CHAPTER 5
CONCLUSION AND FUTURE RESEARCH DIRECTIONS
From the above analysis, it is observed that the data reliability over health care applications is achieved effectually. The methods applied above help resolve the challenging factors over WBASN, which is a kind of sensor network. The experimentation with DCT and Huffman coding gives only break time to people under every category. It is performed to predict whether the data is compressed effectually. Therefore, clustering the workforce is performed optimally. When older people get tired, huge packets are transmitted with higher bandwidth. Therefore, appropriate compression techniques are executed to enhance the sensor lifetime. Integrating the compression with the above scenario can reduce the sensor lifetime because of the hardware complexity. This model decreases the memory space requirement and network bandwidth. Here, metrics like MSE, CR, and SNR are used for comparison with other models.
Subsequently, the compressed data is encrypted with the Improved Secure Symmetric Force Encryption (SSFE) algorithm for enhancing the privacy and security of the data. This SSFE uses basic mathematical operations to protect and reduce the complexity encountered in existing approaches like AES and LEA. Metrics like code and input size, execution time, throughput, and energy consumption are measured. Therefore, it is observed that the anticipated compression and encryption process works effectually with the EEG data, and better simulation is achieved with WBSANs. In the future, the anticipated model is extended in various real-time applications with reduced computational complexity in hardware components. In the end, a productive routing protocol has to be designed using a Meta-heuristic optimization approach to acquire global outcomes.
5.1. Research Findings
Based on the proposed compression and encryption process, the following are the research findings:
- The construction time of HC-DCT is 0.86 seconds, PRD is 9.56, CR is 11.27, MSE is 0.2, and PSNR is 37.89. It shows that the error rate is comparatively lesser with better PRD. Thus, this model helps in efficient signal reconstruction and compression.
- The execution time of SSFE is 0.000285 seconds, which is comparatively lesser than the AES and LEA, respectively.
- Similarly, the energy consumption of SSFE is 0.0001102J, which is lesser than the techniques above.
- The throughput of SSFE is 87500.55 bytes per second, which are higher than AES and LEA. Also, the power consumed during the encryption process is 0.13 mA with a 5.02 V voltage drain.
5.2. Research Summary
The ultimate target of this research is divided into two major parts. They are 1) Compression and 2) Encryption.
1) In this dissertation, a framework is anticipated for selecting an appropriate dictionary for EEG signals, which is known as the Discrete Cosine Transform (DCT). It is adopted for the reconstruction of EEG signals effectually and makes the transmission over the network in an efficient manner. After signal reconstruction, the compression process needs to be performed to make communication easier. Here, Huffman coding is employed for attaining a higher compression ratio.
2) The secondary target of this research aims at analyzing the data for the encryption process. Therefore, the medical data can be provided securely and privately. Followed by the efficient execution of compression sensing over the EEG signal, the data needs to be encrypted and decrypted with reduced computational complexity. Thus, the Secure Symmetric Force Encryption algorithm is employed to make this process simpler and attain better results than various prevailing approaches.
During the EEG signal transmission process, the energy consumption over WBASN is also analyzed with the exploration of proposed model efficiency. The investigations are performed by numerical analysis for realizing the computational complexity and implementation of compression and encryption techniques over the WBASN. The anticipated model facilitates better compression and encryption of data with reduced computational and time complexities. Therefore, it gives substantial effort to save the node’s battery lifetime and improves the overall performance.
5.3 Future research directions
There are diverse possible scopes for future research direction based on this dissertation. Some are listed below.
- The possibility of applying this work for the clinical evaluation of the EEG signal and efficiently reconstructs the signal for analyzing the patients’ present condition.
- This method can be used as an efficient Clinical Decision Support System during a time of disaster.
- Similarly, appropriate classifications of EEG signals are essential to predict the disease and enhance the treatment process.
- The anticipated model can be applied under hardware execution, and validation can be performed for measuring the system efficiency.
- The algorithmic model can be further improved, which contributes to the overall system efficiency.
- The signal correlation has to be measured among the multi-channels.
- Adoption of various algorithms for analyzing the EEG features.
- Hybridization techniques can be used to improve system performance.