As technology advances, new parameters and measures are necessary while some older ones fade in importance.
Back in the day, there were sophisticated, detailed papers created at telecom companies and labs—such as the venerable Bell Telephone Laboratories (Bell Labs)—which analyzed the many voice-phone traffic patterns and situations, and resources needed (local loops, line interface units, routing switches) in terms of erlangs of traffic, how long a user would have to wait to get service, and many other performance metrics. Much of this analysis involved probability-based assessments and the law of large numbers. It’s probably hard for us to fully appreciate how important the erlang was as a metric in this system analysis.
Before you dismiss the erlang as an anachronistic holdover from those ancient days of wired telephony, many of the concepts it inspired to form the basis for much of today’s communications and networking environment. Seminal works such as Harry Nyquist’s “Certain Topics in Telegraph Transmission Theory” (sampling theory, 1928) and Claude Shannon’s “A Mathematical Theory of Communication” (channel capacity, 1948) were closely aligned with issues related to both wire-based telephony as well as wireless links.
Now, with the transition from circuit-switched to data-switched connectivity, the erlang has fallen away and is largely unknown. In its place, we have much more meaningful metrics for data rate and capacity, such as megabits/sec (Mbps); miles (or km)/minute for EV recharging effectiveness, bits per second per hertz for spectral efficiency, and mA/MHz to characterize the processing power and computational efficiency of a CPU, DSP, or co-processor (although I’m seeing much less of that one lately).
Some metrics and figures of merit (FoM) are very specific, even though they are small-scale. For example, an op amp’s input bias current, IB, is an important FoM in some applications and not at all in others.
The need for new measures continues, of course. I came across a figure of merit that was new to me in an article in Microwave Journal, “mmWave Will Be the Critical 5G Link.”. This FoM was Gbps/km2/MHz, or gigabits per second, per square kilometer, per MHz of spectrum (shortened to GkM), and it measures the level of wireless traffic density. It’s not clear to me if this is a widely used system-loading metric, or one promoted by the article’s author at Mobile Experts (Campbell, CA), but it still seems quite useful.
How so? GkM indicates how intensely the available spectrum is used in a given geographical area. This area can be highly localized, such as a stadium or arena, or a neighborhood, or a city; (Figure 1) shows one data set.
GkM helps decide if more microcells are needed to adequately that area and of what type to ensure an acceptable level of user access and connectivity. By looking at GkM, you have one quantitative measure that can help determine the service and resource situation – somewhat analogous to what erlangs helped do in the circuit-switched world.
Undoubtedly, there are other useful and necessary factors for determining wireless connectivity performance; some may have industry-wide and accepted definitions, while others may be more “proprietary” and promoted by one or a few vendors or market analysts. In my spare daydreaming time, I’ll be trying to think of metrics and FoMs that used to be cited regularly but now rarely are, if at all, and some newer ones that the technical community is now using.
Related EE World Content
- High-speed data challenges physical-layer optics
- What metrics can bioelectrical impedance analysis provide?
- What metrics can bioelectrical impedance analysis provide? Pt 2.
- First, go for accuracy, then precision
- ADCs: sufficient sampling at Nyquist’s rate
References
- NIST, “For All Times, For All Peoples: How Replacing the Kilogram Empowers Industry”
- NIST, “A Turning Point for Humanity: Redefining the World’s Measurement System”
- NIST, “Toward the SI System Based on Fundamental Constants: Weighing the Electron”
- NIST, “Universe’s Constants Now Known with Sufficient Certainty to Completely Redefine the International System of Units”
- Erlang Ecosystem Foundation, “Erlang programming language”
- Harry Nyquist, “Certain Topics in Telegraph Transmission Theory”
- Claude Shannon, “A Mathematical Theory of Communication”
- Microwave Journal, “mmWave Will Be the Critical 5G Link”
- Mobile Experts, “5G – Is it Economically Viable?”
- Mobile Experts, “When does Mobile Traffic Density force the operator to use small cells?
Leave a Reply
You must be logged in to post a comment.