Smart Meters, Electricity Usage, and Privacy

My current apartment building has smart meters that measure and report each unit’s electricity usage. While you can get real time high-resolution usage data if you’re willing to stand in front of the meter, the utility company also provides daily usage data through their website.

I got tired of logging in every day, and the visualizations they provide are a little clunky, so I wrote my own Python script to automatically log in, download the data, and graph it using Plotly. The code is available on GitHub, though it’s specific to my utility company (United Illuminating). doesn’t allow arbitrary embeds, so below is a static image of the Plotly output. Here’s a live version that’s updated every 24 hours. It has mean lines and trend lines (simple linear regression) by month. I also built an Android app for my phone that fetches a vertical layout of this graph in a WebView.


I went to a talk this past week by Elizabeth Wilson, an associate professor of Energy and Environment Policy at the University of Minnesota. It was a really interesting talk that covered a lot of different topics, including some of the privacy implications (and perceptions) of smart meters.

One of her slides contained the following image, which documents electricity demand in a British household:

Screen Shot 2014-11-23 at 11.20.29 AM

I found this fascinating; just a few weeks earlier, I had been discussing with a friend visiting from out-of-town how much usage one could infer from reasonable resolution electricity usage data. I was skeptical about how much we could really learn.

  • My day-interval data can probably give you an idea of whether I’m home or not, although it might not be any better than just distinguishing between weekdays and weekends. You also might guess when I turned on the heat for the winter.
  • This project uses the rtl-sdr dongle to decode the messages sent on the 900 MHz ISM band by a particular brand of smart meter, and receives data every few minutes. I would ultimately like to do this for my brand of meter, which, AFAICT, is broadcasting on the 2.4 GHz ISM band every two minutes or so.
  • The graph above uses minute-interval data, derived not from a smart meter, but from a data logger installed at the customer’s home.
  • Proof-of-concept research shows that with two-second interval data, it’s possible to determine what television programs are being viewed (assuming you generate fingerprints of viewable content ahead of time).

The graph above has found its way into presentations, papers, and even anti-smart-meter websites, often as an example of how smart meters can reduce privacy and reveal significant details about personal habits.

Curious, I tracked down the original source of the graph. It comes from a paper published in 1999, which tracked the electricity usage habits of 30 British households. But the methodology section contains some critical details that are always omitted when this graph is bandied about:

Questionnaires were designed for completion at the time the logger was installed in order to identify the primary characteristics of the household and in particular which types of appliance were fitted. Some householders also completed tick-box ‘appliance utilisation’ diaries for one weekday and one weekend day during the logging period; these served to improve our ability to ‘read’ the demand profiles… User behaviour is generally deducible from the database of monitored electricity demand profiles and the knowledge of a surveyed household’s appliance stock… the most apparent classification was by periods of occupation.

In other words, the researchers not only had access to minute-interval data about demand, they also knew what appliances people owned, and in some cases knew when people were using these appliances—all because people volunteered this information. The researchers also logged power factor data, and the internal and external temperatures at the home being monitored.

Note the numerous errors on this Stop Smart Meters website, which includes the graph:

  • They call it a “hypothetical graph” (it’s a real graph, from a real research paper)
  • They imply it was generated by a current-era smart meter (it was generated by a data logger installed at the consumer’s home, which consisted of a 90’s era Acorn PC, and which measured several other variables)
  • They claim that the appliance labels in the graph are based on “signatures” derived from variation in power factors (in fact, the original paper notes that power factors are almost always close to 1, except for the refrigerator and washing machine in spin mode; and they omit that these households answered questionnaires and completed diaries about appliance usage)

Without knowing details about appliance ownership, it would be far harder to determine what is going on in household usage. For example, I suspect there is little way to distinguish between an electric kettle and a toaster based solely on usage, other than to guess that one might make tea before one makes toast.

Certainly, some things will always be obvious (you have a refrigerator, you cook using an oven); but others are less transparent. I have an unusual appliance that pulls 1.5 kW for about 90 seconds a few times a day (it’s not a kettle), and I doubt anyone at the utility company will ever guess what it is.

Consumers should realize that smart meters can benefit them. Apart from lowering costs for the utility company and making them more aware about outages, knowing how much you’re using can help you save money. Where I live, electricity costs about $0.21 per kWh, so savings really add up. Utility companies can help by making sure consumers have access to their data. I wish my company offered higher resolution data, but even the day-interval data has been very informative.


About Gabriel

Ph.D. in political science. Postdoc and resident fellow at Yale Law School's Information Society Project. Tech geek. Mechanically inclined. I study the politics of intellectual property.
This entry was posted in General and tagged , , , , . Bookmark the permalink.

11 Responses to Smart Meters, Electricity Usage, and Privacy

  1. Bob Bichen says:

    Consumers should also realize that smart meters can harm them. Thousands of peer-reviewed scientific studies (mostly NOT funded by industry purveyors of microwave emitting technologies) show dozens of harmful effects to the human body. Low level chronic microwave radiation is the new asbestos or tobacco of the 21st century, complete with the well funded lobby groups to bury the real science showing its harm.

    • Gabriel says:

      Without addressing the issues of actual effects, most consumers will have far more exposure to EM radiation from their cell phones and various WiFi devices around the house than from their smart meter.

      • Bob Bichen says:

        That’s not necessarily true and represents an assumption on your part. Dr. Daniel Hirsch calculated smart meter exposure to be over 100 times that of typical cell phone use. More importantly, cell phone and wifi are optional, not mandated by law. Smart meter exposure is whole body, 24/365 and impossible to turn off by the user. Some people feel effects from low-level microwave exposure immediately, others don’t, but blood analysis shows unequivocally that everybody suffers effects from it due to its influence on biochemical processes that rely on specific calcium ion exchanges through cell membranes.

      • Gabriel says:

        It looks like you’re referring to this document. I don’t want to go down this rabbit hole very far, but a quick glance through it reveals several things:

        • It’s not a publication in a peer-reviewed journal, so its claims have not been submitted to the scrutiny of other experts in the field.
        • A duty cycle of 100% is probably an overestimate for most widely deployed smart meters.
        • Measurements are taken from a smart meter at 3 feet. In the vast majority of real-world situations, a smart meter is going to be much further than three feet away from you (and there will likely be path attenuation from materials, like walls, in between).

        I appreciate your readership, but the focus of my post is about the value of the data smart meters provide, and how privacy concerns are probably being overstated, so I’m going to try to moderate comments to keep the focus on those issues.

  2. For a more expanded on how smart meters invade privacy, I invite you to read some of the articles at my website. My most recent deals with reviewing how “Smart Meter Privacy Invasions Are Not Justified in a Democratic Society” at: Also, “Updated Report on How Smart Meters Invade Privacy” at: There are many other articles as well. For those people who care little about privacy in their lives, the articles will be of little interest. On the other hand, for others, they may be enlightening.

    • Gabriel says:

      Thanks for your comment! At some point in the future, I may write a longer post about smart meters and privacy. I don’t deny that there are important privacy implications of smart meters. However, I have read multiple studies (including some that you linked to in your writings), and none of these studies actually demonstrate real-world effects. There are a few reasons why:

      1. None of the studies actually use smart meter data. Every study I have seen uses a data logger installed by the researchers. This is typically because the researchers want highly granular data, which leads to the second point:
      2. The studies use a far higher-level of granularity than smart meters currently provide. Many recent studies use one-second granularity. This is far more granular than deployed infrastructure provides, which may report at 15 minute or one hour intervals. As one of the studies notes, granularity “directly correlates” with the amount of information one can infer.
      3. Most of the studies employ additional information beyond electricity consumption to reach their conclusions. This is what I noted in my post, and it is true of other studies, which have monitored consumers keep usage logs, or record video of their activities. This data is then used as a training set in a supervised learning model. Alternatively, researchers assume access to other, unrelated information, such as workplace locations, TV program signatures, and lists of appliances that households own.

      Notably, none of the studies I have read suggest we do away with smart meters. They all recognize the value of smart meters and an advanced metering infrastructure in reducing overall and peak electricity demand. E.g., from one study: “depending on the extent of the distribution of AMI, the potential savings in energy in the United States during the peak summer period for electrical demand ranges from 4% – 20% of total load. The subsequent positive impact on the U.S. need for foreign oil and related resources would be difficult to overstate.”

      From my perspective, it is odd how much effort some people expend on opposing smart meters, when credit card activity and grocery store loyalty cards reveal FAR more detail about people’s lives, behavior, habits, diet, health, etc. One of the studies mentions that the researchers could guess when a consumer ate outside the home rather than cooked at home (since cooking at home generates spikes in demand in the evening hours). Yet credit card activity currently reveals not only this, but also the location and time of the meal, the amount spent, and a good guess at the amount consumed.

      • Many of studies use a data logger in place of smart meter as they develop proof of concept and also to collect what sometimes is called “ground truth” in order to validate the algorithms which can be later used for smart meter applications. This necessitates collecting more information about the home than otherwise could be obtained through the smart meter itself. Later, once algorithms have been validated and libraries of appliance signatures compiled, specific appliance use and identification can be made with confidence without collecting the additional ground truth information.

        Granularity is certainly key for disaggregation algorithms. Hourly interval data collection would generally allow occupancy determination and general categories of use. There is a company in Canada that advertises: “Ecotagious Inc.,… has spent years developing sophisticated algorithms which help identify how much electricity major appliances are using based only on smart meter data without the need for any additional hardware. … using proprietary smart meter data analytics that disaggregate HOURLY smart meter data into major appliance loads.” They claim to break down hourly data into categories of electric space heating, air conditioning, base load, occupancy load, and other appliances. This capability is enough for me to say “no” to such technology invading the home space. There are other companies developing more sophisticated algorithms and I could provide more information if desired.

        Where I live, the smart meters collect both real and reactive power data parameters at 15 minute intervals. The reactive power component aids in determining major appliance identification.

        Another called Bidgely issued a “white paper” document that states “every smart meter installed in California (also applicable to most smart meters across the globe) has a ZigBee radio that when turned on by the utility can communicate directly with the home. Numerous In Home Display (IHD) pilots have been performed by utilities across the globe using this interface. … The California IOUs turned on the ZigBee radios of the smart meters in 2013, thus allowing consumers to directly access their real-time energy consumption data. The data streams every few seconds (vary from 7.5 to 30 seconds) from their smart meter.” Link at:; see page 4.

        Thus, although activation of the (secondary) HAN transmitter within the smart meter is generally advertised as optional by the utility at the request of the customer, there are serious concerns that the HAN transmitter might somehow become activated inadvertently or without the customer’s consent, thereby transmitting real-time and extremely detailed granular appliance-related data to anyone able to access and interpret the data transmission.

        As to being odd for people to oppose smart meters as compared to other activities, it is a matter of perception of risk relative to perceived benefits of the technology. Also, there are those who possibly do not use credit cards as much you think. I sometimes use “cash only” where I don’t want a certain transaction recorded. I don’t use a cable box because I don’t want the cable company to know what TV show I am watching. That is my choice and it should be the same with smart meters; I shouldn’t have to go without power as my only option when the utility does not need the added granular data to bill me for monthly usage at a fixed rate.

        Studies I have seen purporting savings to the consumer skew the results by including other initiatives as a part of the study; like now that you have a smart meter, install new insulation at a discount and see how much energy you save. The smart meter was not necessary to achieve the savings that is later reported.

        Earlier this year there was a news report that in Texas where millions of smart meters are deployed, less than 1% of people have ever logged in one time to check their online granular data utility records. So how useful is that?

        What is necessary is consumer choice; some prefer to keep old analog meter but other options could alleviate privacy concerns such as smart meter being programmed to only collect data necessary for billing purposes. This is actually the approach in the Netherlands and Great Britain.

      • Gabriel says:

        I shouldn’t have to go without power as my only option when the utility does not need the added granular data to bill me for monthly usage at a fixed rate.

        While smart meters do reduce billing costs (by not requiring meter readers to visit each meter individually), they also facilitate variable rates based on the time of day, which is an important component of load shifting, thereby reducing peak demand. This in turn reduces the need to turn on dirty and expensive peaking plants, resulting in less pollution and less expensive electricity. They also can help identify outages far more accurately, resulting in quicker restoration of service. I.e., there are more reasons than simply billing to use smart meters, and many of these reasons directly benefit the consumer.

        Earlier this year there was a news report that in Texas where millions of smart meters are deployed, less than 1% of people have ever logged in one time to check their online granular data utility records. So how useful is that?

        First, even if the number was 50%, I doubt this would change your opposition. Second, it sounds like the utility in question needs to do a better job of advertising the possibility of viewing usage, or perhaps making it easier to do so. Texan’s cost per kWh is about the U.S. average, or slightly below. Where I live, I pay twice the national average, so conservation and reduction in usage is financially important. Having daily reports helps me save electricity, thereby saving money. When I take some conservation action, I can see the results a few days later, and have some confidence that the change is related to my action.

  3. Fabrio says:

    This is a few years old but it shows how a hacker mapped power consumption of TV to particular movie…. using man in the middle attack on smart meter ….

    • Gabriel says:

      Yes, this is the research I linked to in my post. They needed two-second interval data, and had to generate fingerprints of the video ahead of time.

  4. Mark Miller says:

    Your post was referenced over at Energy Matters recently.

    I am glad I followed the link to your post as your discussion on the high degree of granularity (sampling rate) that was used to generate the detailed demand curves was very informative.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s