Enhancing Flexible Unequal Error Control Method to Improve Soft Error Tolerance
Because of soft error, the reliability of modern computing systems is facing many problems. As the area of application increases, the demand for detecting and correcting soft errors increases proportionally. Several error-correcting methods have been deve
- PDF / 619,872 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 58 Downloads / 212 Views
Abstract Because of soft error, the reliability of modern computing systems is facing many problems. As the area of application increases, the demand for detecting and correcting soft errors increases proportionally. Several error-correcting methods have been developed to solve this problem. However, high redundancy, low correction ability, large circuit complexity, space and time overhead of most of the existing methodologies are creating the demand for enhancement in the area of soft error tolerant researches. To improve the soft error tolerability in computer systems, we have enhanced the performance of Flexible Unequal Error Control (FUEC) method in this paper. Our method has increased the correction ability for six more erroneous bits than FUEC. Moreover, the proposed method has low bit-overhead and high correction coverage compared to the previous methods. Keywords Soft error tolerance · Reliability · Flexible unequal error control · Parity check matrix · Generator matrix
1 Introduction With the advancement of technology, computer systems and circuits are becoming more complex [1]. The size of the computer chip is decreasing and so the circuits are becoming much denser. Although more transistors are present on a single chip, Md. Atik Shahariar (B) · M. Islam · M. S. Sadi · S. Ghosh Khulna University of Engineering & Technology, Khulna 9203, Bangladesh e-mail: [email protected] M. Islam e-mail: [email protected] M. S. Sadi e-mail: [email protected] S. Ghosh e-mail: [email protected]
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 M. S. Kaiser et al. (eds.), Proceedings of International Conference on Trends in Computational and Cognitive Engineering, Advances in Intelligent Systems and Computing 1309, https://doi.org/10.1007/978-981-33-4673-4_39
489
490
M. Atik Shahariar et al.
there is a decrease in the supply of power and noise margins. As a result, electronic systems are becoming more vulnerable to radiation-induced errors [2], [3]. Memory devices can easily become corrupted due to this radiation induction [4]. This can lead to corruption of data or malfunction in the computer circuits [5]. Such errors can be more catastrophic in critical environments such as space stations, space shuttle, power plants, etc. [6], [7]. Nowadays, computer chips are being used in medical science to monitor patients in intensive care units, where even a little change of data can be very harmful to the patient. As soft error poses a high threat to the system, necessary steps should be taken to make the system error resistant or less vulnerable to soft error. There are some error protection techniques based on hardware but they suffer from high area, time, and power overheads [8]. Some prior works known as Error Correction Codes (ECCs) [9] have been done for detecting and correcting soft errors. Examples of ECCs are BoseChaudhuri-Hocquenghem (BCH) [5], Golay codes [10], SEC–Double-adjacent Error Detection (SEC–DED) [11], Matrix Codes [12], Hamming code [13],
Data Loading...
