By Steve Hickling, Spirent
GNSS have been with us for more than 30 years, giving rise to a wealth of positioning and navigation technologies for military, civilian, and consumer use. Today, we’re entering a new era of experimentation and innovation in satellite and hybrid positioning. In turn, this drives new test challenges and introduces an ever wider group of engineers to the art and science of GNSS test.
Where Is the Testing Panacea? I am sometimes asked, “What is the best way of testing a GPS receiver?” — as if there existed some laboratory panacea to all GNSS test and characterization woes. Well, there is a saying, “There are horses for courses,” meaning what performs well in one situation may not deliver in another, and nowhere is this more true than in the field of GNSS test. Not only is there a wide range of different test equipment available, but there are no universally applicable test objectives, test methods, or parameter definitions, in exactly the same way as there is not one universally applicable GNSS receiver. Just as the rapid time-to-first-fix of an automotive receiver may be less relevant in a maritime environment, so different test approaches have their place.
A Systematic Approach. If there is one thing, it is this: be systematic in your test design. Consider the purpose of the test, the test conditions, and the measurements you plan to take, and be wary of generic tests that may not achieve your objectives.
Equipment. A wide range of GNSS testing equipment is available, ranging from basic single-constellation RF simulators to highly configurable, multi-GNSS constellation simulators. Single-channel, multi-channel, and record and playback systems all have their place, and to get the best results in the fastest time, it’s essential to choose the kit that’s right for the kind of testing you need to do.
Vulnerability, Fidelity, Integrity, and Time Travel. More and more, receivers need to be tested for their vulnerability to interference, jamming, and spoofing. As GNSS-derived position and time become more ubiquitous, so the motivation for confounding the system grows. This has a double impact on test.
First, performance requirements around vulnerability may be introduced, with tests to match. Second, and perhaps less obvious, is the way in which this concern is reflected in the receiver’s design and potential rejection of the laboratory test signal. Yes, I mean receivers getting more fussy about the signals they lock onto. Anyone who has tested a receiver with an out-of-date recording or simulation scenario will have experienced a receiver refusing to track a satellite showing a time and date prior to its firmware release date. The receiver, discounting time travel, knows there has to be something wrong with a satellite showing a date before it was born. With the risk of spoofing, receivers will only get more picky and likely to reject poorly simulated signals. To avoid such effects, it is important to have very high integrity and fidelity in any simulator system. Getting these details right is not esoteric, but is essential to allow the proper attribution of any problems observed and if test results are to have any meaning.
Conclusion. Be systematic in your approach to test; beware the universal and generic; “good enough,” it rarely is.
Steve Hickling is lead product manager for Spirent’s GNSS simulator business and is based at the factory in Paignton, England. Previously he held a variety of marketing, technical, and management roles in the telecoms and optical components industries. He holds a bachelor of science degree in physics and electronic engineering from Birmingham University, an MBA from Open University Business School, and a post-graduate diploma from the Chartered Institute of Marketing.