Cybersecurity service · Founded 2024

Intrusion detection that's built for SMEs, not just big business.

SecureSME is an affordable, tailored intrusion detection service for small and medium enterprises. An adaptive deep learning model watches your network traffic and flags attacks as they happen.

See how it works View the results
96%detection accuracy
3deep learning models
Adaptiveto evolving threats
The threat

Attacks don't check company size first.

Cyber attacks hit businesses of every size — and smaller organizations usually have the least protection and the fewest security staff.

50%

of businesses reported a cyber security breach or attack in the last 12 months.

70%

of medium-sized businesses reported a breach or attack in the same period.

84%

of businesses that identified an attack were hit by phishing — the most common type.

Source: UK Government, Cyber Security Breaches Survey 2024

The protection gap

Enterprise security, without the enterprise price tag.

Typical enterprise tools

  • Priced and sized for large organizations
  • Need dedicated security staff to run
  • Complex setup and technical alerts

SecureSME

  • Affordable for small and medium budgets
  • Delivered as a service — no in-house security team needed
  • Tailored to each organization's network
Why adaptive

Threats evolve. Static rules don't.

Earlier intrusion detection methods struggled to keep up with constantly changing attacks. SecureSME's deep learning model learns patterns in network traffic, so it can recognize threats that fixed rules miss — and is designed to address future threats too.

How it works

From traffic to alert, automatically.

  1. 01 // MONITOR

    Watch the network

    Network traffic flowing in and out of the organization is monitored continuously.

  2. 02 // ANALYSE

    Deep learning analysis

    CNN, ANN and MLP models examine traffic patterns for signs of intrusion.

  3. 03 // CLASSIFY

    Normal or attack?

    Each traffic scenario is classified, separating everyday activity from threats.

  4. 04 // RESPOND

    Alert and report

    Suspected intrusions are flagged clearly so the business can act fast.

The technology

Three deep learning models, working together.

The adaptive framework combines three neural network types to improve both detection accuracy and scalability.

CNN

Convolutional Neural Network

Finds local patterns and structures in network traffic features that point to an attack.

ANN

Artificial Neural Network

Learns complex, non-linear relationships between traffic characteristics and threats.

MLP

Multi-Layer Perceptron

Stacks layers of neurons to classify traffic scenarios as normal or malicious.

Results

Accurate detection across real traffic scenarios.

CLASSIFICATION ACCURACY 96%

in classifying varied network traffic scenarios with the adaptive deep learning model.

Higher detection accuracy

Deep learning improves on the accuracy of earlier intrusion detection methods.

Scalable

The framework is designed to grow with the network it protects.

Ready for future threats

An adaptive design built to keep pace as attacks change.

Source: Fenjan et al., Proceedings of the Cognitive Models and Artificial Intelligence Conference, ACM (2024), pp. 279–284. doi:10.1145/3660853.3660928

The service

Security as a service, sized for your business.

Affordable

Priced for the budgets of small and medium organizations.

Tailored

Configured around each organization's own network and risks.

Always watching

Continuous monitoring, so threats are spotted when they happen.

Clear alerts

Plain-language warnings that non-specialists can understand and act on.

Part of a layered defence

SecureSME adds the layer most SMEs are missing.

Intrusion detection works best alongside the basics — it doesn't replace them.

FirewallBlocks unwanted connections at the edge of the network.
AntivirusProtects individual computers from known malware.
SecureSME IDSWatches traffic for suspicious behaviour that slips past the other layers.
Backups & trainingRecovery plans and staff awareness against phishing.
The team

Cybersecurity engineers on your side.

Faculty of Engineering Technologies

Cybersecurity Engineering Technologies

SecureSME was founded in 2024 by a research team specializing in cybersecurity engineering, combining network security with deep learning to protect organizations that big security vendors overlook.

Network securityIntrusion detectionDeep learningThreat analysis

Frequently asked questions

What is an intrusion detection system?

An IDS monitors network traffic and alerts you when it spots activity that looks like an attack or unauthorized access.

How accurate is SecureSME?

The adaptive deep learning model behind SecureSME classified varied network traffic scenarios with 96% accuracy in published research.

Which AI models does it use?

A framework combining Convolutional Neural Networks (CNN), Artificial Neural Networks (ANN) and Multi-Layer Perceptrons (MLP).

Do we need our own security team?

No. SecureSME is delivered as a service designed for small and medium organizations without dedicated security staff.

Does it replace our firewall or antivirus?

No. It adds a detection layer on top of existing protections such as firewalls, antivirus, backups and staff training.

The research

Built on open, peer-reviewed research.

Read the open-access ACM study behind SecureSME's adaptive intrusion detection.

Read the paper
Open access

Fenjan, A., Almashhadany, M. T. M., Ahmed, S. R., Fadel, H. A., Sekhar, R., Shah, P., & Veena, B. S. (2024). Adaptive intrusion detection system using deep learning for network security. In Proceedings of the Cognitive Models and Artificial Intelligence Conference (pp. 279–284). ACM.

https://doi.org/10.1145/3660853.3660928