detection rate | ||
| (Wang, Wen & Zhao, 2018) | *Deep learning | *Detection Accuracy | *94.39% classification accuracy |
| *Stacked Denoising Autoencoder (SDA) | |||
| *Back propagation algorithm | |||
| *Complex network theory | *More robust and efficient even in the existent of huge baneful beacons. | ||
| (Jamshidi et al., 2017) | Behaviour Based | *True Detection Rate (TDR) | *Identify 94% of Sybil nodes |
| *False Detection Rate (FDR) | |||
| *False detection rate | |||
| (Li et al., 2017) | Localization Based | *Localization Error | *Superior in terms of malicious nodes identification and performance improvement. |
| (Razaque & Rizvi, 2017) | Data Aggregation | *Detection Accuracy | *Prevent and detect both sinkhole and Sybil attacks in the presence of static and mobile sensor nodes |
| *Communication Overhead | |||
| *Energy Consumption | |||
| (Raja & Beno, 2017) | Security mechanism and Fujisaki Okamoto (FO) algorithm | *Throughput | *Increase the performance of the network. |
| *Energy Consumption | |||
| *Packet Delivery Ratio (PDR) | |||
| (Alsaedi et al., 2017) | Energy Trust System (ETS) | *True Detection Rate (TDR) | *Robust in detecting sybil attacks in terms of the true and false positive rates. *70% detection at the first level, which significantly increases to 100% detection at the second level *Reduces communication overhead, memory overhead, and energy consumption |
| *Energy Consumption *Communication Overhead *Memory Overhead *False Positive Rate (FPR) *Resource Consumption | |||
| (Khan & Khan, 2016) | Signed response (SRES) authentication | *Power Consumption | *Lesser computational cost |
| *Computational Cost | *Power consumption | ||
| (Singh, Singh & Singh, 2016a) | Trust Based | *True Detection Rate (TDR) | *Significant attack detection rate |
| (Kumar et al., 2016) | *Signature Based | *Throughput | *Improves the data reliability |
| *Packet delivery Ratio (PDR) | |||
| *Rule base | |||
| (Vamsi & Kant, 2016) | Neighbour Based | *False Positive Rate (FPR) | *Robust in detecting Sybil attacks with very low false positive and false negative rates. |
| *False Negative Rate (FNR) | |||
| (Saleem et al., 2016) | Encryption Based | *Resource Consumption | *Efficiently protect WSNs Sybil attacks. |
| *Energy Consumption | |||
| *Computation Overhead | |||
| *Packet Delivery Ratio (PDR) | |||
| *Computational Cost |