- Neha

Ms. - Neha

University of Newcastle Australia, Newcastle, New South Wales, Australia
orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 February 2025
Verified
Conference Paper
Closed access
- Neha- Neha
- Neha
TBTarunpreet Bhatia
Tarunpreet Bhatia
Intrusion Detection Systems (IDS) are critical components in safeguarding 5G/6G networks from both internal and external cyber threats. While traditional IDS approaches rely heavily on signature-based methods, they struggle to detect novel and evolving attacks. This paper presents an advanced IDS framework that leverages adversarial training and dynamic neural networks in 5G/6G networks to enhance network security by providing robust, real-time threat detection and response capabilities. Unlike conventional models, which require costly retraining to update knowledge, the proposed framework integrates incremental learning algorithms, reducing the need for frequent retraining. Adversarial training is used to fortify the IDS against poisoned data. By using fewer features and incorporating statistical properties, the system can efficiently detect potential threats. Extensive evaluations using the NSL-KDD dataset demonstrate that the proposed approach provides better accuracy of 82.33% for multiclass classification of various network attacks while resisting dataset poisoning. This research highlights the potential of adversarial-trained, dynamic neural networks for building resilient IDS solutions.


14 August 2026
Verified
Preprint
Open access
PDF file
- Neha- Neha
- Neha
Bioinformatics workflows routinely combine small relational/metadatafiles (sample manifests, phenotype tables, GWAS summary statistics)with large, dense numeric arrays (expression matrices, genotypearrays). Existing practice typically forces a choice between arelational database, which handles metadata well but scales poorlyfor dense arrays, or a chunked array format, which handles heavynumeric data well but has no native relational query layer formetadata. DZH-Protocol defines a reproducible method for routing eachinput file to whichever storage layer suits its shape: lightweighttabular data is ingested directly into DuckDB, while heavy numericarrays are encoded into chunked, compressed Zarr stores and linkedback into DuckDB through a lightweight pointer table. This keepsrelational queries fast and array access lazy and chunk-wise, withoutrequiring either format to do the other's job. The protocol specifiesan explicit, configurable classification rule, a defined chunking andcompression convention, a pointer-table schema, and checksum-basedvalidation, so that independently written implementations can bechecked for conformance against the same rules. A reference Pythonimplementation (DZH protocol) is available on GitHub(https://github.com/NehaAr/DZH_protocol) and has beenvalidated against a real 21-million-row summarystatistics dataset.