Editor’s Note: Welcome to our weekly Reader Forum section. In an attempt to broaden our interaction with our readers we have created this forum for those with something meaningful to say to the wireless industry. We want to keep this as open as possible, but we maintain some editorial control to keep it free of commercials or attacks. Please send along submissions for this section to our editors at: [email protected].
Security continues to be a hot topic in all areas of technology, including machine-to-machine applications. While most analysts agree that the current security risk is relatively low, the consensus is that as the M2M space continues to grow, with projections as high as 25 billion connected devices by 2015 and 50 billion by 2020, security concerns will become more and more relevant.
A September 2012 GigaOM Pro report on the subject stated, “As the volume of
M2M devices and associated data increases, so too will the probability of hackers and malware writers targeting these systems to exploit networks, steal data, hijack systems, and compromise workflows.”
The report details a number of scenarios researchers are studying that represent potential avenues of attack, from wirelessly transmitted computer viruses infecting onboard systems of vehicles, to “man-in-the-middle” attacks between wireless endpoints to intercept and manipulate data. The question is what should enterprises using M2M and original equipment manufacturers developing M2M solutions be doing right now to protect their applications?
One might conclude “everything they can,” but the goal for an M2M application shouldn’t necessarily be to provide the “most” security, but rather provide the right level of security for each specific device and application. In fact, the question “How secure should this application be?” is more complicated than it may appear, because the phrase “M2M application” encompasses a full ecosystem that extends beyond the device itself. It includes the cellular network, the machine subscription identification module, the M2M cloud management platform and the back-end enterprise application, all of which have unique security capabilities and considerations.
Even security on the M2M device itself is not straightforward. Unlike a PC or mobile phone, most M2M devices don’t have operating systems and the processing power to run advanced antivirus or anti-malware software. Indeed, even full-scale encryption mechanisms can overwhelm the microprocessors and available bandwidth in many deployed M2M devices.
Simply put, building every M2M device – every connected sensor in every car, truck, toll station, appliance, industrial equipment, utility meter, etc. – with a full operating system and high-powered chipset isn’t justifiable from a business perspective. Rather, the smartest M2M security is security tailored for each specific application. As a starting point, one needs to understand three essential truths about security:
1. Some level of security is always necessary. No enterprise would deploy a system that afforded free reign for attackers, and no legitimate M2M vendor in the marketplace today would build one.
2. No system can ever be 100% secure. Security is less a goal to be achieved than a calculation of the time and resources necessary to crack a system. No hacker would spend more time and resources breaking a system’s security than the value of what that security is protecting. Put another way, it doesn’t make sense to install a $1 million security system to protect a $100,000 house. A banking application therefore requires much more security than a simple power monitoring application for a consumer home. Additionally, no system will be secure forever. Security must align with the expected lifetime of the solution. It should be secure enough for the threats faced today, but upgradable to defend against future threats.
3. Security always means constraints. The more security a system employs, the higher the costs to build and operate it, the more usability will suffer and the more restricted it will be in the features it can support. Ultimately, too much security can be just as problematic as too little. Consider a real-world example such as two-factor authentication (i.e., requiring a user to enter a password plus a one-time code sent via text message to access an application). It may make sense to use this level of security for applications that make credit card purchases online. It does not make sense to require it every time a user opens a connected e-reader, which would, in fact, marginally increase security at best, but significantly diminish the user experience and increase costs.
Keeping these three truths in mind, what does the “right” level of security mean? It means:
–The devices, applications, and data are secure enough for all use cases they will support.
–The application can offer powerful features.
–The application is easy to develop, deploy and maintain, in a cost-effective way.
Enterprises and OEMs use a variety of mechanisms and techniques to address security threats in each segment of the M2M chain. Ultimately, however, all of these techniques revolve around four key areas of M2M system security: trust, upgradability, robustness, and encryption. To summarize:
–Trust: In an M2M application it’s imperative to verify that commands or instructions coming in to a device or server are legitimate and coming from a verified source.
–Upgradability: To maintain security in a constantly evolving environment, enterprises and OEMs must be able to update M2M devices and applications quickly and remotely, across thousands or millions of deployed devices.
–Robustness: A robust M2M application should be built using an M2M-specific software stack that includes design elements and fault tolerance mechanisms that allow M2M applications to remain available even when operating at massive scale.
–Encryption: An M2M application must protect the transmission of private and confidential data by utilizing data encryption and secure transmission technologies across multiple segments of the application – between deployed devices, the M2M cloud management platform and the enterprise application.
Securing the cloud data center
Just as important as securing communications between M2M devices and servers, enterprises must assure that the M2M cloud management platform they use employs strong security measures. This is especially important when relying on a third-party cloud management provider.
A secure M2M cloud offering should provide:
1. Secure shared data warehouse that assures that deployed devices can only register to the appropriate server.
2. Flexible data retention policies that can be defined and implemented based on the enterprise’s unique requirements.
3. Robust disaster recovery features to protect the M2M application, including:
–Highly resilient infrastructure that assures 99.982% uptime (i.e., tier-three data center).
–Daily backup of all data to a remote site.
–Nonstop monitoring and alerting.
–Fast recovery times (24 hours or less) in the event of a disaster.
–Ongoing testing of disaster recovery attributes.
Enterprises should also look for M2M cloud partners that are certified for compliance with industry standards for data center security (i.e., SAS 70 Type II, ISO 27001) and that are a member of the Cloud Security Alliance.
Conclusion
Today, it’s clear that companies looking to deploy M2M solutions need to address both the current and anticipated security concerns to ensure their applications are protected from outside threats across the full M2M ecosystem, and that their own customers’ information and assets are kept safe. Fortunately, any enterprise or OEM can achieve the “right” level of security for its M2M application, both today and in the future. The key is to work with vendors that understand not just security, but have deep expertise and extensive field experience in M2M itself.
