(6 min read)
Urbanisation and high population density in modern cities, combined with the lack of efficient public transport infrastructure, has led to the extensive use of public space for vehicle parking. In order to optimize parking space occupancy, municipalities deploy zoning systems to balance the use of the public space between the citizens’ and businesses’ needs.
1. In areas with high parking demand, metered parking zones are defined where citizens can park during a short amount of time, usually for a fee (e.g., commercial areas, public buildings, recreational areas, etc.)
Fuelics parking space occupancy monitoring devices in metered parking zone (© Swarmchestrate consortium 2024-2026)
2. In order to support the logistical operations of local businesses, designated loading and unloading areas allow short term parking for transport vehicles at specific times within the day, usually during off-peak hours.
Fuelics parking space occupancy monitoring device in loading-unloading zone (© Swarmchestrate consortium 2024-2026)
3. In residential areas, restricted parking zones are defined where only local residents are allowed to park.
4. Especially for citizens with disabilities, designated parking spaces are allocated where parking is allowed for special permit holders only.
Fuelics occupancy monitoring device in parking space for disabled (© Swarmchestrate consortium 2024-2026).
The monitoring and management of these zones becomes increasingly difficult as the city size increases. The authorities do not have the capacity to patrol and enforce the proper use of the designated parking zones, which leads to poor experience for citizens and businesses alike, as well as to economic loss in case of metered parking zones. In order to improve the efficiency of the monitoring of parking spaces occupancy, municipalities deploy parking space occupancy monitoring IoT devices city wide.
Fuelics is a Greek based deep-tech company that designs and develops battery operated edge computing sensors which are utilised to provide real time monitoring of parking space occupancy. The devices utilise both radar and magnetic flux sensors to identify if an object is occupying a parking space. In addition, Bluetooth Low Energy (BLE) tags are identified by the device in case the parking space is designated for specific users. The purpose of the parking space occupancy monitoring IoT devices is twofold:
On the one hand, to provide real-time view of the occupancy status of selected areas and generate an alert in case parking conditions are violated, e.g., overrun parking time, parking at a space without the proper permit, etc. This allows the municipalities to use their patrolling teams more efficiently.
On the other hand, to collect occupancy data in order to adapt the zoning requirements as the ways in which citizens and businesses use the public space evolve.
Due to the large geographical area the parking space occupancy monitoring sensors are deployed, and in order to reduce the cost of their installation, the devices are battery operated and use the Narrowband Internet of things (NB-IoT) Wide Area Network (WAN) for transmitting the occupancy information. The reason battery operated devices are preferred is to avoid the additional cost of the public works required to connect these devices directly to the power grid. In terms of WAN technology used, NB-IoT, which uses the mobile operators’ cellular network is preferred as it does not require the deployment of additional wireless network infrastructure. In order to support identification of parking events in spaces where a specific permit is required, the parking space devices support BLE and can detect proximity tags to identify if the vehicle is allowed to park or not. The BLE mode is also used during device commissioning and allow field engineers to connect to the devices via their mobile phones, change their configuration and also retrieve the current occupancy status for testing purposes.
The primary factor that affects the operational cost of these devices is the battery life, as the cost of replacing them can become prohibitively high when thousands of devices are deployed all over a city. Related to the energy consumption of the devices, most of it is spent on WAN data transmission. The parking space occupancy monitoring devices connect and transmit their data when a vehicle occupies or leaves a monitored parking space. The connection frequency cannot be known in advance and it is therefore not possible to predict how long the battery of the device will last. In terms of specifications, the common practise is to provide an estimate of the number of connections a device can support with a single battery pack. If the frequency of parking events is known, the device’s expected operational time can be calculated.
The energy required to transmit a message over NB-IoT (Wide-Area communication) in ideal conditions is ~15 times higher than the energy required to transmit a message over BLE (Local-Area gateway), as Fuelics have established during measurements in their lab. This insight may lead to a deployment strategy that can increase the life expectancy of parking space occupancy monitoring sensors.
The primary difference between NB-IoT and BLE is that the former has adequate power to transmit information directly form the device to the cloud in a push fashion (hence the increased energy needs), whereas the latter requires a local gateway to poll the status of the device and take over the push of the information to the cloud. The wireless nature of BLE gives some limited flexibility on the placement of the local gateway which can be close to a power source thus not requiring to be itself battery operated.
The choice between these two operational modes is not clear-cut and may vary depending on the parking zone requirements as well as the time of day for each individual device. The primary metric to determine the optimal selection is the expected frequency of the parking events. In case the parking spaces are in a busy area and allow only for short-time parking (e.g., 10-15 minutes) the number of connections per hour will be high. Similarly, if the parking spaces are used for parking occupancy avoidance in busy areas, the connection frequency can also be high due to vehicles occupying the space for just a few minutes. This favours the use of a local BLE gateway operating in polling mode. On the other hand, outside of busy hours, during the night, or in residential parking zones, the parking events frequency will be low which favours the use of direct transmission via NB-IoT in push mode.
In an attempt to optimise the battery usage and maximise the operational lifetime of the parking space devices, the Parking Space Management demonstrator will leverage the Swarmchastrate cloud-to-edge continuum orchestration capabilities to automate the re-configuration of the parking space devices and alternate their operational mode between NB-IoT/push and BLE/poll. When the connection frequency is expected to be high, the configuration will dynamically switch from Wide-Area communication technologies (NB-IoT) to the use of a Local-Area gateway (BLE), which has lower energy requirements for connection and data transmission. When the connection frequency is expected to be low, the opposite will happen and the configuration will dynamically switch from Local-Area (BLE) to Wide-Area (NB-IoT) communication. This will allow the maximisation of the battery life without jeopardising the real time identification of parking events. By leveraging the Swarmchestrate cloud-to-edge continuum orchestration capabilities, it is expected that the battery life of the deployed sensors will consistently exceed the current 5 year expected operational lifespan. An additional benefit of leveraging the Swarmchestrate orchestrator in this use case is that once the application is deployed, it can scale from hundreds to thousands of devices in widely dispersed geographical areas supporting different parking zone configuration requirements.
Swarmchestrate is a very promising orchestrator ecosystem that can be utilised in Smart City IoT applications and enable the implementation of use cases that are currently not feasible with existing Smart City Platforms. The Fuelics bi-modal parking space occupancy monitoring demonstrator is a case study that will demonstrate these capabilities and at the same time have a tangible impact on the operational efficiency of Smart Municipalities.
Editors:
Evangelos Angelopoulos, PhD, Fuelics CEO & Head of Business Development
Ioannis Nikolaou, PhDc, Fuelics Head of Cloud & Platforms