The AMMOD project, funded by the Friedrich‑Alexander‑Universität Erlangen‑Nürnberg under grant 16LC1903G, ran from 1 January 2020 to 30 April 2023. Its aim was to enable autonomous, energy‑autonomous monitoring of biodiversity by integrating a wide range of sensor modules into a single base station that can store, process and transmit data to the AMMOD cloud. The work was carried out by the Chair of Hardware‑Software Co‑Design at the Friedrich‑Alexander‑Universität Erlangen‑Nürnberg in close cooperation with the Technical University of Hamburg (TUHH) and the Leibniz Institute for Innovative Microelectronics (IHP). The design of the base station (Module 6) was developed jointly with TUHH and IHP, while the connection to the AMMOD cloud was established together with partners from Module 7. Sensor‑module teams were actively involved in integrating their modules into the base station, for which a lightweight software interface was designed. The project was extended in response to the COVID‑19 pandemic and a global chip shortage; the Chair received approval to reallocate funds and successfully met all objectives within the extended period.

Technically, the project produced a methodology for optimising the dimensioning of AMMOD base stations. By analysing transmission and power‑consumption profiles of the sensor modules and the station’s location, the method calculates optimal sizes for photovoltaic arrays, battery capacities, storage capacities and energy‑management strategies. This allows a trade‑off between energy demand, cost, maintenance intervals and data availability. The methodology was validated through laboratory and field tests in Bonn and Hamburg, covering all seasons, and yielded satisfactory results.

A key deliverable was a system architecture for data processing on the base station. It comprises a lightweight interface to the sensor modules and a separate interface to the AMMOD cloud. A central system manager monitors battery level and can pause energy‑intensive cloud communication, compress data, and buffer it locally until sufficient energy is available. This adaptive behaviour reduces power consumption while ensuring that data are eventually transmitted. The architecture was prototypically implemented and integrated into the base station hardware. The use of Raspberry Pi boards, substituted for originally planned FPGA‑based boards due to chip shortages, required redesign of many concepts but was successfully adapted.

The project also addressed the challenge of handling large data streams from visual and acoustic monitoring. For example, an HD camera operating at 15 frames per second generates more than 90 MB per second; a stereoscopic camera would double this rate. Limited mobile‑bandwidth and storage capacity necessitated on‑site processing to reduce data volume before storage or cloud upload. The signal‑processing work package developed algorithms that compress and filter data in real time, thereby lowering the amount of data that must be stored or transmitted. Energy‑management strategies were designed to allow sensor modules to enter low‑power modes during periods of low solar generation, while still capturing and buffering data for later transmission.

Overall, the project achieved its goal of creating a flexible, energy‑aware base station capable of integrating diverse sensor modules, processing data locally, and reliably transmitting biodiversity information to the AMMOD cloud. The collaboration among the university, TUHH, IHP, and sensor‑module partners, combined with adaptive project management in the face of external constraints, ensured that all technical and scientific objectives were met within the extended timeframe.