Dissolved oxygen (DO) in water is one of the most important ecological factors for the healthy growth of aquaculture organisms. The too high concentration can easily cause the fish bubble disease. The low concentration can slow down the growth of fish, where the prolonged floating head can cause the fish to float. There are temporal and spatial variations in the DO of aquaculture water under various environmental factors in large-scale aquaculture. The existing predictions cannot accurately describe the spatial distribution of DO, due to the time series. In this study, an improved prediction model of spatiotemporal DO was presented for the intelligent control of water quality in ponds. The high aquaculture efficiency with fewer risks was then realized using an improved gated recursive unit using a dual attention mechanism (IDA-GRU) and an improved inverse distance weighted interpolation algorithm (IIDW). Firstly, the feature and temporal attention mechanism were introduced to improve the prediction accuracy of DO using the gated recurrent unit (GRU), according to the correlation between parameters and the dependence relationship of temporal information. The key environmental features were also strengthened using the feature attention mechanism. The contribution rate of each environmental feature was then calculated to continuously modify the weight of each environmental feature in real time. The time attention mechanism autonomously extracted the key historical moments, and then enhanced the expression of the key moment, thus improving the stability of DO time prediction. Secondly, an improved inverse distance-weighted interpolation (IDW) was applied to achieve spatial prediction using the prediction of DO time series. The distance weight correction coefficient was also introduced using the traditional inverse distance-weighted interpolation. The interpolation weights were adjusted in real time, according to the DO content at each monitoring point. An improved accuracy was achieved in the three-dimensional spatial interpolation of DO. The local environment was considered around the monitoring points. As such, the interpolation weights were changed concurrently rather than that in the traditional algorithm, once the interpolation distance was determined. Finally, the improved model was validated at the digital eco-farming site of Shanghai City electric development Co. The experimental results showed that the mean square error (MSE), root mean square error (RMSE), and mean absolute error (MAE) of the evaluation indexes were 0.0687 mg²/L², 0.2621 mg/L, and 0.2051 mg/L, respectively, in the IDA-GRU model for the DO time series prediction. Additionally, the MSE, RMSE, and MAE of the IIDW were 0.2088 mg²/L², 0.4570 mg/L, and 0.3835 mg/L, respectively, for the spatial prediction of DO. A better performance was achieved, compared with the previous models. The present model had improved the accuracy of spatiotemporal DO prediction. The finding can also provide a strong reference for disaster prevention and intelligent regulation of water quality in large-scale aquaculture.

Long range radio (LoRa) can be expected to be widely applied in large-scale and standardized aquaculture. A low-power and low-cost communication solution can also be provided for the Fishery Internet of Things (IoT). Particularly, the reason can be its long transmission distance and independence from the network operators. Furthermore, the stability of the wireless signals can depend mainly on the complex environment of the large outdoor farming areas and seasonal variations in the air temperature and humidity. Moreover, the scaling of the aquaculture can lead to a sharp increase in the number of IoT devices. Data collision problems have been more prominent in the wireless network performance. In this research, the multi-spreading factor (SF) characteristics of the LoRa were fully balanced to design a LoRa multi-channel communication mechanism suitable for the fishery IoT. A joint Spreading Factor and time slot allocation strategy was also proposed, termed AOA-CB-SS. The SF allocation was realized for the different node types in order to avoid the mutual interference between the communication links of the sensor and controller nodes, thereby enhancing the stability and reliability of the network system. Firstly, the node type, Received signal strength indicator (RSSI), and signal-to-noise ratio (SNR) were used as the inputs into a CatBoost (CB) model. The arithmetic optimization algorithm (AOA) was employed to optimize the hyperparameters of the model, thus enabling more accurate and rapid allocation of the spreading factors (SF) into the nodes. According to the current SF distribution within the network, the hierarchical slot selection (SS) strategy was utilized to assign the time slots to the sensor nodes, thus ensuring the resource allocation of the balanced network. The SS strategy of the network allowed for multiple devices with different Spreading Factors, in order to communicate simultaneously within the same time frame. While the devices were equipped with the same Spreading Factor communicated in different time slots. Simulation and experimental results demonstrate that the AOA-CB-SS strategy was achieved in an SF allocation accuracy of 98%. Its packet delivery ratio (PDR) increased by over 9.89%, and average energy consumption was reduced by 16.8%, compared with the similar strategies. The time slots were reallocated to permit the nodes using different SFs. Thereby, the packet collisions were reduced to simultaneously increase the wireless bandwidth for the scalability of the network system. Field tests showed that an average PDR of 96.7% was obtained for the sensor nodes. Furthermore, the success rate of the controller nodes reached 96.5% for the single-attempt transmissions of the gateway control messages. New perspectives were also offered for the application of the large-scale fishery IoT. Therefore, the LoRa IoT systems can be effectively extended to the agricultural domains (such as large-scale fruit and vegetable cultivation). It is only required to train the model with the specific datasets in the application scenario, and then modify the upload period, time slot length, and number in the SS component.