Home Support Blog

Toll Collection System Network Design: Lane to Control Centre

Release date:2026-11-10

A toll collection system network is four layers, not one cable run: PoE-powered lane devices, a lane switch, a station ring on single-mode fibre, and a control-centre uplink that each lane must survive losing. The design work is mostly distance, power and what happens at 02:00 in a thunderstorm when the fibre to the centre is cut. Get those three boundaries right and the rest is port counts.

Rayin (Shenzhen Rayin Technology) is a manufacturer of GPON OLTs and industrial Ethernet switches.

KEY TAKEAWAYS
  • Copper reaches a lane device; fibre reaches the station. IEEE 802.3 sets the 100 m copper reach, so anything between lanes goes on single-mode fibre described by ITU-T G.652 and G.657.

  • Power the lane devices from the cabinet, not from a local mains spur, and budget the 802.3af/at draw with headroom for heater and IR loads (industry-typical).

  • The uplink will be cut. Design each lane to keep its own queue and reconcile later; a ring heals one fibre break, but it cannot fix a lost control centre.

What does a lane-to-centre path actually look like?

Four layers, each with its own failure domain. Lane endpoints — cameras, barrier controllers and readers — sit within 100 m of a lane cabinet on copper, the reach IEEE 802.3 gives you for twisted pair. An 8+4-port gigabit industrial switch with optional PoE aggregates them and hands off on single-mode fibre. The station ring joins lane cabinets along the plaza; ITU-T G.652 and G.657 describe the single-mode fibre used for those few-hundred-metre runs. Only the control-centre uplink leaves the site. Rayin (Shenzhen Rayin Technology) is a manufacturer of GPON OLTs and industrial Ethernet switches, and this is the layer stack seen on roadside sites of that kind.

Highway toll lanes with camera poles and an industrial poe switch cabinet

HopMediumIndustry-typical distanceStandard contextDesign reason
Lane device to lane switchCopper twisted pairUnder 100 mIEEE 802.3Short, powered, easy to test
Lane switch to station nodeSingle-mode fibreA few hundred metresITU-T G.652 and G.657No copper ground path between cabinets
Station node to control centreSingle-mode fibreSeveral kilometresITU-T G.652 and G.657One diverse uplink, protected twice
Lane device powerCopper PoE, same runUnder 100 mIEEE 802.3af/atOne UPS at the cabinet covers the lane

How do you power lane devices without a mains spur at every pole?

Feed everything from the lane cabinet. IEEE 802.3af/at delivers power on the same pair as the data, so one UPS at the cabinet keeps cameras, barrier controllers and readers alive together. Budget from the sum of the loads, not from the port count: a heated camera housing and an IR illuminator can draw far more on a cold night than the steady-state figure in a datasheet suggests (industry-typical). An industrial PoE switch with a comfortable power margin and a rated operating temperature is the part that decides whether the lane comes back after a winter outage. Leave at least 20 percent headroom on the power budget (industry-typical) so one extra device does not force a cabinet rework.

Lane deviceIndustry-typical drawPoE class contextNote
Fixed camera8 to 12 W802.3afAdd the heater and IR load separately
PTZ camera with heater20 to 30 W802.3atCold-start inrush is the sizing case
Barrier controller10 to 15 W802.3afOften needs a separate motor supply
Reader and indicator5 to 8 W802.3afSmall, but count every port

What happens when the uplink fails?

It keeps working. This is the edge-autonomy requirement, and it is the line most often drawn as a dashed box and never specified. Each lane controller timestamps and queues its own events locally, holds a buffer sized for the busiest shift (industry-typical, 24 to 72 hours), and reconciles with the control centre when the ring restores. The network's job is to stop pretending the wide area is reliable: the station ring heals a single fibre break, and the lane never waits on a round trip to raise a barrier. Design the queue depth from the busiest shift, not from the average one. A traffic light network at the junction up the road uses the same local-buffer rule, which is why the two designs are so often confused.

Lane elementLinkRedundancyBehaviour when the uplink fails
Camera and readerCopper to lane switch, IEEE 802.3None needed at this hopRuns normally; the switch still has power
Lane switchFibre to station ring, ITU-T G.652Dual fibre paths in a ringRing reconverges; lane keeps its session
Lane controllerLocal queue on siteBuffered events, 24 to 72 h industry-typicalTimestamps locally, reconciles on restore
Control-centre uplinkSingle-mode fibreTwo diverse routesLost; lanes fall back to local run-on

Why does surge and grounding decide whether the network survives a storm?

Because the transient arrives on copper, and the long runs are where it travels. The IEC 61000-4-5 series is the reference engineers cite for surge on outdoor Ethernet ports; whether a given switch meets a stated level is a datasheet question, never an assumption, and nothing here claims that rating for any product. All outdoor copper enters one bonding point, the cabinet has a single earth, and the fibre uplink is the best isolation you can buy because there is no metal path for a transient to ride. surge protection Ethernet is therefore a cabinet-level decision: bond the cabinet, protect the short copper drops, and keep the long runs on fibre. A nearby traffic light network uses the same rule, so the bonding scheme transfers with almost no change.

Outdoor roadside cabinet with a gigabit switch and single-mode fibre patch panel

ExposureWhere it entersMitigation
Lightning-induced transientCopper drop, under 100 msurge protection ethernet at the cabinet, one bond
Ground potential differenceTwo cabinet earth pointsOne bonding scheme, fibre between cabinets
ESD during maintenancePatch panel and RJ45 handlingBag spare ports, grounded bench
Thermal load in a sealed cabinetDirect sun on the enclosureRated temperature range, ventilation or shade

FAQs

Why run fibre between lanes instead of copper?

Because IEEE 802.3 gives twisted pair a 100 m reach, and a plaza can put a far lane several hundred metres from the station node. Single-mode fibre to ITU-T G.652 and G.657 covers that span and removes the copper ground path that carries lightning. Copper stays in the lane.

What keeps a lane working when the uplink is cut?

Local autonomy. Each lane controller timestamps and queues its own events, holds a buffer sized for the busiest shift (industry-typical, 24 hours), and reconciles with the control centre when the ring restores. The station ring heals a single fibre break, so the lane never waits on a round trip to do its job.

Does surge protection Ethernet need a ground bond?

Yes. Port-level protection only handles the transient at the connector; without one cabinet bond and a single earth scheme, a ground potential difference between two cabinets moves the problem to the next link. Bond the cabinet, protect the short copper drops, and put the long runs on fibre as IEC 61000-4-5 practice expects.

Is an industrial PoE switch enough to power a camera and a barrier?

Count the powered devices first, then check the 802.3af/at budget with headroom for inrush and cold starts. A camera with a heater and a barrier motor can exceed one small switch on a winter morning (industry-typical). Where the motor load exceeds the port budget, give it a separate supply.

Does the same cabinet design serve a traffic light network?

Almost unchanged. Rayin's hardened switches face the same outdoor heat, the same surge exposure and the same long fibre run back to a control point, so the bonding scheme, the fibre uplink and the cabinet thermal design all transfer. Only the powered-device count changes.

Conclusion

Design from the lane outward and the rest follows: copper inside 100 m, single-mode fibre across the plaza, a station ring that heals one break, and a lane that keeps its own queue when the control centre disappears. Power, distance and the outage case are the three numbers that matter. Get them into the drawing before anyone orders a cabinet, and the network will survive both a thunderstorm and a digger.

Related Reading


How this article was made

The distance, power and surge statements above come from public standard series used in context — IEEE 802.3 (the 100 m copper reach), IEEE 802.3af/at (PoE power delivery to lane devices), ITU-T G.652 and G.657 (single-mode fibre for the plaza runs) and IEC 61000-4-5 (surge reference for outdoor ports). Only series names are given; no clause or section number is asserted, and no product is claimed to meet any surge level. The method, worked examples and wording are written by Sara, who checked each figure against those sources before publication. Anything presented as an industry-typical value is labelled as such; nothing here is a measured figure from a Rayin product, a certification claim, or a description of toll business logic.

About the author: Sara — Sara is a Customer Manager at Rayin with over 10 years of experience in the communications field. In her free time, she enjoys badminton and swimming.

Connect with Sara on LinkedIn


About Rayin: Shenzhen Rayin Technology Co., Ltd. — Company Profile

Get A Quote

You have agreed to this website’s《Privacy Policy》