Xandem Next Gen Intrusion Tested

Published Apr 26, 2017 10:00 AM

Xandem's "full coverage motion tracking technology" is unlike any intrusion technology we have seen.

We bought their new system and tested it extensively to better understand its strengths and weaknesses. Below is one of the examples from our testing showing movement within our test environment:

 

The plus side is that the technology works and is legitimately different. The negative side is how and where the technology is built to be used.

In this report, we analyze Xandem's technology in depth.

High Performing but Wrong Market

Xandem's offering is marketed toward the price sensitive low-end residential market, but given that Xandem's tracking and sensor technology is altogether different and outperforms what is currently available, we believe the better market is high-end systems that places a premium for the motion tracking features the technology offers.

Xandem's Market Impact

Xandem is unlikely to displace much of the existing intrusion market with this offering, but the company could be attractive either (1) repositioned as a higher-end enterprise offering or (2) a valuable acquisition target for incumbent intrusion providers like Bosch, DMP, Honeywell, Tyco, etc. who would benefit from their technology's differentiation.

Key Test Findings

Xandem does not work like traditional intrusion systems, and our testing illustrated some big differences:

  • Overall, the system worked well in our test, although motion detection was weaker at the edges and often misindicated movement by 2 or 3 feet because of signal strength weakness at maximum range of nodes.
  • Humans were detected with consistent accuracy every time, with no observed false detection or fail detection events during our test.  In some cases, human movement suddenly appeared in the middle of a room at extreme distances, but once detected movement was fluidly tracked.
  • We timed (using stopwatches) Latency between actual movement and displayed movement at 0.75 - 2.0 seconds, generally varied based on local network activity.  Busier network periods resulted in longer latency.
  • Sensitivity of the sensors was 'turned up', or made more sensitive to detect small children and pets under 40 pounds. At default sensitivity, small children and pets may not be detected.
  • The system itself still lacks fundamental features available in basic intrusion alarm systems, like zoning, conditional/area arming, partitioning, and central station integration.
  • The 'armed' feature does not indicate alarm events or triggers onscreen, and unless users are monitoring email or have a local sounder connected, notification of an alarm condition may not be obvious.
  • Detection at the perimeter of a building was weak because of blind-spots, that could only be addressed by locating nodes in outlets adjacent to outside walls.  This outlet location may not always be available, or even installed, in many floor plans.
  • In our testing, Xandem nodes did not exhibit 'line of sight' limitations or general false alarm susceptibility of typical motion PIR sensors.

Theory of Operation

The primary difference is that instead of magnetic contacts, Motion Sensors, and Glass Break Sensors, Xandem uses radio waves (2.403-2.480 GHz) in part of the same radio spectrum used by WiFi and ZigBee devices.

The net impact of the different approach is that sensors can 'see through' obstructions like walls and furnishings, instead detecting only organic mass movements with specific location resolution. Movement tracking accuracy depends on how well and how many system sensors relate to each other via a mesh, with higher sensor density resulting is better accuracy.

The company's demo and marketing video below describes the general function and operation of the system:

However, sensor placement on household plugs essentially limits the current Xandem offering to only detect motion and movement inside a building:

This presents a big difference and potential weakness compared to tradition intrusion systems. While those systems are designed to detect intrusion at the perimeter of a building, Xandem essentially only detects movement inside a perimeter.

System Overview

The system itself is composed of hardware devices that plug in to common AC voltage outlets and communicate to an internet connected hub that uploads real-time data to a web portal. The overview video provides a demo of the system and an overview of the features of the portal:

Kit Contents

The kit itself comes with the hardware nodes and the central hub needed to install the system:

The nodes are plugged into standard two-pronged outlets and communicate wirelessly to the hub.

Node Coverage Areas

Variations of the 10-node kit include both smaller and larger sizes based on the total square footage of coverage.

  • 6 node: Good for 500 sq. ft. of coverage
  • 10 node: Good for 1,000 - 1,500 sq. ft.
  • 20 node: Good for 2,000 - 2,500 sq ft.

Larger areas can use more nodes and multiple kits joined together in the web portal as a single system.

In addition to the 'Plug-In' node version, a version that uses DC power supplies and external wiring is available for commercial or areas that may not have outlets installed in the monitoring area. 

Kit Pricing

Xandem sensors kits [link no longer available] are available in several versions, with either 'plugin' power or external low voltage power supply options.

We tested the plugin, 10 node 'Standard' sized kit that cost $495, about $50 per node. The site claims the 10 node kit covers about the size of a small house or commercial office space and is the most common size. Larger 15 node (XL) kits cost $595, while small 6 node (MINI) kits cost $430.

Xandem disclaims that the number of nodes needed "May vary depending on construction materials and furniture. Heavier walls like brick or cement will require more nodes per total area."

For applications where convenience wall outlet power is not available, DC-wired version including power cable are offered at higher prices: 6 nodes cost $695, 10 nodes cost $995, and 15 nodes cost $1,295 in that configuration.

Xandem also offers the same kits in 'Pro' configuration that includes battery backup options for the nodes via AA batteries, 4 relay output contacts on the Hub for external integration to intrusion alarm systems, and access to the API. We tested the 'Home' kit version that does not include the external relay points or battery backup option.

Installation and Configuration

In our test, we plugged nodes into wall outlets and made note which node was installed where per Xandem's install instructions. Typical installed appearance occupies one outlet plug, and can be installed leaving one open in a standard duplex outlet:

For our ten-node system, installation took about 15 minutes total, including the time to connect and register the central node.

From there, a scaled floor plan must be drawn in the web portal, and nodes placed in the specific locations they are physically installed on the plan.  Xandem uses a 'drag and drop' wall tool on a grid, which is easy to use but very basic in detail. Nodes must be placed with regards to their factory numbering scheme on the floorplan in order to register movement accurately.

While unit placement is fairly precise (down to 1 foot), the level of floorplan details are essentially limited to basic lines. Drawing common features like doors, windows, or appliances is difficult and hard to decipher given everything uses the same line type, layer, and color, so most users are only going to draw basic wall locations. Xandem tells us that floorplan uploads are a development item for future releases:

Overall sensitivity of the nodes is adjusted with a slider as a general setting.  Individual node sensitivity cannot be adjusted, so if unwanted movement is detected (pets), then turning the gross sensitivity down is recommended.  We actually made the system 'more sensitive' based on sensor layout and distances, but the adjustment is very coarse and may not help in all situations.

Nodes vs. PIRs

Compared to standard motion PIRs, Xandem's nodes are slightly smaller. The image below is a side-by-side of a 2Gig wireless PIR, typically sized and comparable to other sensors:

Unlike PIRs intended to hang 7- 12 feet above the floor, Xandem nodes plug in close to the floor, about 1.5' - 2' above the ground. Despite the low elevation, our test did not detect node problems detecting motion at normal pedestrian heights, and could detect arms waving even if the subject stood still.

Another key difference: the mesh created between Xandem's nodes do the sensing, unlike the PIR that uses a traditional IR FoV for detection.

In general, signal strength of nodes is significantly important to control, with Xandem saying that node strength of less than '70%' resulting in poor performance. We tested our 10 node system with averages of 70%+ strength, and then spread out to far under that threshold and found those claims accurate.

The portal shows users a diagnostic in real time the signal strength, and node location adjustment updates on the fly.  The power quality and availability for each node is shown as well, with 'power lost' conditions being tracked:

The diagnostic feature is generally easier to use than a panel-based alarm system, since the real-time graphic displays make sensor fine tuning and placement changes easier to see and optimize.

Alarms, Notifications, and Rules

Our testing showed that while detection could be excellent, the portal lacked granular reporting and configurabililty of many intrusion systems.  For example, 'rules' that dictate action when motion is detected must be created, and output activity is limited to three basic default actions including 'send email', 'sound external speakers', or 'go to URL' for potential integrations with home automation platforms:

The central hub itself contains a 1/8" / 3.5mm "mini stereo" or "TRS" (tip/ring/sleeve) audio jack where computer speakers or similar externally amplified sounders can be plugged in.  Without these speakers, the system itself does not have local annunciation.

Weaknesses of System

While Xandem touts high-performing and uncommon detection sensors, it lacks basic features that would be problematic if used as a primary intrusion system:

  • No zones support
  • No suppression of individual nodes
  • No discrete arming
  • No partitions
  • No Central Station integration

While Xandem tells us the improvements like zones and discrete arming support is coming, the lack of these features now mean that Xandem systems are not suitable replacements for traditional alarm systems, even with the advanced and superior motion tracking ability.

Strengths of Systems

While Xandem lacks the depth of operational features needed for commercial intrusion, both the company and technology is relatively new and still being refined for commercial use.

The sensing technology is not vulnerable to many of the common sources of false alarms that plague PIR based motion sensors, and the system does not have the line-of-sight or blindspot issues that foil IR based technology.

Given the time to refine and add management features, Xandem sensors could be a clear advantage for protecting 'high security' locations where tracking granular movement in restricted or hazardous areas are needed, and the technology is higher performing and may be cheaper to install given the modest power and location requirements than existing sensor methods and devices.

Comments are shown for subscribers only. Login or Join