Abstract: Wireless sensor networks (WSNs), composed of sensors, microprocessors, and wireless communication interfaces, are a technology that is increasingly attracting research interest. Their broad application prospects have led to rapid development in fields such as healthcare, environmental monitoring, and the military. This paper summarizes the key issues involved in WSN research, introduces some major solutions, points out their advantages and disadvantages, proposes some perspectives and ideas, and looks forward to the future of WSNs in light of current international and domestic technological frontiers. Keywords: Smart sensor; Wireless sensor network; Routing protocol; Ad hoc network I. Introduction Wireless sensor networks are a type of independent computer network. Their basic unit is the node, which integrates four modules: sensor, microprocessor, wireless interface, and power supply. Mature solutions from traditional computer network technologies can be applied to WSNs. However, based on the specific uses and advantages of WSNs, developing dedicated communication protocols and routing algorithms has become an urgent research topic in the current field of WSNs. II. Characteristics of Wireless Sensor Networks 1. Wireless sensor networks involve large-area spatial distribution. For example, in military applications, wireless sensor networks can be deployed on the battlefield to track enemy military operations. Intelligent terminals can be mass-produced and embedded in propaganda materials, bullets, or shell casings, and scattered at target locations to form a large-area surveillance network. 2. Energy-constrained: The power supply of each node in the network is limited. Since the network often operates in uninhabited areas or harsh environments harmful to humans, power replacement is almost impossible. This necessitates low power consumption to extend the network's lifespan and minimize power consumption as much as possible. 3. Automatic network configuration and node identification: This includes automatic network formation, authentication of incoming terminals, and prevention of unauthorized user intrusion. Compared to sensor networks deployed in pre-designated locations, wireless sensor networks can be configured using ad hoc methods, provided there is a suitable communication protocol to ensure automatic operation without human intervention. 4. Automatic Management and High Collaboration: In wireless sensor networks, data processing is handled by the individual nodes. This reduces the amount of data transmitted over the wireless link, as only information relevant to other nodes is transmitted. This data-centric nature is another characteristic of wireless sensor networks. Because nodes are not pre-planned and their locations are not predetermined, some nodes may cease operation due to numerous errors or inability to perform designated tasks. Configuring redundant nodes is necessary for monitoring target objects within the network. Nodes can communicate and collaborate, sharing data to ensure comprehensive data on the monitored objects. For users, sending a data request to all sensors within the observation area and then sending the collected data to the designated node for processing can be achieved using a multicast routing protocol. This requires a unique address table. Users do not need to know the specific identification number of each sensor, allowing for a data-centric networking approach. 5. Differences from Mobile Ad Hoc Networks: As a distributed sensor network, wireless sensor networks share similarities with mobile ad hoc networks, but also have many differences. Mobile ad hoc networks can be used in situations where wireless infrastructure is unavailable or inconvenient to set up due to cost and security considerations. However, sensors are often deployed in near-ground environments, where ground wave absorption cannot be ignored, and the multi-user interfaces in high-density sensor networks also result in high bit error rates. As one of the two basic networking modes of mobile communication, the transmission model in mobile ad hoc networks is a typical many-to-many model, while the transmission model in sensor networks leans more towards a hierarchical model (many-to-one transmission). Generally speaking, nodes in wireless sensor networks have more resource-constrained requirements than typical mobile terminals or handheld devices, but computational requirements are optional. When computational tasks are required, if the communication cost is lower than the computational cost, the computational task is sent to the central node for execution. III. Key Issues in Wireless Sensor Networks 1. Network Security Protocol Issues Sensor networks face different security threats than mobile ad hoc networks, so existing network security mechanisms are not suitable for this field, and specialized protocols for wireless sensor networks need to be developed. One approach is to start from the perspective of maintaining routing security and find the most secure routes possible to ensure network security. Reference [1] points out that if the routing protocol is broken and the transmitted message is tampered with, then there is no security for the data packets on the application layer. The paper introduces a method called "Security Awareness Routing" (SAR), which is to find out the relationship between the real value and the node, and then use these real values ​​to generate a secure route. This method solves two problems, namely how to ensure that the data is transmitted in a secure path and the information security in the routing protocol. The paper assumes that two officers use the Ad Hoc On Demand Distance Vector Routing (AODV) protocol to communicate through an ad hoc network. Their communication is based on the Bell-La security model (PadulaBell-La Padula Confidentiality Model) [2]. In this model, when the security level of a node does not meet the requirements, it will automatically withdraw from the routing selection to ensure the routing security of the entire network. Reference [3] points out that the robustness of the system can be improved by multipath routing algorithms. Data packets are transmitted forward in the multipath path through the routing selection algorithm and reconstructed in the receiving end by forward error correction technology. The number of sensors in wireless sensor networks is large and their functions are limited. The routing scheme in mobile ad hoc networks cannot be directly applied to wireless sensor networks. Therefore, this paper presents a mesh multipath routing protocol. This protocol uses selective forward transmission of data packets and end-to-end forward error correction decoding technology. Combined with a mesh multipath search mechanism suitable for sensor networks, it can reduce signaling overhead, simplify the node database, and increase the system throughput. Compared with data packet duplication or finite flooding, this method consumes less system resources (such as channel bandwidth and power). Another idea is to focus on security protocols. A large number of research results have emerged in this field. In the literature [4], the author assumes that the task of the sensor network is to provide security protection for high-ranking officials. Providing a security solution will bring a universal model for solving such security problems. In terms of specific technical implementation, it is first assumed that the base station is always working normally and is always secure, meeting the necessary computing speed and memory capacity. The base station power meets the requirements of encryption and routing. The communication mode is point-to-point, and the security of data transmission is guaranteed by end-to-end encryption. The radio frequency layer is always working normally. Based on the above premises, typical security problems can be summarized as follows: (1) Information is intercepted by illegal users; (2) A node is destroyed; (3) Identifying fake nodes; (4) How to add legitimate nodes to the existing sensor network. The proposed scheme does not use any routing mechanism. In this scheme, each node and the base station share a unique 64-bit key Keyj and a public key KeyBS. When the distance between the node and the base station exceeds the predetermined distance, the network will select a node between the node and the base station as a relay node; the sending end will encrypt the data, and the receiving end will select the corresponding key to decrypt the data according to the address in the data after receiving the data. This double encryption method can prevent the exposure of the number and address of nodes, and can also prevent the data from being illegally intercepted. Even if individual nodes are decrypted, only their own key will be leaked, and the entire network can still work normally. Reference [5] introduces two dedicated security protocols in wireless sensor networks: SNEP (Sensor Network Encryption Protocol) and µTESLA. The function of SNEP is to provide authentication, encryption and refresh of data between the node and the receiver, and the function of µTESLA is to authenticate broadcast data. 2. The problem of node mobility management in large-scale sensor networks is essentially the node query problem in wireless sensor networks without wireless infrastructure. The simplest resource query method is global flooding, but it is not suitable for wireless sensor networks with limited resources. Therefore, global flooding should be avoided as much as possible in the design work. Expanding ring search repeats flooding by increasing the time-to-live (TTL). This method and the methods derived from it are also not suitable for wireless sensor networks. In terms of improving the efficiency of flooding, the scheme proposed in reference [6] is to reduce the redundant messages that occur when querying each node to reduce the inherent redundancy of flooding. In the absence of obvious redundancy, this scheme does not contribute much to improving efficiency. In ad hoc networks, node query is implemented through a hierarchical table based on clusters and landmarks. This method requires setting up a complex coordination mechanism between nodes. When a node moves or the cluster head or landmark fails, the hierarchical table needs to be reconfigured. Moreover, cluster heads usually become a bottleneck, so we usually avoid such hierarchical coordination tables and avoid using cluster heads. The technique proposed in GLS[7] is based on a network mesh graph that all nodes know. Nodes use a location server to store their locations and use an ID number-based algorithm to update their locations. When a node looks for the location of a node with a specified ID number, it also uses this algorithm to look for the location of the target node on the server. This method is good for nodes that know the network mesh graph and their own locations and know the ID number of the target node. Reference [8] introduces a query method for large-scale mobile sensor networks. This method borrows the concept of small worlds, uses the mobility of nodes to improve query efficiency, and introduces the concept of contacts. Its working principle is to first establish contacts between adjacent nodes, and then associate new adjacent nodes when they move, thus improving query efficiency. Unlike traditional routing query methods, the basic goal of this design is not to optimize routing or response delay, but to reduce the system overhead of communication. This is very important in energy-constrained environments, especially for one-time queries in networks with a large number of sensors (the communication lifespan is very short). The protocol given in the paper is scalable and automatically configured, which is very suitable for the mobility requirements of nodes. Simulation results show that it improves efficiency by 60-70% compared with edge flooding, 80-90% compared with flooding, and has a greater improvement than the extended ring search method. For distributed positioning in wireless sensor networks, reference [9] compares three positioning algorithms: ad hoc, robust positioning, and N-hop multilateration. The specific choice of which algorithm to use depends on certain network parameters, such as error distribution and connectivity. 3. Automatic configuration and automatic recovery of the network and maintenance of system energy efficiency When wireless sensor networks are deployed in unattended environments, it is almost impossible to change the energy source. In order to save energy, the transmission power should be as small as possible, the transmission distance should be short, and the communication between nodes requires intermediate nodes as relays. In earthquake relief or unmanned aerial vehicles, the automatic configuration and automatic recovery functions of the network are extremely important, and the scalability of large-scale multi-hop wireless sensor network systems is also a key issue. One approach to achieving measurability is "divide and conquer," or hierarchical control. This involves dividing network nodes into clusters using a clustering criterion, selecting a leader within each cluster to represent the cluster at a higher level. The same mechanism is applied to the cluster leaders, creating a hierarchy where each level applies local control to achieve a global objective. Most wireless network classification approaches assume the network is geographically independent, classifying clusters based on the number of nodes and the logical diameter between clusters (relative to geographical diameter). However, when the links between the cluster leader and other nodes within the cluster are long, adjacent clusters have significant geographical overlap, and the routing traffic load between different clusters is unbalanced, communication between a non-leader node and its cluster leader via their only long link will consume more energy. Furthermore, parallel communication conflicts between adjacent clusters become frequent, leading to unbalanced energy consumption. This results in a significant reduction in network lifetime, communication quality, and effectiveness. Therefore, in order to save energy and improve communication quality and effectiveness, the geographical radius of the cluster should be considered when designing the cluster algorithm. Reference [10] proposes to use a simple cell clustering structure to construct the routing protocol within the sensor node, so as to maintain a measurable energy-efficient system. The key issue is to make this cell cluster structure have automatic recovery capability. The authors propose a distributed algorithm for automatic configuration and automatic recovery of large-scale multi-hop sensor networks. This algorithm can guarantee that the network nodes are automatically configured into a cell cluster structure in two-dimensional space. Its cell units have a compact geographical radius and the overlap between cell units is also small. This structure is automatically recovered under various disturbances, such as node joining, leaving, dying, moving, being captured by the enemy, etc. Reference [11] gives a distributed algorithm for clusters called LEACH, which handles disturbances by repeating cluster operations globally. However, this algorithm cannot guarantee the location of the clusters in the system or the number of clusters. Reference [12] gives another cluster algorithm, which only considers the logical radius of the cluster and does not consider the geographical radius. When there is a large overlap between clusters, this method will reduce the effectiveness of wireless transmission. In addition, its recovery is not processed locally, but depends on the message being looped multiple times throughout the system. Reference [13] gives an access-based cluster algorithm that focuses on the stability of the cluster, does not consider the size of the cluster, and requires each node to have the support of the Global Positioning System (GPS). 4. System power consumption problem When wireless sensor networks are applied to special occasions, the power supply cannot be replaced, so the power consumption problem is crucial. In the power consumption model of the system, we are most concerned about: (1) the operating mode of the microcontroller (sleep mode, operating mode, potential slowing down of the clock rate, etc.), the operating mode of the wireless front end (sleep, idle, receive, transmit, etc.); (2) the power consumption of each functional block in each mode, and which parameters it is related to; (3) the mapping relationship between the transmit power and the system power consumption when the transmit power is limited; (4) the conversion time and power consumption of switching from one operating mode to another (assuming that it can be directly converted); (5) the receiving sensitivity and maximum output power of the wireless modem; (6) additional quality factors (such as the temperature drift and frequency stability of the transmitting front end, the standard of the received signal strength indication (RSSI) signal, etc.). Based on the above considerations, reference [14] proposed a protocol for self-organizing low-power networks, i-Beans, and specifically explained the power consumption of this network. For example, if powered by a small 220mAh button battery, the average current consumption of the network is 100µA, and the sampling rate is once per second, then the battery can last for 80 days; if the sampling rate is once every two minutes, the average current consumption drops to 1.92µA, and the battery life can be extended to 13.1 years. In order to overcome the problem of short battery life faced by remote wireless sensor networks, the US company Millennial Net has combined its i-Bean wireless technology with the "energy harvesting" technology from the emerging company Ferro Solutions. The two companies recently demonstrated an i-Bean wireless transmitter that works by an inductive oscillation energy converter. This converter can generate a voltage of 1.2mV to 3.6mV from 28Hz to 30Hz oscillations under the action of 50mg to 100mg force, and allows data to be transmitted at a rate of 115Kb/s over a distance of 30m (without battery). The company also collaborates with other companies to develop solar panels to power wireless sensors. In terms of energy optimization research, Huang Jinhong et al. of Xi'an Jiaotong University proposed an energy-optimized adaptive organization structure and protocol ALEP for wireless sensor networks in reference [15]. Compared with traditional wireless micro-sensor network protocols, ALEP takes into full consideration practical applications. It introduces an efficient energy control algorithm into the networking protocol, improves the energy utilization rate of the network, significantly extends the life cycle of the wireless network, and enhances the robustness of the network. Through OPNET simulation of the ALEP protocol, the results show that compared with the traditional wireless micro-sensor network protocol, the protocol has more efficient energy characteristics and information transmission characteristics under the condition of transmitting the same amount of data. IV. Conclusion Although the application prospects of wireless sensor networks are very promising, due to several technical difficulties, they cannot be widely used yet. Researchers have encountered severe challenges in integrating MEMS and other electronic devices into a single chip. The various algorithms mentioned in this paper still need to be tested for their practicality in engineering implementation. References: [1] S. Yi, P. Naldurg, R. Kravets. Security-aware ad hoc routing for wireless networks[C]. Proc. of 2001 ACM International Symposium on Mobile Ad Hoc Networking and Computing, 2001, pp.299-302. [2] DE Bell, LJ LaPadula. Secure computer systems: mathematical foundations and model[R]. Technical Report M74-244, Miter Corporation, 1975. [3] Swades De .Chunming Qiao, Hongyi Wu. Meshed multipath routing with selective forwarding: an efficient strategy in wireless sensor networks[M]. [4] Sasikanth Avancha, Je.rey Underco.er, Anupam Joshi, John Pinkston. Secure sensor networks for perimeter protection[M]. [5] A. Perrig, R. Szewczyk, V. Wen, D. Culler, JD Tygar. SPINS: security protocols for sensor networks[J]. Wireless Networks 8 (2002) 521–534. [6] WR Heinzelman, J. Kulik, H. Balakrishnan. Adaptive protocols for information dissemination in wireless sensor networks[C]. Proceedings of the Fifth Annual ACM/IEEE International Conference on Mobile Computing and Networking (MobiCom_99), Seattle, WA, August 1999, pp. 174–185. [7] J. Li, J. Jannotti, D. Couto, D. Karger. R. Morris, A scalable location service for geographic ad hoc routing[C]. ACM Mobicom, 2000. [8] Ahmed Helmy. Mobility assisted resolution of queries in large scale mobile sensor networks MARQ[J]. Computer Networks, 43 (2003) 437–458. [9] Koen Langendoen, Niels Reijers. Distributed localization in wireless sensor networks: a quantitative comparison[J]. Computer Networks, 43 (2003) 499–518. [10] Hongwei Zhang, Anish Arora. GS3: scalable self-configuration and self-healing in wireless sensor networks[J]. Computer Networks, 43 (2003) 459-480. [11] W. Heinzelman, A. Chandrakasan, H. Balakrishnan. Anapplication-speci.c protocol architecture for wireless microsensor networks[C]. IEEE Transactions on Wireless Networking 1 (4) (2002) 660–670. [12] Suman Banerjee, Samir Khuller. A clustering scheme for hierarchical control in multi-hop wireless networks[C]. IEEE INFOCOM 2001, pp. 1028–1037. [13] Ting-chao Hou, Tzu-Jane Tsai. An access-based clustering protocol for multihop wireless ad hoc networks[J]. IEEE Journal on Selected Areas in Communications 10 (7) (2001) 1201–1210. [14] Sokwoo Rhee, Deva Seetharam, Sheng Liu, Ningya Wang, Jason Xiao. I-Beans: An Ultra-low Power Wireless Sensor Network. [15] Huang Jinhong, Zuo Fei, Zeng Ming. An energy-optimized adaptive organization structure and protocol for wireless sensor networks[J]. Telecommunication Engineering, 2002, Vol. 42, No. 6. About the authors: Xiao Jian, Master's student of the School of Information Science and Technology, Nanjing University of Aeronautics and Astronautics, whose main research directions are digital communication, spread spectrum technology, wireless sensor networks, etc. Lü Aiqin, Master's student of the School of Computer Science, Xi'an University of Electronic Science and Technology, whose main research directions are wireless communication simulation design, wireless sensor networks, etc. Chen Jizhong is an associate professor at the School of Information Science and Technology, Nanjing University of Aeronautics and Astronautics. His main research areas are digital communication and spread spectrum technology. Zhu Minghua is an associate researcher at the Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences. His main research area is wireless communication simulation design.