Optimization of energy saving for wireless sensors (1)

After introducing a series of wireless sensors, I will continue to elaborate on the Energy Saving Problem of wireless sensors in the last node article. To save energy, you need to optimize system problems, solve the energy consumption problems of various problems, and improve the system through computing.

1. Energy Saving Optimization for a Single Node
After analyzing the composition and energy consumption characteristics of wireless sensor nodes, let's see what measures can be taken on a single node to save energy consumption.

1) energy-saving computing

In addition to using low-Power hardware in the node design, Dynamic Power Management (Dynamic Power Management, DPM) and other technologies allow all parts of the system to run in the energy-saving mode, which can also save a lot of energy. The most common power management policy is to disable idle modules. in this state, wireless sensor nodes or some of them will be shut down or in a low power state until any event of interest occurs. The core issue of DPM technology is the State scheduling policy, because different states have different power consumption characteristics, and status switching also has energy and time overhead.

In the active state, Dynamic Voltage Adjustment (Dynamic Voltage Scaling, DVS) technology can be used to save energy. Computing load changes over time on most wireless sensor nodes, so you do not need the microprocessor to maintain peak performance at all times. DVS uses this technology to dynamically change the operating voltage and frequency of the microprocessor so that it can meet the current operation requirements and strike a balance between performance and energy consumption.

2) energy-saving Software

If system software such as the operating system, application layer, and network protocol are specially optimized for energy consumption, the survival time of the wireless sensor network can be effectively extended.

In the operating system, dynamic power management and dynamic voltage adjustment are the most suitable, because the operating system can obtain the performance requirements of all applications and directly control the underlying hardware resources, thus, the necessary compromise between performance and energy consumption control is made. The core of the operating system is the task scheduler, which schedules a given set of tasks to meet their respective time and performance requirements. By considering the energy-saving problem in task scheduling, the system survival time can be significantly prolonged.

In view of the inevitable data loss during transmission, wireless sensor networks should be able to provide data of different precision based on the current network environment, so as to obtain a certain degree of elasticity. On the other hand, the attributes of monitoring objects change over time, resulting in changing computing and communication requirements in the network. In this way, we can make certain predictions in real-time scheduling algorithms to actively manage energy consumption. In addition, the application layer can be designed to execute major computing tasks as early as possible, and then stop the task before the algorithm ends normally. In this way, the energy consumption can be reduced without affecting the data accuracy.

3) wireless transceiver circuit energy consumption management

Although the power management of embedded processors has been thoroughly studied, the energy-saving design of Wireless Transceiver systems is not enough. Wireless Communication accounts for the majority of the energy consumption of the entire wireless sensor network, so it is very important to manage the energy consumption of the Wireless Transceiver System.

The power consumption of the wireless transceiver system mainly comes from two parts, depending on the transmission distance, the RF part of the modulation parameters, and the baseband circuit part for frequency synthesis, filtering, and other operations. The energy-saving design of the wireless transceiver system is complicated because the power overhead of the RF part and the baseband circuit part are equivalent. Reducing the rate of the RF part will lead to an increase in energy consumption. In addition, the initialization overhead of the wireless transceiver system is very high, which makes it more difficult to design the energy-saving Wireless Transceiver System. The Energy Consumption Optimization for wireless communication will be discussed in detail in the next section.

4) energy-saving message forwarding

In addition to sending self-sensing data, each wireless sensor node is a router and needs to forward packets to other nodes. In a typical wireless sensor network, most of the messages received by wireless sensor nodes are about 65%. Generally, most of the protocol processing functions of wireless sensor nodes are handed over to MCU. In this way, no matter where the destination is, each received packet will go through the same processing steps to reach the computing subsystem and be processed, resulting in unnecessary energy consumption. With the smart Wireless Transceiver System, packets to be forwarded can be identified and forwarded directly in the communication subsystem, even when the computing subsystem is sleeping.

2. Energy-saving Optimization of wireless communication
Similar to the energy consumption management of a single node, energy-saving measures during inter-node communication also play an important role in improving the power usage efficiency of the entire system, to make the communication process sensitive to energy consumption, the range of energy-saving optimization can be expanded from a single node to multiple nodes involved in the communication.