Sensors are inherently of their context: the physical context they sense and the human one they often infer, but also of the corporations that manufacture them, and the organizations that install them and base decisions on the supposedly objective data that they create.
They’re everywhere. We attach them to our wrists, embed them into our medical devices, and mount them onto the lampposts that dot every block of our city. Some sensing technologies capture our imagination and attract our constant attention. Yet many go unnoticed, their insides packed with unknowable electronic components, ceaselessly counting, measuring, and transmitting. For what purpose, or to whose gain, is often unclear.
These unseen sensors are the instantiations of large invisible systems that have a very real impact on the lived urban experience, and we want to examine them, their contexts, and their consequences from the vantage point of a single human being at street level in New York City.
They are invisible by default. Housed in anonymous plastic or metal boxes, these sensors rarely give away even a hint as to their purpose, intention, or ownership. Only sometimes adorned with a manufacturer’s label or an owner’s logo, there is seldom any information to explain what these barnacles of our urban landscape are or what they are doing.
This panoply of boxes falls under the control of countless public bodies and private owners. In New York these include the city and state Departments of Transportation, The New York City Police Department, and federal entities such as the Department of Homeland Security. Retail outlets, businesses, and private residences have their own suites of sensors and security systems.
For all their ubiquity, the aesthetics of these public objects are generally unconsidered, and their function and provenance are entirely inscrutable. It’s one thing to be able to recognize a CCTV camera. It is generally impossible, however, to know the fate of its data. Perhaps it’s streaming live video to a far-off police control room, being scanned in real time for flagged license plate numbers. Or it may be recording footage that will later be viewed by a private security contractor working the night shift. Some may never been seen by a human, instead archived to a remote server under an entirely different jurisdiction – and kept for 24 hours, or for 5 years, or indefinitely. Even if unexamined today, it could be scanned with facial recognition algorithms or mined for patterns years later.
The Domain Awareness System, for example, was created by the NYPD together with Microsoft in the wake of the 9/11 attacks, funded in part by the Department of Homeland Security. Its stated goal: to monitor lower Manhattan by making sense of the many streams of signals flowing from a variety of devices, including thousands of radiation sensors, publically and privately owned surveillance cameras, and 911 calls and reports from officers on the street.
The system automatically runs the video feeds captured by these cameras through an automatic license-plate reader and cross-references them against a database of missing or flagged vehicles throughout Manhattan. Facial recognition algorithms could be run on that same video stream, but (current) policy is not to implement or use that capability.While the actual sensors may remain unchanged when policies and technologies shift, that which is qualitatively “sensed” changes dramatically as the systems analyzing the data are installed, integrated, and upgraded.
But for all the futuristic optimism and deep-seated anxiety these technologies can arouse, most of them perform simple and mundane tasks. An external thermometer helps a building regulate its heat. A street-embedded parking sensor detects if a car is present, and signals the spot’s availability. A traffic light stays green a fraction longer as a bus approaches and emits an infrared beam to announce its presence.
This is the reality of the 21st century city. Our future won’t arrive in a sweeping, clearly defined transformation represented by an inevitably shiny rendering or seductively simple diagram, even if those are the futures that make headlines. Rather these changes happen incrementally, infiltrating our everyday bit by bit as technology falls in price and grows in utility, availability, and reliability, and becomes socially and politically palatable.
The sensing devices multiplying around our world are the instantiation of ethical, aesthetic, and political choices made by a myriad of organizations and individuals over the course of decades. Each has their own incentives to follow and problems to solve – including the prosaic and profound, the widespread and those applicable only to a select few.
Tomorrow’s networked, instrumented, sensing city will still function in messy, complicated, and contested ways, as cities always have. An ever-changing palimpsest of infrastructures and technologies will spur cities onward and, as ever, inhabiting, interpreting, and interpolating these systems will be the most wonderfully unpredictable and most significant variable of them all: people.
As you examine the sampling of sensors from just a small portion of New York City below, we challenge you to consider them carefully. What decisions led to these additions to the built environment, and what decisions are based on the data they create? Should private companies own the data that is created in this most public of contexts? In this realm of multi-billion dollar technologies that can sometimes seem like science fiction, where is the line between what is real, and hype? And why should these newer additions to the urban fabric be any less carefully thought through than the steel, cast iron, or brick structures onto which they are grafted?
Once you start noticing them, you’ll see many more. Keep looking.
Remote Traffic Microwave Radar Sensor
This remote traffic microwave radar sensor attached to a roadside traffic light pole at a major intersection is an RTMS K4 manufactured by Image Sensing Systems of St. Paul, Minnesota. Part of the “Midtown in Motion” program and administered by the New York City Department of Transportation (NYCDOT), and the New York City of Information Technology & Telecommunications (NYCDOITT), this device registers the presence of passing vehicles and bicycles in each lane, allowing the city’s department of transportation to adjust traffic signals in response to accidents, major events, and unexpected volumes of traffic.
RFID E-ZPass Reader
This device on a lamppost at an intersection is an “Encompass 4” RFID reader manufactured by TransCore of Pennsylvania. It is used to sense E-ZPass transponders in passing vehicles as part of the City and State Departments of Transportation and the NYCDOITT “Midtown in Motion” program. Although E-ZPass is primarily used for toll-collection across the Northeast, these sensors track traffic flow in midtown Manhattan, allowing traffic signals to respond to changing conditions.
Horizontally Polarized Panel Antenna
Used together with the Encompass readers shown above, this horizontally polarized panel antenna (model AA3110-101) is also manufactured by Transcore and forms part of the “Midtown in Motion” system. Mounted directly over an intersection, it broadcasts and receives radio frequency signals in the 902 to 928 MHz band, picking up the RFID-based E-ZPass transponders in vehicles passing in lanes below.
NYC Street Light Status
The red lights on the top of most New York City streetlights are repair status indicators. A sensor in the lamp detects changes in the electric current and, if an anomaly or problem is detected, the red light flashes. In some cases, a signal is sent to the NYCDOT to notify them that a repair is necessary, and the flashing light serves as an indicator to repair crews.
Water Meter Transmission Unit
These grey boxes are attached to the exterior of over 800,000 buildings across New York City. They are Meter Transmission Units, manufactured by Aclara (owned by ESCO technologies of St Louis, Missouri) and part of a 2008 New York City Department of Environmental Protection (DEP) program titled “Citywide Advanced Metering Infrastructure Program.” They transmit data from meter sensors inside the building to nearby Network Data Collectors, which in turn send it to the DEP for more accurate billing compared to estimating bills based on irregular manual meter readings, and near-real-time leak alerts for building owners.
Private Security Camera (Hikvision)
This weatherproof CCTV camera on a private building is a DS-2CC1173N-VFIR, manufactured and sold by Hikvision of Hangzhou, China. This recent model incorporates a 1/3” Sony CCD sensor, and a 2.8 - 12mm lens with an aperture of f/1.4 and a viewing angle of 92° - 27.2°. Visible around the lens on the front of the device are infrared LEDs for improved vision in low light. The embedded software has options including night vision, masking areas as ‘privacy zones,’ and motion detection. As it is part of a private security system, there’s no way to know whether it is being monitored live or simply stored for future reference.
Private Security Camera (Panasonic)
This CCTV camera is a WV-CW964 model from Panasonic, a technology giant from Osaka Prefecture, Japan. This camera cost upwards of $1,000 per unit, and is sold with the tagline that it “Sees Reality and Thinks Actively”. It incorporates a video feed, an ambient light sensor, and motion tracking capabilities, with a motor capable of panning 360 degrees horizontally and 190 vertically in less than one second, automatically following the largest movement detected in the image. The brochure for this model describes these as “Handy Features for More Efficient Surveillance.”
NYPD Security Camera
This assemblage of devices and cameras is prominently marked as being a security camera from the New York Police Department (NYPD). It includes two cameras from Pelco, and mounted higher on the pole, a “Tsunami QuickBridge” from Proxim, a proprietary point-to-point wireless system. There’s no way to tell what other sensors are included in the large metal box in the center of the image. The clear marking by the NYPD suggest it is part of the “Lower Manhattan Security Initiative”, a post- 9/11 counterterrorism program to increase surveillance, with a combined budget of well over $100m from the City, NYPD, and the Department of Homeland Security. Feeds from the approximately 3000 cameras, together with other inputs including 911 calls, radiation sensors at bridges and tunnels, and mapped crime patterns, are monitored from a central command center in New York’s Financial District, and forms part of the “Domain Awareness System”, which was co-developed with Microsoft.
This camera installed on Federal Plaza in the Civic Center district of lower Manhattan - home of the NYC field offices of the FBI, the Department of Homeland Security, and the Social Security Administration, and falling under the jurisdiction of the United States Federal Protective Service rather than the NYPD - is prominently marked with the logo of the Department of Homeland Security. Also visible is a mark reading “CrimeEye,” a product name of Total Recall Corporation, which is a “video-centric security technology provider specializing in surveillance solutions” from Suffern, NY.
In a recent similar installation in New York, Total Recall Corp assembled a team of subcontractors including Axis Communications of Sweden and Scallop Imaging of Massachusetts providing cameras; Israeli company Briefcam and the Danish-based Milestone Systems contributing software; the Canadian company Dragonwave and Californian Proxim Wireless supplying networking components; Texas-based Pivot3 providing storage and processing capability; and California’s RGB Spectrum and Minneapolis’ Winsted Corporation combining to create a command center from which to track activity captured by the security system.
NYC DOT Camera
This CCTV camera manufactured by Sweden’s Axis Communications, and owned and administered by the New York City Department of Transportation, is suspended directly over an intersection. In a 2011 upgrade of old analog equipment, NYCDOT purchased video servers and several network cameras to better monitor traffic, and stream video and images to other government agencies and to the public over its website. The stream from this particular camera can be found at http://nyctmc.org/multiview2.php?listcam=441
It is impossible to tell whether it also feeds an Automatic License Plate Reader/Recognition (ALPR) system consisting of computer vision and optical character recognition software to recognize flagged vehicles.
This device mounted on top of a street light pole is a Vantage Vue Wireless Weather Station, manufactured by Davis Instruments of Hayward, California. It measures wind speed, temperature, humidity, barometric pressure, and rain rate, and transmits its results wirelessly. Although it is attached to standard street furniture on a public street, it’s impossible to tell if it belongs to or is administered by the city or by a private entity, and other than the branding on the side, has no markings to indicate ownership or purpose.
3D Tracking Radar & Camera
This 3D tracking Radar and camera are a red-light safety system manufactured by Arizona’s “American Traffic Solutions.” The radar portion of the system tracks passing vehicles’ location and, if a car runs a red light, the camera is activated. That image is then processed to recognize the license plate number, and a ticket is mailed to the registered address. A ticket issued by this system is treated differently to one issued by a police officer, with a lower fine and no points on the driver’s license, as it is difficult to prove who was driving the vehicle at the time. Due to state legislation, the number of red light systems is heavily regulated.
Outside Temperature Sensor
This small beige box mounted at a small distance from an exterior apartment building wall is a temperature sensor, manufactured by Heat-Time Corporation of New Jersey. It senses the external temperature and controls the heating of a building, most commonly by cycling a boiler found in the basement of the building. NYC Department of Housing Preservation and Development mandates that all apartment buildings in the city provide heating when the outside temperature drops below given temperatures, and these sensors help buildings abide by those laws.
New York City’s bike share system has many sensors embedded in both the docking stations, and the bikes themselves. Most prominently, the stations incorporate RFID readers to release a bike when a Citibike key is inserted, and to register when a bike has been docked, as well as to report system usage to a central server to allow it to be load-balanced by manually moving bikes from one station to another. The bikes themselves incorporate GPS chips, to enable missing bikes to be located.
Magnetometer / Parking Sensor
This unmarked disk sunken into the asphalt at the side of the street is a parking sensor from California-based Streetline, forming part of a pilot project run by the Roosevelt Island Operating Corporation. Using a magnetometer and unspecified “other sensors” together with wireless low power mesh networking, it reports the presence of a vehicle directly above and, by implication, whether the parking space in question is occupied or vacant. This potentially allows for mobile payment for parking, dynamic pricing, and for the data to be used to find empty spots using a smartphone app.
The small box protruding form the light fixture above a building entrance is a direct wire-in photocontrol manufactured by Mulberry Metal Products of New Jersey, costing in the region of $5-15 apiece. This simple technology senses ambient light, and is hardwired into the electrical circuit of the light itself to switch it on and off at specified light levels.
This apartment building intercom system is the Vario model from Siedle, a company from Furtwangen, Germany. When looking at the street through the lens of “sensors,” a common intercom system like this one at the entrance to a private apartment building presents itself as a networked camera and microphone combined with a speaker and a user interface.
Wireless sensors and fast processors are popping up everywhere, allowing us to generate volumes of real-time data about human behavior and our world. At frog we define sensing as the ability to harness these real-time data streams to identify patterns, generate insights, and design better experiences for people. As engineers crack the technical challenges, from ultra-cheap sensors to exabyte-scale data processing, designers must discover how we can adapt these technologies to human life.